Abstract:The soil-contact components in agricultural machinery, whose performance directly determined operating efficiency and service life, and whose materials development was key to advancing agricultural equipment were investigated. It presented a systematic review of the research progress and current applications of soil-contact component materials. Firstly, it analyzed the loading and damage mechanisms of these components under typical operating conditions, such as sand and clay soils, and paddy fields, and clarified core performance requirements, including wear resistance, impact resistance, and anti- adhesion properties. Secondly, it traced the technology system from traditional steel materials ( spring steel, boron steel) to modern surface coating technologies ( hardfacing, cladding, thermal spraying, chemical surface treatments, etc. ) and to new materials (polymers, high-entropy alloys), and compared the gaps between domestic and international research and applications. Drawing on representative domestic and international case studies, it summarized evaluation methods for material performance in field tests, laboratory experiments, and numerical simulations. Finally, it identified the challenges that current materials faced in extreme operating conditions, multi-property synergy, and industrialization, and outlined future development directions in terms of industrial upgrading, condition-driven design, intelligent and green manufacturing, and differentiated competition. The aim was to provide a systematic theoretical reference for the development and engineering application of high-performance soil-contact component materials.

Abstract:High-quality seeding operation is a key link in the seeding process of precision agriculture and smart agriculture, which is mainly completed through the coordination of processes such as furrow opening, seed metering and guiding, soil covering and soil compacting. The working performance of soil- engaging components such as furrowing, soil-covering, and pressing affects the final seeding quality by acting on the various stages of the seeding process. Based on the structure and types of the soil-engaging components in seeding operations, the basic structures and forms of existing typical and a typical products were summarized. Focusing on the improvement of operational quality such as low disturbance, anti- adhesion and anti-sticking of soil-engaging components, the research on component-soil interaction mechanism, the adjustment mechanism, and the intelligent monitoring of operating status, the research status of relevant technologies in China and abroad were elaborated in detail. Based on the above research status, the main existing problems of soil-engaging components in seeding operations were identified as follows: a single structural form of components, insufficient industrialization transformation, weak adaptability to high-speed operations, inadequate intelligent regulation and control capabilities, and lagging material and process levels. Totally five development prospects were summarized: modularization and adjustable technology for soil-engaging components tailored to meet the needs of multiple agricultural modes, bionic drag reduction and lightweight structure technology for efficient and low-consumption soil- engaging components, dynamic matching and speed adaptive control technology suitable for high-speed operation scenarios, closed-loop regulation and perception decision-making technology for active operation and intelligent linkage, and advanced wear-resistant material preparation and high-precision forming process technology.

Abstract:Agricultural machinery serves as the backbone of modern agricultural production, as its operational performance and reliability directly impact national food security. As the mechanization of agriculture and the upgrading of agricultural equipment continue to advance, there is an increasing demand for enhanced performance in soil-engaging component, such as plowshares, tillage tools, and trenchers. Soil adhesion during operation leads to increased resistance and energy consumption, resulting in irregular furrowing, inadequate soil breakup, and uneven compaction, ultimately diminishing operational quality and efficiency. Thus, addressing soil adhesion in soil-engaging component is essential for maximizing the performance potential of agricultural equipment. The mechanisms of soil adhesion formation and the factors influencing it were systematically analyzed. A comprehensive review of the current state of research, both domestically and internationally, on technologies for reducing adhesion and facilitating detachment in agricultural machinery was provided. These included vibration, pneumatic methods, heating techniques, structural optimization, surface treatments, and biomimetic designs. Despite significant advancements in research aimed at reducing soil adhesion on agricultural machinery, current technologies still had considerable room for improvement to meet the high-quality development demands of agricultural equipment in China. While vibration, pneumatic, thermal, and structural adhesion reduction techniques showed effectiveness, they often require additional power sources or auxiliary systems, complicating machinery design and increasing energy consumption. Furthermore, many of these techniques were optimized for specific soil conditions, lacking comprehensive studies on their adaptability across diverse soil types and moisture levels. Surface engineering technologies, while capable of producing coatings with excellent initial adhesion reduction properties, often suffered from performance degradation due to wear and mechanical impacts during actual operations. Existing research tended to focus too heavily on optimizing single properties of coatings, with insufficient attention to the synergistic enhancement of wear resistance and anti-adhesion capabilities. Additionally, biomimetic approaches, though promising, often lack universal design models that can adapt to various conditions, and they required further exploration of cost implications and complexity in manufacturing processes. The operational conditions for soil-engaging component were harsh, and understanding the complex mechanisms of soil adhesion was vital for enhancing their adhesion reduction and detachment performance. Future research should focus on dynamic, multi-field coupling studies of soil-component interfaces, the development of specialized low-energy adhesion reduction materials, and the transition from static to dynamic biomimetic designs, fostering innovative, cost-effective solutions that ensure high reliability.

