Abstract:The process of fertilization and gender determination in avian eggs directly impacts the economic efficiency and ethical considerations within poultry incubation and breeding operations. Historically, the detection of fertilization status and gender information in eggs has represented a significant bottleneck in the advancement of the poultry and egg sectors. The non-destructive identification of egg sex stands as a particularly formidable challenge on a global scale. Non-destructive detection methodologies that are both suitable and efficacious promise not only to unveil pertinent, previously inaccessible information but also to do so without inflicting any harm upon the eggs, hence the surge in research interest in recent times. Nevertheless, the quest for precise detection is complicated by factors, including but not limited to the variability in eggshell thickness and coloration, the dynamic nature of the internal fluids, and the intricacies of embryonic development. A comprehensive comparative analysis of the array of non-destructive detection techniques show cased in extant literature was conducted, encompassing machine vision, spectroscopic technologies, acoustic resonance frequency analysis, bioelectrical signal analysis, percussive vibration methods, dielectric constant analysis, and olfactory signature analysis. Moreover, it delineated the persistent challenges faced in the nondestructive assay of fertilization and sex data in avian eggs and articulated a forward-looking discussion on the potential integration of nascent technological applications in forthcoming investigative endeavors.

Abstract:A crop line feature detection method based on semantic segmentation network was proposed to realize stable and reliable visual servo control of plant protection robot. Based on the semantic segmentation network which was termed with ESNet, pixel-wise labeling in farmland images was performed for ribbon regions detection, and least mean squares algorithm was utilized to find out all the crop line feature parameters in real time. Among the derived candidate lines features, a key route line was chosen as the valid navigation path which was responsible for subsequent robot motion control. Kalman filter was subsequently employed to smooth geometrical parameters of the previously specified key route, which effectively suppressed the fluctuation of navigation parameters caused by jolt behavior of plant protection robot generated from uneven ground and measurement noises incorporated in visual images. Afterwards, the sophisticated Ackermann steering kinematic model which was characterized by robot front-wheel steering and rear-wheel differential was introduced. A pure tracking controller was designed in Cartesian coordinate system to realize the servo motion control of plant protection robot. The field experiment conducted in real farmland scenarios verified the effectiveness of the proposed method.

Abstract:In order to investigate the influence of different positions of apple tree branches on the picking effect under different combinations of vibration parameters, the main factors affecting the apple drop were obtained as vibration frequency, amplitude and holding position through the establishment of a secondary branch vibration mechanics model of the apple tree and analytical analyses. Measuring the morphological characteristics of apple trees and based on the principle of dwarf anvil dense planting shaping to establish a three-dimensional model of fusiform apple trees, using ANSYS software to carry out finite element simulation analysis of the apple tree model, the simulation results showed that the vibration frequency of 4~8Hz,amplitude of 20~30mm, the holding position of 0.35l~0.65l (l was the length of secondary branches), the damage to the fruit tree was small and easy to fall off the apples. In order to determine the best combination of vibration parameters for different positions of branches in apple trees, a four-factor, three-level field experiment was designed, and the experimental data were analyzed and response surface was optimized by using Design-Expert software, and the results of the parameter optimization were as follows: when picking apples on the upper layer of the apple tree, the vibration frequency was 5Hz, the amplitude was 28mm, and the clamping position was 0.40l. When picking apples in the middle layer of the apple tree, the vibration frequency was 4Hz, the amplitude was 30mm, and the clamping position was 0.43l; when picking apples from the lower layer of the apple tree, the vibration frequency was 8Hz, the amplitude was 20mm, and the clamping position was 0.65l; it can be seen through the validation test that the picking rates of the upper to the lower layer of the apple tree were 96.4%, 94.8%, and 93.2%, which were similar to the optimized values, indicating that the optimized model was reliable.

Abstract:To address the issues of residue accumulation and blockage, significant soil disturbance, low cleanliness of seedbed, and poor seeding quality caused by maize stubble and straw residues during no-tillage wide-row wheat sowing in maize residue-covered fields, a seedbed cleaning method combining passive cutting and subsequent active throwing was proposed. A cutting and spreading combined residue-breaking and residue-preventing device was designed, which utilized longitudinally arranged inclined notched discs and rotating spreading devices to clean seedbed residues and straw. A diversion-type furrow opener was designed to stabilize row spacing and achieve one furrow seeding two rows. The tilt angle and skew angle parameters of the notched discs were determined based on force and kinematic analyses of straw residues during device operation. The straight and inclined rotary blade parameters were designed and calculated based on the sliding-cutting theory. Using the orthogonal rotation combination experimental method, key parameters of the spreading device were optimized through simulation experiments on the established MBD-DEM combined simulation platform, with seedbed cleanliness rate, residue breaking rate, and soil disturbance as indicators. Based on simulation results, regression models for each indicator were established and response surface analysis and multi-objective optimization were conducted. The optimal parameters were determined to be as tilt angle of 20° and rotation speed of 310r/min, resulting in seedbed cleanliness rate of 96.9%, residue bonds breaking rate of 26.9%, and soil disturbance width of 139mm. Wheat seeding experiments in maize residue-covered fields showed that the sowing machine equipped with the cutting and spreading combined anti-breaking and residue-preventing device had good pass ability, with seedbed cleanliness rate of 90.1%, residue-breaking rate of 96.2%, soil disturbance width of 127mm, soil disturbance rate of 6.9%, stable overall operation quality, and uniform wheat emergence after seeding, meeting the agronomic requirements for wide-seedbed seeding of wheat.

Abstract:Aiming at the problems of long R&D cycle and high R&D cost of no-tillage planter for maize, it was proposed to apply the meta-universe technology to the simulation of corn no-tillage planter operating unit. Based on the architecture of the meta-universe system, the simulation platform was constructed, and the scenes in the virtual farm were highly restored under the meta-universe environment by using the marking line method and the transformation method, and an immersive corn no-tillage planter driving platform was constructed and tested for the interactivity, to verify that the immersive corn no-tillage planter driving platform had better interactivity. The interactive performance of the immersive corn no-till planter driving platform was verified to be better. On this, seeding performance simulation tests were carried out at different driving speeds, and the overall seeding quality showed a decreasing trend when the forward speed of the corn no-tillage planter unit was increasing, and the overall trend was close to the actual field test results, and the spacing pass rate was greater than 88.52%, the reseeding rate was less than 6.97%, the omission rate was less than 4.51%. The test results showed that the virtual seeding simulation platform designed and implemented can be used for corn seeding experimental research, which was of great significance to reduce the research and development cost of no-tillage seeding machine and shorten the research and development cycle of no-tillage seeding machine for corn.

Abstract:Considering at the problem of small particle size and light quality of rapeseed, which makes it difficult to displace individual seeds, a positive and negative pressure combination roller type precision seed-metering device was designed. The working principle of the seed-metering device was clarified, and the force analysis of the seed absorbing, seed carrying and seed discharging links of the seed-metering device was carried out, as well as the analysis of the dragging process of the seed discharging roller on the seed population. The mechanism of preventing dragging accumulation by controlling the height of the seed layer in the seed filling area and the pressure of the seed population was proposed, and a seed filling chamber structure with lateral seed filling and free fall of the dragging seed was designed. The influence of the height of the seed layer and the structure of the seed filling chamber on the performance of the seed filling in the seed filling area of the seed-metering device and the solution of the dragging accumulation of the seed population were investigated by using the discrete element simulation. The simulation results showed that the average kinetic energy of the seed population inside the seed dispenser was increased with the increase of the filling seed layer height, and the average disturbance capacity of the seed population was gradually increased. Under the condition of 50mm seed filling layer height, the designed anti-drag strip accumulation seed filling chamber reduced the pressure on the seed population at the bottom of the seed filling area, and the phenomenon of population drag strip accumulation did not occur, and the disturbing effect on the seed population in the seed filling area was maintained. Seed metering tests were conducted on the JPS-12 seed-metering device inspection test bench, the test results showed that in the roller speed at 15~30r/min, suction seed negative pressure in the range of 1.0~1.2kPa seed-metering device qualified index could reach more than 90%. The results of the field test showed that the coefficient of variation of plant spacing stability was 4.4% and the coefficient of variation of seedling number consistency in each row was 8.14% in the actual field operation of the planter equipped with positive and negative pneumatic pressure combination roller-type small-size seed single-seed precision planter, which met the requirements of precision sowing.