Abstract:Aiming to improve the wear resistance of the 30MnB5 ditch knife, three types of Fe-based alloy cladding layers were fabricated on its surface using rectangular spot laser cladding technology. The phase composition and microstructure of the cladding layers were analyzed by using X-ray diffraction and scanning electron microscopy. Wear resistance was tested through wet sand rubber wheel and self- developed soil incorporation component wear tests. The results indicated that the three cladding layers exhibited good metallurgical bonding with the substrate. and there were no defects such as cracks or pores in the No. 2 and No. 3 cladding layers. There were a large number of cellular crystals and hard phases inside the cladding layers. The cladding layers were primarily composed of Fe C solid solutions, Mo61 C39, Mo18C7, V4C3, (Fe, C), WC, SiC, and Ni2 B, among other phases. Both the wet sand rubber wheel wear test and the self-developed soil incorporation component wear test demonstrated that the 3 cladding layers exhibited the best wear resistance, attributed to its high Mo content which promoted a fine cellular crystalline structure. Additionally, the appropriate amounts of V and W elements resulted in a dispersed distribution of VC, WC, and Mo-based carbides, enhancing the wear resistance. The wear mechanism for all three cladding layers was predominantly abrasive wear, accompanied by adhesive wear and oxidative wear. The composition of the No. 3 Fe-based alloy significantly improved the wear resistance of the ditch knife, combined with the substrate of 30MnB5, providing technical support for the manufacturing of low-power and long-life agricultural chain trenching components.

Abstract:When paddy field pulping machine blades operate in a muddy environment, they are subjected to friction, impact, and erosion by soil particles, which easily lead to severe wear and a short service life. The microstructures on the shell surface of intertidal Scapharca broughtonii—possessing excellent wear resistance—were selected as the bionic prototype to carry out design and experimental research on the bionic wear resistance of pulping machine blades. The “ rib-groove” structural features of the Scapharca broughtonii shell surface were investigated by using a 3D microscope, and the surface flow field and particle movement trajectories were analyzed to reveal the wear resistance mechanism. Based on the discrete element method (EDEM), a “blade-soil” interaction model was established to analyze the influence of “rib-groove” structural parameter variations on normal and tangential contact energy. The Box-Behnken response surface method (RSM) was adopted for multi-objective parameter optimization of groove depth, groove width, and rib width. The simulation results showed that when the groove depth was 0.27 mm, the groove width was 0.68mm, and the rib width was 0.18mm, the normal cumulative contact energy (1.406J) and tangential cumulative contact energy (0.85J) of the bionic blade were minimized. Bionic blades were prepared based on the optimized parameters, and field comparative tests were conducted. The results indicated that the wear mass and wear rate of the optimized bionic blade were 2.048g and 0.383% , respectively. Compared with conventional smooth blades, the wear rate was reduced by 65.53% . The research result can provide a theoretical basis for the wear resistance research of soil-contacting components in paddy field machinery.