Abstract:In order to solve the traditional mechanical vegetable seeder can not realize the problem of precision sowing and the existence of injured seeds, a radish spoon type precision seed discharger was designed. Seed filling, seed cleaning and seed feeding were accomplished by using a seed scoop suspended from the side of the seed tray, realizing a non-touch operation process. The movement of seeds into and out of the seed scoop was analyzed theoretically, and to clarify the rationale for its non-injurious nature, the basic structural parameters of seed disks, seed scoops and seed tubes were determined, EDEM discrete element simulation software was used to simulate the working process under different structural dimensions of the seed-rowing spoon. The test factors were the diameter and depth of the hole in the seeding spoon and the rounding ratio of the release surface, single grain rate, multiple grain rate and empty grain rate were used as test indicators. Simulation experiments were conducted by using a 3-factor, 5-level secondary general rotary combination design, the optimal structure parameters of the seed discharge spoon for the hole diameter of 5mm, the depth of 4.3mm, the release surface fillet ratio of 0.12 were determined. Based on discrete element simulation of this parameter, bench testing was done via a homemade seed dispenser test bed and attachment of seed displacer to planter for field trials. The results of the simulation test were as follows: single grain rate was 93%, multiple grain rate was 4%, empty grain rate was 3%. The results of the bench test were as follows: the mean value of the pass index was 92.2%, the mean value of the reseeding index was 4.6% and the mean value of the leakage index was 3.2%，the relative errors were 0.86%, 15% and 6.67%, respectively. The results of the field trial were as follows: the mean value of the pass index was 90.5%, the mean value of the reseeding index was 6.9% and the mean value of the leakage index was 2.6%，which demonstrated good performance of precision seeding with this seed dispenser. Comparison test of simultaneous damage rate with brush fossa wheel seed discharger showed that the injury rates were 0.43% and 1.27%, respectively，relative error was 66.14%，indicating a significant reduction in seed damage.

Abstract:To enhance the accuracy of simulating charcoal loading, hole pressing, and overlaying in the hole tray seeding process, based on the material properties of charcoal, the Edinburgh Elasto-Plastic Adhesion (EEPA) model was selected to establish a discrete element simulation model of charcoal in EDEM software. The peat parameters were calibrated through shaft-closed compression and virtual simulation tests, while the density, particle size distribution, and contact parameters of peat were measured through physical experiments. Significance analysis experiments were designed by using Plackett-Burman Design and the steepest climbing test to determine the restitution coefficient, static friction coefficient, tangential stiffness factor, and shear of peat, with modulus having a significant effect. A quadratic polynomial regression model was established between the response value and four significant parameters using the central composite design test. The axial pressures of 3.83N and 91.45N corresponding to the uniaxial closed compression of 20% and 50% axial strain were used as target values for testing the significant parameters. After optimization, the optimal combination was determined to be a recovery coefficient of 0.202 between peat plants, a static friction coefficient of 0.595 between peat plants, a tangential stiffness factor of 0.667, and a peat shear modulus of 0.613MPa. Finally, the simulated values and the measured values under this parameter combination were compared and verified. The average error between the measured values and the simulated values in the axial strain range of 20% to 50% was approximately 8.08%, with the relative error reaching the maximum value of 15.34% at about 40% of axial strain. These results demonstrated that the EEPA model parameters calibrated based on the response surface method can be used for discrete element simulation research.

Abstract:In response to the current issues of the uneven application, difficulty in near-root application, limited operational functionality in liquid manure fertilizer shovel, a layered near-root and basic application of the liquid manure fertilizer shovel tracking the integration of the fertilizer was designed from the perspective of improving fertilizer efficiency,reducing emissions, and improving the performance. Fertilizer, obstacle avoidance device and other key components were developed. The fertilizer shovel-soil mechanics simulation model was designed by using EDEM discrete elements to optimize the parameters of the lateral discharge pipe of the fertilizer shovel during the follow-up application process, and to build the performance test platform of the fertilizer shovel. Fertilizer shovel layered fertilization and the effect of near-root fertilization were tested by clear water simulation method. The results showed that when the backward inclination angle of the lateral discharge pipe of the fertilizer shovel was 15°,and the cutting edge angle of the lateral discharge pipe was 18°,the resistance of fertilizer shovel basal fertilizer operation was the minimal. With the layered fertilizer application,when the operating speed of the fertilizer shovel was 3km/h and the discharge volume was 5L/s,the longitudinal spreading depth of the fertilizer in the soil was 235mm,which was 65% higher than that of the pre-improvement single-port fertilizer discharge method.When the fertilizer shovel was operated at a speed of 1.2km/h and a discharge rate of 3L/s, the fertilizer was distributed within a radius of 100mm from the center of the crop root by about 80% effectively achieving near-root fertilization.

Abstract:Aiming at the problems of the existing agricultural equipment digital twin system development difficulty, high configuration requirements and poor portability, a lightweight network-based digital twin system construction method for combine harvester was proposed, which contained the realization of multiple subsystems such as physical, virtual, data interaction, model computation and human-computer interaction. Based on the technical characteristics of digital twin and the operational characteristics of combine harvester, a lightweight digital twin system framework was designed based on JavaScript language. By adopting Solidworks and CMdevelopment kit tools for the model lightweight processing and coordinate system integration of the digital twin system, it achieved a significant reduction of the system’s hardware requirements and memory occupation without affecting the model’s accuracy and functionality. The lightweight network-based combine harvester digital twin system was developed by using a Lovol GM100 combine harvester as an object to provide a joint simulation, analysis, and validation platform for performance analysis, real-time monitoring, instantaneous computation, and remote manipulation of the combine harvester digital twin system. To verify the performance and functionality of the digital twin system, twin system performance tests and fuel consumption prediction experiments were conducted. Tests showed that the response speed was within 78ms at a data update frequency of 20Hz, and the memory occupation was within 331MB in the performance test, and the average occupancy of the system’s CPU and GPU in the running state was 17% and 30%, respectively; and the system’s frame rate can be maintained at 75.6f/s even under high-intensity operation. Under normal operation, the average error of the fuel consumption prediction model was 0.34L/h, with an average relative error of only 2.51%. This system can provide a low-cost, high-efficiency digital twin lightweight construction scheme, which provided a useful reference for the further promotion and application of digital twins in the field of agricultural equipment.

Abstract:Aiming at the field corn harvester harvesting seed corn is prone to the phenomenon of injury to ears and seeds, debris blockage, etc., a large-scale seed corn combine harvester for the biological characteristics of seed corn in the suitable harvest period was designed. The machine adopted small row spacing to row flexible plate picking cutting platform and replaceable combined peeling device, to ensure low-loss picking, conveying, peeling operation and reduce seed loss and damage, which was equipped with steel rubber-covered curved picking plate on the upper part of the cutting platform, “rubber+steel” clamping conveyor chain and six-pronged low-speed pulling rollers, and the replaceable combined peeling device adopted flexible peeling+rubbing+speed reduction. The main factors affecting the machine indexes were extracted by screening through the Plackett-Burman experimental design, and the Box-Behnken experimental design principle was applied, taking the forward speed of the machine, the rotational speed of the stem pulling roller and the rotational speed of the peeling roller as the experimental factors, and taking the total loss rate and the impurity rate as the experimental indexes, the machine was examined through the field test, and the best operating parameters of the machine were optimized. The test results showed that the total loss rate of cob was 1.61% and the impurity rate was 0.55% when the optimized machine forward speed was 4.87km/h, the stem pulling roller speed was 877.27r/min and the peeling roller speed was 442.52r/min. It was verified in the field test: when the forward speed of the harvester was 4.9km/h, the rotational speed of the stalk pulling roller was 880r/min, and the rotational speed of the peeling roller was 450r/min, the total loss rate of the cob was 1.64%, and the rate of impurity was 0.57%, which met the requirements of the mechanized joint harvesting of corn for seed production, and it can be used as a reference for the design and test of the joint harvesting machine for corn for seed production.

Abstract:In view of the fact that the current corn grain harvester can not meet the need of large feeding quantity cleaning above 15kg/s, a kind of cleaning device with pre-cleaning function was designed. Firstly, the stress analysis of corn grains before the extruder leaved the screw conveyor and reached the pre-cleaning screen was carried out, and then the motion model of the crank linkage mechanism was simplified. Secondly, the motion state of corn grains on the sieve surface was analyzed. The impeller and volute of centrifugal fan were designed and calculated. The single factor test was used to determine the fan speed, vibration frequency and the value range of the upper screen opening. With the fan speed, vibration frequency and the opening of the upper screen as test factors, and the grain impurity content and cleaning loss rate as evaluation indexes, the three-factor and three-horizontal center combined test was designed to establish a regression model between the factors and indexes. The test results were analyzed by response surface method, and the regression model was optimized by DesignExpert 12. When the feeding amount of corn extract was 16kg/s, the optimal combination was obtained. The test results showed that the fan speed was 1202.50r/min, the vibration frequency was 5.41Hz, and the opening of the upper screen was 18mm. Under these conditions, the grain impurity content was 0.79% and the cleaning loss rate was 1.10%. The verification test showed that the speed of the fan was 1200r/min, the vibration frequency was 5Hz, and the opening of the upper screen was 18mm. At this time, the kernel impurity content was 0.82%, the cleaning loss rate was 1.14%, the relative error between the test value and the optimization value was less than 5%, and the kernel impurity content was 2.07 percentage points lower than that of the traditional double-layer reciprocating vibrating screen cleaning device. The loss rate of cleaning was reduced by 2.13 percentage points, which proved that the design was reasonable and met the national cleaning standard.

Abstract:Aiming at the oil bean mechanized harvesting process of lifting and transporting vibration device to remove impurities is not obvious, resulting in the oil bean and soil impurities and other separation is not complete, the soil leakage rate is low, the rate of injury to the fruit is high, as well as the soil clogging and back to the belt in the process of lifting and transporting and so on, combined with the oil bean fruit-soil-seedling agglomerates, a hinge sieve sheet lifting device was designed through the sieve sheet bending part of the agglomerate to provide the same direction of movement with the chain sieve to increase the area of the screen holes and increase the effect of soil leakage and effectively avoid the phenomenon of back to the belt. Through the bending part of the sieve plate to provide the agglomerates with the same thrust as the chain sieve movement direction, the sieve area was increased, so that the hinge-type lifting device can increase the effect of soil leakage and effectively avoid the phenomenon of back to the belt. The inclination angle and fixed position were analyzed and designed, and the simulation test was carried out by EDEM. Taking the linear speed of the chain sieve, the bending height of the chain sieve plate, and the vibration frequency of the chain sieve as the test factors, and the soil leakage rate and the fruit injury rate of the oleaginous soybean agglomerates in the conveying process as the test indexes, the optimal structural parameter combinations of the conveying sieve were obtained through the analysis of three-factor and three-level orthogonal simulation test. The test results showed that the optimal combination of chain sieve line speed was 1.151m/s, bending height was 27.779mm, and vibration frequency was 9.561Hz, at this time, the soil leakage rate of the lifting device was 96.524%, and the fruit injury rate was 2.439%. The results of the field verification test showed that the average soil leakage rate of the hinge-type lifting device was 96.05% and the average fruit injury rate was 2.38% when taking the chain sieve linear speed of 1.2m/s, bending height of 28mm, and vibration frequency of 9.5Hz, which was basically the same as the results of the simulation test, and it met the working requirements of the chain sieve for lifting and transporting the oil soybeans.

Abstract:The self-propelled sprayer for high ground clearance is prone to wheel slippage in complex operating environments, which can affect the flow and pressure stability of the static hydraulic drive system. In severe cases, this can lead to a loss of maneuverability, necessitating the implementation of anti-skid control to ensure stable driving performance and the ability to escape from difficult situations. A static-hydraulic drive system anti-skid control scheme was proposed for a large-scale self-propelled sprayer for high ground clearance. It defined the relationship between slip ratio and adhesion coefficient by using a bilinear model, and a sliding mode controller was designed. Field off-road tests were conducted to validate the control performance of the anti-skid system. The experimental results demonstrated that the system can control the sprayer’s slip ratio within 0.15. Under conditions of initial acceleration and constant speed, the mean slip ratios were 0.020 and 0.019, respectively. Moreover, under ditch-crossing conditions, the entire machine can be rapidly freed within 2s. These results confirmed that the designed sliding mode anti-skid system for the sprayer exhibited excellent anti-skid performance, ensuring smooth operation under typical conditions and effectively reducing the impact of unfavorable ground conditions on overall stability. The research did not utilize flow control valves or diverter valves, ensuring high control precision. It held the potential to achieve comprehensive and continuous anti-skid control for large-scale high ground clearance sprayers, thereby enhancing their driving stability, ground traversability, and active escape capability.

Abstract:With the aim to study the characteristics of complex vortices in the vortex pump, quantify the volume and intensity of vortices in pump, and analyze the influence of vortex structure on energy conversion and energy loss, the unsteady numerical simulation and external characteristic experiments were carried out to analyze the single-stage vortex pump, the Ω method, and Liutex method were used to identify and characterize the intensity of vortex in the pump, the average vortex intensity was proposed for the quantitative study, and the analysis was carried out by combining the vortex dynamics decomposition formula of kinetic energy equation and the vortex decomposition theory. The results showed that there was a spiral vortex structure in the fully developed vortex region of the pump, which flowed out from the impeller channel into the side channel. With the increase of flow rate, the number of vortex tubes was decreased and the vortex strength was decreased. With the increase of flow rate, the volume and intensity of the vortex in the impeller was decreased, while the relative change in the side channel was relatively small. The average vortex intensity in the impeller was much greater than that in the side channel under the same working conditions. Pressure gradient contributed the most to the kinetic energy conversion of fluid, and the proportion of momentum transport and dissipation loss caused by vortex structure was small, but the rigid vorticity, that was, the strength of vortex structure, was positively related to momentum transport, while the deformed vorticity was strongly related to the energy loss of enstrophy.

Abstract:Analyzing the spatial distribution characteristic of rural settlements and their influencing factors on a macro scale is conducive to comprehensively understanding the spatial differential law of rural settlements and its forming reasons, and providing a research basis for village planning and other work. Taking Hebei Province as the main research area, the overall changes and spatial distribution characteristics of rural settlements were analyzed with the help of landscape pattern indices, the nearest neighbor analysis and kernel density estimation, and the main factors influencing rural settlement spatial distribution were explored based on the integrated application of distribution indices, factor analysis and multiple linear regression analysis. The results showed that since 2000, especially after 2010, rural settlements in Hebei Province had mainly expanded their area through extension development and gradually showed a large-scale and concentrated distribution. They were aggregated in the southeast of Hebei Province and their degree of agglomeration was continuously increasing. The limitations of natural factors such as towering terrain, soil that was not conducive to agricultural production and rivers and lakes of different scales on the spatial distribution of rural settlements were weakening. There were differences in the degree and radius of agglomeration effects that locational factors such as towns or roads of different levels on rural settlements. The rural production and housing scale, the regional economic development and the regional infrastructure completeness were the main social and economic factors that affected the spatial distribution of rural settlements in Hebei Province, and the main areas affected by different factors were distinct. The research results can provide a theoretical basis for the differentiated optimization of rural settlement in relevant regions of northern China.

Abstract:In view of the problems of high computational complexity, large memory consumption, and difficulty in deployment on mobile platforms with limited computing power in DeepLabV3+ segmentation model, an improved lightweight DeepLabV3+ algorithm was proposed to realize the segmentation and recognition of buckwheat by UAV at seedling stage. The algorithm adopted the fusion of re-parameterization visual geometry group (RepVGG) and mobile vision transformer (MobileViT) modules to establish the backbone network for feature extraction. At the same time, the squeeze-and-excitation networks (SENet) attention mechanism was introduced into the RepVGG network structure to capture more global semantic information by using the correlation between channels, and ensure the performance of buckwheat segmentation. Experimental results showed that compared with fully convolutional networks (FCN), pyramid scene parsing network (PSPNet), dense atrous spatial pyramid pooling (DenseASPP), DeepLabV3, and DeepLabV3+ models, the improved algorithm proposed greatly reduced the model parameters, making it more suitable for deployment on mobile terminals. The mean pixel accuracy (mPA) and mean intersection over union (mIoU) on the selfbuilt buckwheat segmentation dataset were 97.02% and 91.45%, the overall parameters, floatingpoint operations (FLOPs) and inference speed were 9.01×106, 8.215×1010 and 37.83 f/s, respectively, with the best performance. In the full-size image segmentation, the mPA and mIoU for buckwheat segmentation can meet the requirements at different flight heights, which had good segmentation ability and inference speed. The algorithm can provide technical support for the later buckwheat seed replacement, fertilization maintenance, and growth monitoring, and promote the intelligent development of small and coarse grain industry.