Abstract:Aiming to address the characteristics of fragmented farmland plots, heavy clay soil, and extensive straw coverage ( primarily whole-stalk coverage) in the southwestern region, a bionic drag- reduction anti-clogging device based on the Fermat spiral was designed to meet the requirements of no-till seeding for soybean-corn strip intercropping. The cutting-edge curve of the stubble-breaking blade was constructed based on sliding-cutting theory, and a wedge-shaped biomimetic drag-reducing unit was designed, inspired by the non-smooth surface structure of the Tenebrionidae elytra. A discrete element coupling model of “multi-crop residues-heavy clay soil-residue-clearing device” was established, and a second-order orthogonal rotation combination experiment was conducted to optimize key parameters, including the forward speed of the machine, the rotational speed of the stubble-breaking blade and the blade edge angle. The orthogonal experimental results showed that when the forward speed was 0.91m/s, the rotational speed of the stubble-breaking blade was 225r/min, and the blade edge angle was 25.2°, the number of bonding breaks between the stubble-breaking blade and root residues reached 21470, with an operational power of 5.82kW. Field tests demonstrated that under conditions of an average straw coverage rate of 94% and a whole-stalk coverage of 0.575 kg/m2, the device achieved an average cutting rate of 92.98% for corn-soybean straw, representing a 50.91% improvement compared with that of traditional notched disk blades. The device exhibited good passability during operation, meeting the requirements of no-till seeding in the southwestern region. The results revealed the synergistic mechanism between the sliding-cutting characteristics of the Fermat spiral cutting edge of the stubble- breaking blade and the biomimetic drag-reducing structure, validating the advantages of biomimetic curved stubble-breaking blade in reducing drag and preventing blockage under conditions of high straw coverage and heavy clay soil. The research result can provide theoretical support and technical foundation for the structural innovation and parameter optimization of key components of no-till seeder in hilly regions of southwestern China.

Abstract:Aiming at the problems of uneven seeding depth and excessive soil backfilling of the no-till grass seeder opener,a sliding blade type inverted T-shaped opener for no-till grass seeder was designed based on the principle of sliding cutting. The interaction between soil and the opener was analyzed based on particle dynamics, and the main factors affecting its working performance were determined. The inverted T-shaped forming block was designed based on the sand boa. The simulation results showed that the soil particle flow rate generated by the bionic forming block was 5.48% higher than that of the wedge-shaped forming block,and the working resistance was reduced by 15.88% . An EDEM simulation model was established,and the working speed,sliding cutting angle,and upper inclination angle were taken as the test factors,while the backfilling depth and the stability coefficient of the furrow depth were taken as the test indicators. A quadratic regression orthogonal rotation combination test was carried out. The optimal parameter combination of the opener was determined as: sliding cutting angle of 40.24°,working speed of 7.60 km/h,and upper inclination angle of 12.43°. At this time,the stability coefficient of the furrow depth was 88.64% and the backfilling depth was 19.43 mm. The soil trough test showed that the stability coefficient of the furrow depth was 87.21% and the average backfilling depth of the opener under the optimal parameter combination was 22.00 mm. The working resistance of the bionic forming block was lower than that of the wedge-shaped forming block,with an average resistance reduction of 8.66%. The field test results showed that within the working speed range of 6~10km/h, the stability coefficient of the furrow depth of the inverted T-shaped opener was higher than 80.86% ,and the backfilling depth was less than 26.00mm,indicating good furrow opening performance in grassland.

Abstract:Existing soil wear testing methods face several limitations, including restricted travel distance in linear soil bins, significant centrifugal effects and velocity differences between inner and outer radii in circular soil bins, and difficulties in soil preparation. To address these issues, a novel annular track-type soil wear testing platform was developed. The platform mainly consisted of an annular soil bin, a closed-loop guide rail, and a wear testing apparatus. The wear testing apparatus integrated functional modules for driving, attitude adjustment, and soil preparation, enabling reciprocating motion of soil-engaging components along the guide rail and facilitating long-distance soil wear experiments under multiple operating conditions. The operating speed (0~10km/h) was regulated via a programmable logic controller (PLC), while tillage depth, rake angle, and cutting angle can be precisely adjusted. Using a 1L-225 trapezoidal ploughshare as the test specimen, a long-distance wear experiment covering 140 km was conducted. A discrete element method (DEM) model describing the soil-ploughshare interaction was also established to validate the reliability of the platform in terms of load characteristics, soil flow field, and surface morphology evolution. Experimental results indicated that the fluctuation of penetration depth remained within ±4.5mm, while the coefficient of variation of the operating speed was only 2.1%. In addition, the soil firmness after preparation reached more than 80% of the measured field value. Reliability verification demonstrated that the measured draft force during the stable tillage stage showed a low relative error compared with simulation results, with consistent fluctuation phases. Wear morphologies observed in the soil bin tests, such as edge pitting and parallel furrow patterns, as well as the evolution of the ploughshare outer contour, showed strong agreement with the simulation results. Multi-condition experiments further verified the platform’ s capability to reproduce complex operating conditions. The proposed annular soil wear testing platform provided controllable operating parameters and stable tillage loads, enabling realistic reproduction of long-distance field wear failure processes. It therefore offered significant value for research on drag reduction and wear resistance of soil-engaging components, as well as for service life prediction.