Abstract:Remote sensing technology for ground change detection has been widely used in the fields of agricultural planting planning and disaster situation assessment. For grapes, which is an important economic crop in China, accurately obtaining its spatial change information is crucial for industrial planning and sustainable development. Nevertheless, the dispersed arrangement of the grape growing areas, their diverse sizes, and the intricate nature of feature types, along with the heterogeneity among different temporal images, collectively contribute to a diminished accuracy in detecting areas of change. Therefore, a change detection model (Multiscale difference feature capture net, MDFCNet) based on attention mechanism and multiscale difference features was proposed.The main structure of the network adopted an encoderdecoder structure, which incorporated the squeeze and excitation (SE) attention module on the basis of ResNet101 backbone network to improve the network’s ability to adequately extract change features from remote sensing images, suppressing interference from extraneous pixels. We also designed the cross difference feature capture (CDFC) module, it captured different features with dense contextual information, thereby improving the accuracy of change detection in the case of complex feature types. While the supervised ensemble attention (SEA) module was designed to enrich multiscale features by fusing low-level detailed texture features and high-level abstract semantic features layer by layer to enhance the network’s ability to detect small planting areas. Comparison and ablation experiments were conducted on the constructed change dataset of grape growing area, which was located within the city of Yinchuan, Ningxia Hui Autonomous Region. The experimental results showed that the MDFCNet method achieved the best detection results compared with the current state-of-the-art change detection methods of SNUNet, A2Net, DSIFN and ResNet-CD. Compared with the model with the 2nd highest performance(A2Net), the evaluation metrics of IoU, recall, F1 value and precision were improved by 5.42, 5.62, 3.48 and 0.95 percentage points, respectively. And the ablation experiments also demonstrated the effectiveness of fusing the modules. Compared with the base network, the addition of the three modules resulted in 12.9, 5.63, 8.64 and 11.75 percentage points increases in the evaluation metrics of IoU, recall, F1 value and precision respectively. The model extracted different features with larger sensory fields to provide rich inferential information for change detection, and the fused multiscale features can effectively avoid the problem of false detection and missed detection in the results. The extracted change areas were more complete and retain more edge detail, providing a solution to the task of change detection for the complex background of the wide range of grape growing areas.

Abstract:The identification and selection of drought-tolerant wheat varieties are of great significance for ensuring food security and sustainable agricultural development in China.The root system serves as the primary pathway for water absorption in plants, and the phenotype of the root system is closely related to drought tolerance. In order to obtain root system phenotypic indicators of wheat quickly and accurately,drought stress experiments were conducted by using the soil box method and collected sequential root images at 18 time points. A deep learning-based image processing and analysis workflow was proposed for root image processing. To address the issue of root breakage caused by soil occlusion, a two-stage detection-repair method combining object detection networks and hourglass attention networks was designed to repair the broken root regions. Multi-scale training and adaptive iteration were employed to improve the accuracy and robustness of the repair process. Six phenotypic traits of wheat root systems, including root area, total root length, and root width, were extracted under drought stress and control conditions, and the phenotypic responses of wheat roots to drought stress were analyzed. The results showed that under drought stress, wheat exhibited lower root biomass, deeper rooting depth, and more dispersed root architecture. The drought tolerance index of wheat root systems was calculated, and the drought tolerance of wheat varieties was described and ranked by using principal component analysis. A method of in-situ measurement and analysis of wheat root phenotype was proposed, which can be applied to the study of wheat drought resistance.

Abstract:The in-situ phenotype of sweet pepper is an important reference indicator for fruit breeding and management. Automated measurement of phenotype in-situ through phenotype collection robots is one of the effective ways for digital breeding and management of sweet pepper. However, fruit occlusion during the measurement process seriously affects the success rate of detection. Therefore, a three-dimensional reconstruction method for multi view sweet pepper fruit point cloud was proposed to address the problem of target occlusion in in-situ fruit phenotype measurement. By using the method of virtual leaves, an enhanced dataset was created, and a sweet pepper fruit recognition model based on YOLO v5 algorithm was established to recognize fruits with different degrees of occlusion. At the same time, a fruit phenotype collection algorithm considering fruit position and occlusion degree was constructed to achieve multi view threedimensional data collection of fruits. Finally, the three-dimensional point cloud of sweet pepper fruit was registered, the phenotype parameters of sweet pepper was extracted, and the effectiveness of the point cloud reconstruction method was validated through the greenhouse sweet pepper fruit phenotype. Compared with manual measurement data, the average relative error of fruit width was 1.72%, and the average relative error of fruit height was 1.60%. The experimental results indicated that the in-situ phenotype point cloud reconstruction method proposed for sweet pepper can provide effective solutions and feasible methods for crop phenotypes under occlusion conditions.

Abstract:In response to the problems of low efficiency and strong subjectivity in obtaining information about the current stage of wheat development that relies on manual observation, a wheat image dataset consisting of four key growth stages of winter wheat: winterovering stage, green-turning stage, jointing stage, and heading stage, totaling 4599 images were constructed. A lightweight model FSST (fast shuffle swin transformer) based on FasterNet was proposed to carry out intelligent recognition of these four key growth stages. Firstly, based on the partial convolution of FasterNet, the Channel Shuffle mechanism was introduced to improve the computational speed of the model. Secondly, the Swin Transformer module was introduced to achieve feature fusion and self attention mechanism, it can improve the accuracy of identifying key growth stages of wheat. Then the structure of the whole model was adjusted to further reduce the network complexity, and the Lion optimizer was introduced into the training to accelerate the training speed of the model. Finally, model validation on the self-built wheat dataset with four key growth stages was performed. The results showed that the parameter quantity of the FSST model was only 1.22×107, the average recognition accuracy was 97.22%, the F1 score was 78.54%, and the FLOPs was 3.9×108. Compared with that of the FasterNet, GhostNet, ShuffleNetV2 and MobileNetV3 models, the recognition accuracy of the FSST model was higher, the operation speed was faster, and the recognition time was reduced by 84.04%, 73.74%, 72.22% and 77.01%, respectively. The FSST model proposed can effectively identify the key growth stage of wheat, and had the characteristics of fast, accurate, and lightweight recognition. It can provide a reference for optimizing the application of deep learning models in smart agriculture and offerring information technology support for real-time monitoring of field crop growth on resource-constrained mobile devices.

Abstract:Plant height is an important phenotypic indicator for identifying maize germplasm traits and crop vigor, and maize genetic characteristics are obvious at the seedling stage, so accurate measurement of plant height at the seedling stage is of great significance for maize genetic characteristics identification and field management. Aiming at the problem that traditional plant height acquisition methods rely on manual measurement, which is time-consuming and subjective error, an improved ZoeDepth monocular depth estimation model incorporating mixed attention information was proposed. The improved model added the Shuffle Attention module to the various stages in the Decoder module, so that the Decoder module was more able to pay attention to the effective information in all the feature maps in the process of extracting information from the low-resolution feature maps, which enhanced the model’s ability of key information extraction, and could generate more accurate depth maps. In order to verify the effectiveness of the method, the validation was carried out on the NYU-V2 depth dataset, and the results showed that the ARE, RMSE, LG were 0.083, 0.301mm and 0.036, and the accuracy δ under different thresholds of the improved Shuffle-ZoeDepth model were 93.9%, 99.1% and 99.8%, respectively, all of which were better than those of the improved Shuffle-ZoeDepth model on NYU-V2 depth dataset.In addition, the Shuffle-ZoeDepth monocular depth estimation model combined with the maize plant height measurement model was used to complete the measurement of seedling maize plant height, and maize height measurement experiments were carried out by collecting images of seedling maize at different distances, and when the maize height was in the three height intervals of 15~25cm, 25~35cm, and 35~45cm, the AE were respectively 1.41cm, 2.21cm, and 2.08cm, and the PE were 8.41%, 7.54%, and 4.98%, respectively. The experimental results showed that this method can accomplish the accurate measurement of maize plant height at the seedling stage in complex environments using only a single RGB camera with a complex outdoor environment.

Abstract:In the practice of modern agricultural production, the method of harvesting agricultural products is gradually shifting towards mechanization and intelligence. An increasing number of robots are being introduced into actual production and progressively replacing traditional manual labor. However, in natural environments, factors such as weather,lighting,the similarity in color between fruits and their backgrounds,and mutual occlusion between fruits and branches significantly increased the difficulty of fruit target detection. To accurately detect pears in natural environments, a lightweight pear detection method M-YOLO v7-SCSN+F was designed based on the YOLO v7-S foundational model. This model introduced MobileNetv3 into the YOLO v7-S model as its backbone feature extraction network, thereby reducing the number of parameters in the network. It incorporated a coordinate attention (CA) mechanism in the model’s feature fusion layer to enhance the network’s feature representation capabilities. The loss function CIoU in YOLO v7-S was replaced with SIoU, which was used in conjunction with the normalized Wasserstein distance (NWD) mechanism for small target detection, further improving the detection accuracy for fragrant pears. Based on the Fourier transform (FT) data augmentation method, new image data was generated by analyzing the frequency domain information of images and reconstructing the amplitude components, thereby enhancing the model’s generalization ability. Experimental results showed that the improved M-YOLO v7-SCSN+F model achieved mean average precision (mAP), precision, and recall rates of 97.23%,97.63% and 93.66%,respectively,on the validation set,with a detection speed of 69.39 f/s. The proposed detection model improved performance compared with Faster R-CNN, SSD, YOLO v3, YOLO v4,YOLO v5s, YOLO v7-S, YOLO v8n and RT-DETR-R50 models on the validation set, with mean average precision (mAP) enhancements of 14.50, 26.58, 3.88, 2.40, 1.58, 0.16, 0.07 and 0.86 percentage points, respectively. Furthermore, the improved M-YOLO v7-SCSN+F model reduced its parameter count by 16.47MB and 13.30MB, respectively, when compared with the advanced YOLO v8n and RT-DETR-R50 detection models. The detection model introduced demonstrated a high degree of effectiveness in target detection for mature pears, offering a reference for detecting small objects with backgrounds of similar color, and provided effective technical support for the automation of pear harvesting.

Abstract:To achieve high-precision detection of four major cassava leaf diseases in complex unstructured environments, an improved algorithm for cassava leaf disease neural network detection based on the selective attention mechanism, MAISNet, was proposed. Using V2-ResNet-101 as the base network, the multiattention algorithm was firstly used to optimize the weighting coefficients, adjust the semantic expression of the feature channels, and the semantic feature saliency expression of cassava leaf disease in the feature map was preliminary constructed; then the instance batch normalization method was used after the residual unit to suppress the covariate offset in the feature expression, highlight the target semantic feature expression in the feature map, and realize the high-quality semantic feature expression. Finally, the Squareplus activation function was used to replace the ReLU activation function in the residual branch to maintain the numerical distribution of semantic features in the negative domain, and reduce the truncation errors in the feature fitting process. The results of the comparison test showed that the MAISNet-101 neural network constructed after the above improvement achieved an average accuracy of 95.39% for the detection of four common cassava leaf diseases, which was significantly better than the performance of the mainstream algorithms such as EfficientNet-B5 and RepVGG-B3g4. The results of the visualization and analysis of the extracted features of the network showed that high-quality semantic feature saliency representation of cassava leaf diseases was the key to improve the accuracy of cassava leaf disease detection. The proposed MAISNet neural network model can accomplish high-precision detection of cassava leaf diseases in real scenarios, which can provide technical support for precise drug application.

Abstract:An innovative approach that integrated series fusion of scatter correction techniques with deep learning models was introduced to achieve rapid and precise classification of common bunt in wheat kernels. Manual identification of this disease can be particularly challenging, especially in cases with mild infections. To address this challenge, the high-spectral data was leveraged from a sample set comprising 300 kernels, encompassing healthy, mildly infected, and severely infected kernels. The original spectra underwent preprocessing by using the multiplication scatter correction (MSC) and standard normal variate (SNV) algorithms. Furthermore, two-dimensional correlation spectroscopy (2D-COS) analysis was employed to assess the complementarity between spectra processed by SNV and MSC. Subsequently, the series fusion of scatter correction techniques was applied to amalgamate the original spectra, SNVprocessed spectra, and MSC-processed spectra, resulting in fused spectral sequences that harnessed the complementary information from various spectral preprocessing methodologies. Following this, a classification model for wheat common bunt, based on the ResNet 50 algorithm, was developed by using the fused spectral data. Experimental results demonstrated that the ResNet 50 model achieved the highest classification accuracy of 93.89% and an F1-score of 93.87%, surpassing models based on individual preprocessing methods. To further evaluate the classification performance of the model, partial least squares discriminant analysis (PLS-DA), support vector machine (SVM), and ensemble learning algorithms, random forest (RF), and extreme gradient boosting (XGBoost) models were constructed by using the fused spectral data and comparison was done. The results revealed that SVM, PLS-DA, RF, and XGBoost achieved overall recognition accuracies of 81.67%, 84.44%, 89.44%, and 90.55%, respectively, with corresponding F1-scores of 81.59%, 84.04%, 89.49%, and 90.59%. Importantly, the ResNet 50 model outperformed traditional spectral analysis models in terms of overall accuracy and F1-score. In summary, ResNet 50 outperformed traditional spectral analysis models in terms of both overall accuracy and F1-score. In conclusion, this research underscored the efficacy of combining series fusion of scatter correction techniques with deep learning models for the classification of common bunt in wheat kernels at varying infection levels. This approach held promise for the development of rapid and non-destructive detection methods for common bunt in wheat kernels.

Abstract:Aiming at the problems of feature sparsity, terminology and difficulty in capturing semantic associations within the text in question intention recognition in the tobacco domain, a feature fusion method based on sentence-bidirectional encoder representational from transformers-Attention mechanism-latent dirichlet allocation (SBERT-Attention-LDA) and multi layers-long short term memory (ML-LSTM) feature fusion was proposed. The method first dynamically encoded the tobacco question based on the SBERT pre-training model combined with the Attention mechanism and converted it into semantic-rich feature vectors, and at the same time, the topic vector of the question was modelled by using the LDA model to capture the topic information in the question; and then the joint feature representation with more complete and accurate question semantics was obtained by using the modified model-level ML-LSTM feature fusion method; and then the three-layer LSTM and ML-LSTM feature fusion method was used to identify the intention of the question. Then a 3-channel convolutional neural network (CNN) was used to extract the hidden features in the hybrid semantic representation of the question and fed them into the fully connected layer and Softmax function to achieve the classification of the question intent. Compared with the enhanced representation through knowledge integration and embedding (BERT and ERNIE) CNN models, the improvement was obvious (the F1 values were improved by 2.07 percentage points and 2.88 percentage points, respectively), which supported the construction of the Q&A system for tobacco websites.

Abstract:The behaviors of cows, including eating, lying, standing, walking and tailflicking, which directly or indirectly reflects the health and physiological condition of the cows. It is necessary to monitor cow diseases and detect anomalies in cow behavior. In order to achieve the goals, a multi-branch parallel CAFNet (ConvNeXt-ACM-FAM) cow behavior recognition model was proposed by combining temporal and spatial attention information. The model combined an asymmetric multi-branch convolutional module (ACM) and a feature attention module (FAM) on the basis of a ConvNeXt convolutional network. Firstly, ACM was utilized to partition channel branches for feature extraction, and retained some original features to prevent excessive information loss. And ACM can improve the running efficiency of the model. Secondly, FAM fused the features from different channels and introduced the SimAM attention mechanism, which enhanced the efficient extraction of important features without increasing network parameters and improved recognition accuracy. The result of experiment demonstrated that the CAFNet achieved recognition accuracy of the method for eating, lying, standing, walking, and tailflicking was 95.50%, 93.72%, 90.26%, 86.43%, and 89.39%, respectively. And the average recognition accuracy was 91.06%. Compared with the original model, the number of parameters was reduced by 1.5×106, the computational complexity was reduced by 3×108, and the average recognition accuracy was increased by 8.63 percentage points. The results can provide technical support for cow disease monitoring and prevention.

Abstract:With the rapid development of animal husbandry in China, the transition from farmers-dispersed cattle breeding to precision husbandry has become increasingly important. Efficient management of breeding, behavior monitoring, disease prevention, and health assurance pose significant challenges. Traditionally, farmers have struggled to provide adequate attention to each cow. To address these challenges, a comprehensive approach was developed that accurately identified and tracked cattle behavior by analyzing behavior patterns and visual characteristics. Firstly, the improved YOLO v8 algorithm was employed for cattle target detection. The model’s feature extraction capabilities were enhanced by incorporating the C2f-faster structure into the Backbone and Neck. The upsampling operator CARAFE was introduced to expand the perception field for data feature fusion. To identify small area characteristics of young cattle, the BiFormer attention mechanism was integrated into the detection process, replacing the dynamic target detection head DyHead. This allowed to effectively integrate scale, space, and task perception. Furthermore, the issue of the uneven distribution of positive and negative samples and the limitations of CIoU was addressed by utilizing the Focal SIoU function. Finally, the behavior category information detected by YOLO v8 was incorporated into the BoTSORT algorithm to enable multi-target behavior recognition and tracking in complicated situations. The experiments demonstrated significant performance improvements. The proposed FBCD-YOLO v8n model outperformed both the YOLO v5n, YOLO v7tiny, and the original YOLO v8n models, with an increase of 3.4 percentage points, 3.1 percentage points, and 2.4 percentage points in mAP@0.5, respectively, on the bovine behavior dataset. Notably, the accuracy of bovine back licking behavior recognition was increased by 7.4 percentage points. Regarding tracking, the BoTSORT algorithm achieved an MOTA of 96.1%, MOTP of 78.6%, HOTA of 78.9%, and IDF1 of 98.0%. Compared with ByteTrack and StrongSORT algorithms, the proposed method of MOTA and IDF1 scores demonstrated significant tracking improvements. This research demonstrated that the multi-objective cattle behavior recognition and tracking system developed can provide effective assistance to farmers in monitoring cattle behavior within the cattle barn environment. It offered crucial technical support for automated and precise cattle breeding.