Abstract:The phenomenon of adhesion between the soil-contacting components of tillage implements and the soil leads to a significant increase in energy consumption, severely restricting the overall quality and efficiency of agricultural machinery operations. Therefore, developing technologies that reduce the adhesive force between soil-contacting components and soil is an important research direction for improving the efficiency of agricultural machinery operations. Aiming to explore the effectiveness and feasibility of reducing soil adhesion on soil-contacting components through two approaches: structural design and electro-osmosis technology, based on biomimetic principles, by simulating the unique surface morphologies of toads and pangolins, a series of soil-contacting components with bio-inspired non-smooth surface structures, including press wheels and cylinders, were designed. Tensile tests were conducted to demonstrate that bio-inspired protuberance structures can effectively reduce the normal soil adhesion force on soil-contacting components. Meanwhile, electro-osmos technology was introduced, incorporating key parameters such as voltage value and energization time into the experimental scheme, systematically investigating their regulatory effects on soil adhesion characteristics. The experimental results showed that when no electro-osmosis measures were applied (i. e. , without voltage), the adhesion force between the soil and the soil-contacting components was significantly higher than that when voltage was applied. Moreover, the soil adhesion force was decreased as the voltage value and energization time were increased. Additionally, after conducting the electro-osmosis experiments, the soil-contacting surfaces of the press wheels and cylinders were in a clean state with no significant soil adhesion, and clear water film traces were observed. The research result confirmed the effectiveness of electro-osmosis technology in reducing soil adhesion forces on metal soil-contacting components, providing important theoretical support and technical backing for its potential application in agricultural machinery.

Abstract:Rice is the main food crop in China, and weeds seriously affect the growth and yield of rice. Aiming at the problems of large impact and insufficient structural stability of the existing paddy field weeding knife, focusing on the structural design and parameter optimization of the key components of the weeding knife, based on the bionic principle of canine tooth structure, a weeding knife model with semi-circular tooth top structure was constructed, and its structural composition and working principle were systematically analyzed. Combined with the row spacing and root distribution characteristics of rice cultivation, the key structural parameters were designed. By establishing the kinematics model of the weeding knife, the matching relationship between the rotation speed of the cutter shaft and the forward speed was determined, and the collaborative design of the structural parameters and the working performance was realized. In addition, the discrete element software EDEM was used to construct the weeding knife-soil coupling simulation model. The three-factor and three-level orthogonal test was carried out to investigate the influence of test factors on the performance of the weeding knife by taking the soil depth, rotation speed and forward speed as the test factors, the weeding torque and soil disturbance speed as the test indexes, and the power consumption as the goal. The results showed that the rotation speed of the cutter shaft was the main influencing factor, followed by the depth of penetration, and the influence of the forward speed was relatively small. The results of comprehensive multi-index scoring showed that the combination of working parameters with soil depth of 50mm, rotation speed of 180r/min and forward speed of 0.7m/s was the optimal working condition, which can increase the soil disturbance in the working area and reduce the mechanical load. At the same time, the contour of the disturbed section under different propulsion distances was extracted for geometric comparison, which verified the optimization effect of the bionic tooth structure on the soil disturbance behavior and proved that it had better operation performance.

Abstract:Aiming to address the issues of low stone-picking rate and soil accumulation in stone pickers, a chain-and-scoop type stone picker was designed. Through force analysis of the soil-entering scoop plate, its structural parameters were determined. Kinematic analysis of the soil-shifting and stone-picking processes of the stone-picking shaft was conducted to clarify the motion trajectory of the stone-picking teeth and the key structural parameters. The action of the stone-picking shaft on disturbing the soil accumulation area and the velocity of stones thrown onto the conveyor chain were determined by using EDEM software. To evaluate the operational reliability and performance of the stone picker, response surface methodology was employed to design experiments investigating the effects of forward speed, rotational speed of the stone-picking shaft, and ground clearance on the stone-picking rate and soil accumulation height. A regression model was established for parameter optimization. The experimental results indicated that the optimal parameter combination was a forward speed of 0.844m/s, a stone- picking shaft rotational speed of 338.9r/min, and a ground clearance of 394.26 mm. Under these conditions, the model predicted a stone-picking rate of 93.85% and a soil accumulation height of 250.57 mm. Field validation tests showed relative errors between the measured values and the model predictions were 2.97% and 4.73% , respectively, both below 5%. This confirmed that the prototype design was reasonable and met the operational requirements of a stone picker.