Abstract:Efficient management of irrigation water, nitrogen fertilizer and planting area of rice is helpful to improve agricultural economic efficiency, resource utilization efficiency and ecological environment. Taking 13 cities (districts) in Heilongjiang Province as the study area, Meta analysis was used to quantify the effects of different irrigation methods and nitrogen application rates on rice yield and greenhouse gas emissions (CO2, CH4, N2O),and the water-fertilizer production function was established. Furthermore, a multi-objective optimization model was established with economic benefit, greenhouse gas emission and water and fertilizer utilization efficiency as objective functions to optimize the allocation of water and fertilizer resources in different regions and adjust the rice planting area. The results showed that controlled irrigation and application of nitrogen fertilizer had different effects on yield and greenhouse gas emissions. After optimization, rice planting area was reduced by 3.76% and water use efficiency was increased by 18.4%, the amount of irrigation was 4513.54m3/hm2. The nitrogen application rate was reduced by 11%, and the nitrogen use efficiency was increased by 32%, the amount of nitrogen fertilizer applied was 100kg/hm2. Economic benefits were increased by 8.1%, and environmental pollution was decreased by 10.6%. The model can quantify the response of controlled irrigation based water and fertilizer application to yield and greenhouse gas emissions at regional-scale, synergistically optimize water-land-fertilizer resource allocation in rice paddies, balance economics, greenhouse gas emissions, and resource efficiency, it was helpful to optimize water and nitrogen resources and adjust planting area between different targets of rice in Heilongjiang Province, so as to promote sustainable agricultural development. It can provide reference for the optimization and management of soil, water and fertilizer resources of rice.

Abstract:In order to seek the irrigation mode of greenhouse tomato land water conservation, emission reduction and superior yield, taking tomato (Jinpeng 8) as the research object, the experiment set I1 and I2 (corresponding to the crop-dish coefficient kcp 0.8 and 1.0) two irrigation levels, F1 and F2 (corresponding to the application of nitrogen 180kg/hm2 and 240kg/hm2) two nitrogen application levels with A1, A2 and CK (1 time and 2 times Venturi aeration, respectively, without aeration CK was used as control treatment) as three aeration levels, a three-factor completely randomized design with ten treatments, each treatment repeated three times, was used to monitor and analyze the greenhouse gas emissions during the whole life cycle of tomato by static dark box-gas chromatography, and to investigate the changing patterns of soil CO2, N2.O and CH4 emissions and tomato yield. The effects of irrigation level, nitrogen application level and gas addition level on the yield and greenhouse gas emission of greenhouse tomato were analyzed, and the net global warming potential (NGWP) and greenhouse gas intensity (GHGI) were synthesized, so as to put forward the greenhouse tomato water-fertilizer-airintegrated drip irrigation management mode with the goal of water conservation, emission reduction and high yield. The results showed that increasing irrigation level and nitrogen application level both increased soil CO2 and N2O emission fluxes, with an average increase of 24.8% （P<0.05) versus 14.8% （P>0.05) in the I2 treatment compared with that in the I1 treatment, and an average increase of 86% （P>0.05) versus 34.9% （P<0.05) in the F2 treatment compared with that in the F1 treatment. Aerated irrigation had a significant effect on soil CO2 and N2O emission fluxes, which increased by an average of 5.5% and 10.0% （P>0.05) in A1, 20.9% and 62.9% （P<0.05) in A2 treatments, respectively, as compared with that in CK treatment. Soil CH4 emission fluxes during the whole tomato reproductive period did not have a significant pattern of change, showing the soil as a sink for CH4, increasing irrigation level would increase soil CH4 emission fluxes, while increasing nitrogen application level would reduce CH4 emission fluxes, I2 treatment increased by 27.8% on average compared with I1（P<0.05), and F2 treatment decreased by 25.5% on average compared with F1（P<0.05). Aerating, the fertilization and irrigation significantly increased tomato yield （P<0.05). Considering the economic and ecological factors, the benefits of A1F2I1 treatment were the best, the combination strategy of aerating level A1, applying nitrogen level F2, and irrigating level I1 can take into account the requirements of water conservation and superior yield reduction, and provide a reference for the better irrigation mode of greenhouse tomato in Northwest China.

Abstract:To investigate the effects of flooding stress on the growth and yield of double cropping rice under different sediment contents during the flood period in the Poyang Lake Plain area, a combination of field experiments and indoor analysis was used. Two types of flooding depths, including 2/3 plant height and full flooding, and three sediment contents, S1(0kg/m3), S2 (0.5kg/m3), and S3 (1.0kg/m3), were set up. The growth indicators and rice yield of early and middle rice were observed and studied 6 d and 9 d after flooding. The results showed that moderate flooding stimulated rice elongation, continuous internode differentiation, and increased the leaf elongation and width. The height and internode length of flooded plants were increased by 9.35% and 12.75% during the heading-flowering stage of early rice, and the leaf area under 2/3 flooding was increased by 11.00%. However, excessive stress inhibited rice growth. The plant height and tiller number of middle rice were decreased by 33.49% and 29.28%, and the leaf area was decreased by 30.94%. At this time, it was difficult for middle rice to extend beyond the water surface, leading to severe sand deposition on functional leaves. The increase in sediment content further inhibited rice growth. Flooding resulted in an average reduction of 32.35% and 58.72% in dry matter weight of early and middle rice panicles （P<0.05). The decrease in seed setting rate and thousand grain weight was the main reason for the decrease in yield of early rice after being flooded during the heading-flowering stage, and the influence of sediment content and flooding time was not significant at this time. Under full submergence of sediment, the yield reduction of medium rice was intensified. The yield reduction rate under S2 and S3 was significantly increased by 31.63% and 52.20% compared with that under S1 （P<0.05). At this time, the yield decrease was the result of the comprehensive effect of spike length, effective panicle number, grain number per panicle, seed setting rate, and thousand grain weight. The research results can provide theoretical and technical support for flood disaster management and food security guarantee in the Poyang Lake Plain area.

Abstract:Exploring the impact of diffuse radiation changes on the gross primary productivity (GPP) of paddy ecosystem can provide a reference for the assessment of paddy carbon sink capacity and yield estimation. The eddy covariance (EC) system was used to measure the CO2 flux in the double-cropping paddy field in the Poyang Lake Plain for two consecutive years (2017—2018). The data from the mid-season period of rice was segmented according to the diffuse fraction (DF), and the effect of diffuse radiation and other meteorological factors on GPP under different DF conditions were explored. The results showed that the effects of different types of radiation on GPP were different. For both the early rice and the late rice, GPP and PAR showed a quadratic curve relationship (R2 was 0.49 and 0.70). Under different DF conditions, the variation trend of GPP of early and late rice with diffuse photosynthetically active radiation (PAR-dif) was different. When DF was 0.1~0.4, GPP of early rice did not change significantly with PAR-dif, GPP of late rice was increased with PAR-dif (R2=0.23), and GPP of early and late rice was decreased when DF was 0.4~0.7 (R2 was 0.38 and 0.02). When DF was 0.7~1.0, GPP of early and late rice showed a significant upward trend with PAR-dif (R2 was 0.32 and 0.89), indicating that PAR-dif was an important factor affecting GPP of rice. As for the direct photosynthetically active radiation (PAR-dir), when PAR-dir was between 0~10mol/(m2·d), GPP was increased rapidly with the increase of PAR-dir, and then tended to be stable. With the increase of DF, GPP and DF showed a quadratic curve relationship (R2 was 0.45 and 0.67), while the light use efficiency (LUE) and DF showed a significant linear positive correlation (R2 was 0.68 and 0.82). The optimal DF of the early and the late rice were 0.48 and 0.40, respectively. The changes of DF caused the changes of meteorological factors such as air temperature (Ta) and water vapor pressure deficit (VPD), which had a synergistic effect on rice GPP. The results of path analysis between meteorological factors and GPP showed that the effects of meteorological factors on GPP varied under different DF conditions. Overall, the increase of Ta and VPD had positive and negative effects on rice GPP, respectively. Ta, PAR-dir and PAR-dif were the main meteorological factors affecting GPP for the early rice, and PAR-dif, PAR-dir and PAR-dif were the main meteorological factors for the late rice when DF was 0.1~0.4, 0.4~0.7 and 0.7~1.0, respectively.

Abstract:An aerobic incubation method with a constant temperature was used to investigate the effects of different fertilization measures and soil types on soil nitrogen mineralization characteristics of typical farmland in Northwest China.Four fertilization treatments: no fertilizer (CK), single urea application (U), single organic fertilizer application (M), and urea combined with organic fertilizer (U+M) were set up to explore the dynamic process of nitrogen mineralization in different soil types. And the first-order kinetic equation fitting and correlation analysis were carried out on the soil accumulative mineralized nitrogen. The results showed that fertilization treatments and soil type significantly affected soil ammonium nitrogen, nitrate nitrogen, and accumulative mineralized nitrogen contents. And there were significant interactions between fertilization and soil type. The orders of accumulative mineralized nitrogen content and mineralization rate of different soil types were Lou soil, Loess soil, Yellow River irrigation soil, and Grey brown desert soil. Compared with CK treatment, different fertilization treatments significantly increased soil accumulative mineralized nitrogen, mineralization rate constant (k) and mineralization potential (N0), and the differences among treatments were significant （P<0.05). The accumulative mineralized nitrogen content and mineralization rate of single urea application and urea combined with organic fertilizer were 2.83~6.71 times and 3.83~7.70 times higher than that of CK treatment, respectively. Correlation analysis showed that soil accumulative mineralized nitrogen was significantly and positively correlated with soil organic matter and total nitrogen contents. The results illustrated that combined application of organic and inorganic fertilizers could significantly promote nitrogen mineralization in different soil types in Northwest China, improve nitrogen availability and supply capacity, and help maintain soil mineral nitrogen content, which played an important role in the efficient utilization of farmland nitrogen.

Abstract:Aiming at the problem of low agricultural water price, Heilongjiang Province, the main grain producing area, was selected as the study area. Starting from the full-value and full-cost of agricultural water, the value of agricultural water use was analyzed by using the emergy theory, the cost of water use was determined by using the full-cost water pricing method, and the pricing of agricultural water resources was formulated in different situations by combining with the current price of water. A double logarithmic model was used to establish the price function of agricultural water demand, reveal the potential for water saving in the study area brought about by the increase in water price, and determine the affordability of farmers by combining with the index analysis method, so as to judge the feasibility of the water pricing scheme formulated. The results showed that in 2020, the full-value of agricultural water resources was 0.594 yuan/m3, and the full-costs of surface water and groundwater were 0.180 yuan/m3 and 0.355 yuan/m3, respectively; for the three designed ladder water price adjustment schemes, empirical analyses were carried out by using the years of 2005, 2010, 2015, and 2020, and in the case of the transition period and the long-term water price scheme, the average annual water savings respectively can be up to 2.72×109m3 and 4.45×109m3, the average annual water saving potential was 12.74% and 19.48%, and the first-order water price was in the range of the affordability of farmers, which was determined as a feasible program. The research result can provide support for the comprehensive reform of agricultural water price in the study area, which can be combined with the actual situation to gradually increase the agricultural water price to a reasonable level.

Abstract:The solar greenhouse structure helps indoor lighting and heat storage, ensuring the normal growth of crops. However, there is variability in environmental parameters at different times and locations in the greenhouse, and they vary with weather and seasons. To reveal the spatiotemporal variation patterns of light intensity, air temperature and humidity in the canopy, an environmental monitoring system based on wireless sensor networks was built. Nodes with environmental information sensing functions such as light intensity, air temperature and humidity were deployed in the crop canopy to analyze the temporal and spatial variation of environmental parameters. Firstly, the inverse distance weighted algorithm was used to construct discrete data surface of canopy light intensity, air temperature and humidity. Secondly, the K-means clustering based on centroid coordinates of the interpolation results was carried out to calculate the position of the feature points of connected and non-connected areas in the greenhouse. Finally, the semi-variogram method was used to analyze the spatiotemporal variability of the monitoring parameters of the interpolation nodes. The experimental results showed that the solar greenhouse in summer presented high temperature and high light in the afternoon. The light intensity at 08:00 and 16:00 was 24.2% and 72.9% of that at 12:00, respectively. The air temperature at 08:00 (27.7℃) was about 6℃ lower than that at 12:00 and 16:00, and the air humidity (90%) was about 30% higher. The maximum light intensity in sunny day was 1.4 times of that in cloudy day and 4.6 times of that in rainy day. The maximum air temperature in sunny day and cloudy day was about 6℃ higher than that in rainy day (29.5℃), and the minimum air humidity were lower than that in rainy day (84%). The solar greenhouse presented high temperature and low humidity in both sunny day and cloudy day, and high humidity and low light in rainy day. The range of light intensity was 10.34m, and the spatial variability was strong. The spatial variability of air temperature and humidity was weak, and the overall distribution was relatively uniform. The temporal variability of light intensity, air temperature, and air humidity were moderate. The characteristic points and spatial and temporal variation patterns of environmental parameters contributed to the efficient deployment of solar greenhouse sensors and provided a basis for revealing the interaction between crops and the environment.

Abstract:In order to reveal the law of gas evolved from the pyrolysis and gasification processes of stalk biomass, the typical corn stalk in the rural areas of northeastern China was utilized as the pyrolysis experimental material. Based on the self-established pyrolysis and gasification experiment system in tube furnace, the release characteristics of CO, H2, CO2, CH4, CnHm and other small-molecule biomass gas components during the pyrolysis of corn stalk in nitrogen atmosphere and gasification in oxygen-containing atmosphere were systematically studied. The effects of different pyrolysis and gasification temperatures on the release characterization and yield of each syngas component were compared. The experiment results indicated that CO and CO2 were the first small molecule syngas products released during corn stalk pyrolysis. When the temperature was increased, CH4 and H2 gradually appeared in the syngas, and with the increase of pyrolysis temperature, the peak yield of CO first appeared in the heating stage while the peak yields of CO2, CH4 and H2 appeared almost simultaneously in the constant temperature stage. With the increase of pyrolysis temperature, the volume fraction of CO during pyrolysis was hardly changed. However, the CO2 proportion was decreased with the increase of temperature. The CH4 volume fraction was increased with the increase of pyrolysis temperature between 400℃ and 500℃, and the content basically stabilized at 13% after 500℃. In the oxygen-containing atmosphere with 8% O2 and 92% N2, the volume fraction of CO2 produced by corn stalk gasification showed a trend of first increasing and then decreasing with the increase of gasification temperature, while the volume fraction of CO was increased with the increase of temperature, indicating that the high temperature was more conducive to the release of CO, and the low temperature was favorable to the CO2 production.

Abstract:Nitrogen fertilizer was excessively applied in the agricultural production, which resulted in soil secondary salinization and deterioration. Biochar has been gradually applied in soil restoration due to adsorption capacity. However, solanaceous vegetables straw-derived biochar was rarely reported. The present study aimed to explore the effect of solanaceous vegetables straw-derived biochar on adsorption of nitrate nitrogen and mitigation of soil secondary salinization. Straws of sweet pepper, tomato and eggplant were prepared as biochar by pyrolysis. Adsorption and mechanism for nitrate nitrogen removal by experimental solanaceous vegetables straw biochar was examined. Surface morphology and functional groups were characterized by scanning electron microscopy (SEM) and Fourier transform near-infrared spectroscopy (FTIR). The adsorption process of nitrate nitrogen was simulated and fitted by kinetic model and isothermal adsorption model. The adsorption mechanism of biochar was analyzed according to morphology and model parameters. All of the experimental solanaceous vegetables strawderived biochar showed adsorption capacity of nitrate nitrogen. The maximum adsorption capacity of eggplant, tomato and sweet pepper straw-derived biochar were 114.788mg/g, 29.736mg/g and 9.759mg/g, respectively. The adsorption processes of eggplant and sweet pepper straw biochar were well fitted by quasi-second-order kinetic model, which was controlled by the integrated adsorption of chemical bond, micro-porefilling and internal diffusion. The adsorption process of tomato straw biochar was well fitted by the quasi-first-order kinetic model, which was mainly physical adsorption. Functional groups such as hydroxyl group, methyl group, methylene, carboxyl group and carbonyl group were observed in experimental biochar, according to FTIR analysis. Additionally, ether bonds were observed in eggplant and sweet pepper straw biochar and alcohol hydroxyl groups were observed in tomato straw biochar. Therefore, adsorption capacity of nitrate nitrogen was observed in the experimental solanaceous vegetables straw biochar. The eggplant straw biochar had the greatest adsorption capacity of nitrate nitrogen, which was affected by various physicochemical mechanisms such as pore filling, functional groups and complexation. The results demonstrated that eggplant straw biochar had great potential in amendments of secondary salinization soil. The present study provided insight into the effective utilization of solanaceous vegetables straw and soil remediation in crop production.

Abstract:An effective aeration and mixing system composting system was created to address the problems of slow temperature rise and incomplete composting. The system consisted of a segmented aeration mechanism and a combined mixing device for dissolved oxygen. The composting process was designed in a stepwise composting process through the structural design of the two main devices, and the strength of the aeration device and mixing device of the aeration and mixing system were verified by using EDEM (DEM) and ANSYS (FEM) software. The results showed that the maximum stress value in the aeration system appeared in the aeration pipe with 13.064MPa and the maximum deformation was 0.038126mm, and the maximum stress value in the mixing device appeared in the mixing blade with 190.31MPa and the maximum deformation was 0.34417mm, which was in accordance with the design requirements. On this foundation, a 14 day composting test using food waste and maple leaves as feedstock was carried out to test the performance of the aeration and mixing system, record the changes in key parameters of the composting process and measure the relevant indexes of the fermentation production. The test results showed that the compost pile temperature of the rotary composting reactor with this aeration and mixing system reached 53.34℃ on the third day, the maximum temperature of the compost pile reached 69.56℃, and the high temperature period (＞50℃) was maintained for 6d. At the same time, the water content of the product reduced 27.21%, the pH value was increased to 8.4 and the maximum seed germination index reached 131.4%, which met the standards for organic fertilizers (NY/T 525—2021)，and the rotary composting reactor operating costed only RMB 65.51 yuan/t.

Abstract:With the development of blockchain technology in the field of food traceability, the quality and safety of prepared foods have been effectively guaranteed. However, the numerous industrial characteristics of the upstream and downstream production and processing links of prepared foods made traceability difficult. How to safely share recipes, secret recipes and other private data while tracing the source is very important to enable better collaborative production in the supply chain. In order to solve the above problems, a threshold proxy re-encryption privacy data sharing method for prepared food traceability was proposed, and prepared food traceability chain blocks were designed and the production batch number associated related traceability data was traced. Using the threshold proxy re-encryption, the prepared food manufacturer re-encrypted the recipes developed through the threshold proxy to encrypt the recipe to generate the initial encryption ciphertext. At the same time, the re-encryption key was generated locally, and the initial encryption ciphertext and the re-encryption key were generated locally and uploaded to the prepared food traceability blockchain; third-party semi-honest service providers would re-encrypt the re-encrypted materials obtained from the blockchain and upload the re-encrypted ciphertext to the blockchain; data visitors such as regulatory authorities used the own private key would re-encrypt and decrypt the re-encrypted ciphertext obtained from the blockchain to achieve safe sharing of private data on the traceable blockchain. This method was based on Hyperledger Fabric, and a blockchain prototype system for full-process traceability of the supply chain of prepared foods was constructed for testing. Test results showed that the threshold proxy re-encryption method proposed showed lower computational overhead compared with commonly used data encryption sharing methods in secure sharing of private data in prepared food traceability. The average delays of data uploading, public data query and private data query were 1473.8ms, 63.9ms and 59.9ms, respectively. The system performance was good. Experiments showed that the proposed method achieved the purpose of secure sharing of supply chain privacy data, ensured the safe sharing of business secrets and intellectual property rights, which was of great value to the development of the prepared food industry and the improvement of food safety.

Abstract:In order to improve the quality of low-alcohol Zaosu pear-Merlot wine, Zaosu pear and Merlot grape juice were blended in a volume ratio of 50∶50 as test materials. Yeast cell wall, water soluble β glucan and mannoprotein were added to the blended juice at a dose of 0.25g/L respectively before alcohol fermentation, and Saccharomyces cerevisiae as well as non-Saccharomyces yeast were inoculated for mixed fermentation. The volatile aroma compounds in fermented fruit wine were determined by HS-SPME-GC-MS, and their effects on the aroma quality of wine were analyzed by using fuzzy mathematics sensory evaluation method. The results showed that all three kinds of yeast polysaccharides had positive effect on the alcohol fermentation kinetics and physicochemical index of low-alcohol Zaosu pear-Merlot wine, especially the significant difference in the color parameter CIELab between the fruit wine added with yeast mannoprotein and the control. In addition, the addition of yeast mannoprotein could increase the content of terpenes, esters and higher alcohols in fruit wine, especially ethyl 2-methylbutyrate, ethyl butyrate, phenethyl acetate, ethyl decanoate, citronellol and geraniol with floral and fruity flavor. Based on fuzzy mathematics evaluation, the sensory score of fruit wine added with yeast mannoprotein reached 7.400 points. In summary, exogenous addition of yeast mannoprotein could effectively stabilize the color of lowalcohol Zaosu pear-Merlot wine and improve its aroma quality.

Abstract:Soybean protein isolate (SPI) was used as raw material for hydrolysis with Alcalase (0~180min).The structural changes of enzymatic hydrolysis products were investigated by gel electrophoresis, Fourier transform infrared spectroscopy (FT-IR) and intrinsic fluorescence spectra. The interfacial behavior of enzymatic hydrolysis products was described by surface tension and adsorption of interfacial protein, and the influence of structural changes and interfacial behavior on the foam properties was analyzed. After enzymatic hydrolysis, the typical 7S and 11S bands disappeared and new bands were produced (about 24ku). Compared with SPI, the content of α-helix was decreased, the content of β-turn and random coil was increased in hydrolysate, and the fluorescence wavelength was red shifted. These results indicated that the protein structure was unfolded, which in turn promoted the change of protein function. The results showed that the foamability of the sample was the best (143.20%) at 90min,which may be due to the lowest average particle size (208.10nm), high solubility (90.44%) and lowest surface tension of the hydrolysate at this time, which was conducive to improving the adsorption rate of the hydrolysate at the air-water interface. However, due to the small peptides produced by enzymolysis, the ability of protein network structure was lost, which had a negative impact on foam stability. In addition, the antioxidant activity of the protein was greatly improved by enzymatic hydrolysis. The foamability of SPI can be effectively improved through enzymatic hydrolysis, and the application range of SPI as an effective foaming agent in food was expanded.

Abstract:Engine performance variations and traction fluctuations have a great impact on the adaptability of shift strategies for high-power power shift tractors (PSTs). In order to construct a dynamic accurate model and deal with traction fluctuation to improve the adaptability of PST shift strategy, an adaptive shift strategy development method was proposed based on digital twins. On one hand, the engine state change was regarded as an internal disturbance, and the virtual PST engine was calibrated in real time based on the deep deterministic strategy gradient algorithm, which was combined with the PST mechanism model to realize the real-time dynamic and accurate modeling of the PST. On the other hand, the traction fluctuation was treated as an external disturbance, and a deep Q-network was used to automatically generate the shift strategy. Finally, the virtual PST training simulation under plowing conditions and the speed tracking comparison test between the proposed method and the fuzzy adaptive method were carried out. The results showed that the average tracking errors of engine torque and fuel consumption rate did not exceed 7.28N·m and 1.55g/(kW·h), and dynamic and accurate modeling of physical PST was achieved. After using for a long time, the changes of engine and traction force caused that the shift points and fuzzy rules of the fuzzy adaptive method were no longer fully applicable, and the shift performance was gradually deteriorated. In contrast, the shift performance and speed tracking effect of the proposed method were good throughout, and the mean value of speed tracking error, mean value of fuel consumption rate, and total number of shifts were 0.0125m/s, 229.76g/(kW·h),and 42, respectively, which were reduced by 0.91%, 11.14%, and 34.38% compared with those of the fuzzy adaptive method. The adaptability and superiority of the proposed method were verified.

Abstract:Based on the design theory and methodology of parallel mechanisms (PM) based on position and orientation characteristics (POC) equations, a two-translation-rotation (2T1R) PM was designed. It consisted of low pairs and possesses symbolic forward solutions as well as partial motion decoupling. The primary topological features of the PM, including POC, degree of freedom, coupling degree, and motion decoupling were analyzed. Subsequently, based on the kinematic modeling principle derived from topological characteristics, symbolic position forward and inverse solutions for the PM were obtained. Simultaneously, singularity analysis was conducted by using the inverse position solution while solving for the workspace of the PM based on symbolic solutions. Furthermore, employing a sequential single-open-chain method grounded in virtual work principles enables dynamic performance analysis of the PM along with calculation of actuated forces exerted on its three driving pairs. The maximum driving forces required for the three sliders were -58.52N, 47.28N and 64.10N, respectively. Ultimately, this PM can be utilized as an end-effector and safety lander for UAVs; their conceptual design was elaborated upon. The research can provide a theoretical basis for kinematics and dynamics modeling and analysis of 2T1R parallel mechanism symbolized by positive solution and kinematically decoupled, as well as mechanism performance optimization and prototype development.