Abstract:In view of the problems of high digging resistance, low operational efficiency and severe soil disturbance in the traditional harvesting of buried shellfish in tidal flats, a bionic digging shovel design method was proposed based on the bionic characteristics of crab pereiopods, aiming to improve the operational performance of digging operations. Firstly, the morphological structures and motion characteristics of crab pereiopods were observed and analyzed by means of 3D scanning technology, and key bionic elements such as the sharp curvature of the pereiopod extremities and the segmented surface texture features were extracted. Secondly, the 3D structural modeling of bionic digging shovel was completed through integration of extracted bionic features and physicomechanical properties of tidal flat soil. Furthermore, numerical simulation of digging process was conducted via discrete element method (DEM) to analyze the interaction mechanism and drag reduction performance between bionic digging shovel and tidal flat sediment soil. Finally, a bench test device was established to carry out comparative tests on digging speed, digging resistance and soil disturbance, to investigate the actual operational performance of the bionic digging shovel. Simulation results showed that the bionic digging shovel had an average resistance of 151.42N, with a drag reduction rate of 17.45%. Its average soil looseness was 33.43%, which was 6.91% lower than that of the flat digging shovel. Then, soil trench tests were performed to compare the digging performance of the two shovels, with the average digging resistance measured as 162.41N and 189.27N, respectively. Test results indicated that the bionic digging shovel decreased digging resistance and soil looseness by 14. 19% and 4. 98% compared with that of the flat digging shovel, and its actual drag reduction rate was consistent with the simulation results, meeting the drag reduction requirements for tidal flat shellfish harvesting.

Abstract:Aiming to meet the agronomic requirements for hoeing, root-cutting, and deep fertilization during the winter wheat greening stage, and address the need for integrated inter-row root-cutting and fertilizer application, an integrated hoeing, root-cutting, and topdressing device was designed. With the objectives of reducing draft force, increasing the root-cutting rate, and ensuring fertilization depth, structural design and parameter optimization were systematically conducted. Firstly, based on operation mechanism analysis and mechanical modeling, key components, including the furrow opener, root-cutting shovel, and fertilizer tube were designed. A design scheme was proposed, featuring an exponential curve for the leading edge of the furrow opener and a brachistochrone curve for the fertilizer tube. Secondly, a discrete element method ( DEM) simulation model for the “ integrated device-soil-root-fertilizer” system was established. Using the root-cutting rate and draft force as evaluation indicators, key parameters such as the shovel blade profile, penetration angle, and wing angle were analyzed, and reasonable parameter ranges were determined based on wheat agronomic requirements. Subsequently, a Box Behnken design simulation experiment was employed for parameter optimization. The optimal structural combination for the root-cutting shovel was determined to be an Archimedes spiral blade profile, a penetration angle of 6.43°, and a wing angle of 54.99°. The optimized structural parameters effectively increased the root-cutting rate and reduced draft force. Simulation results for the brachistochrone fertilizer tube showed a maximum fertilization depth qualification rate of 94% with a stable discharging process. Field comparative experiments indicated that at an operating speed of 4km/h, the integrated device performed well, achieving an average root-cutting depth of 117 mm, a root-cutting rate of 82.9% , and a fertilization depth qualification rate of 90.6%. The results confirmed that the designed device successfully fulfilled the functions of hoeing, root-cutting, and deep fertilizer application.

Abstract:Aiming to address the challenges associated with heavy clay soil, high stubble retention, insufficient stability in furrowing depth, and stubble interference with seedling emergence in Southern China, a seedbed preparation device integrating rotary pressing for ridge formation, stubble cutting, furrowing, soil covering, and compaction was developed. Accordingly, a compatible multi-purpose planter was designed to achieve combined sowing of rapeseed, wheat, and corn in wet fields with stubble retention. The overall design scheme of the planter was elaborated, featuring an active rotary pressing seedbed preparation device. Through an analysis of the interaction mechanism between the components and stubble, as well as the optimization of critical component designs, the configuration of the ridge-forming and furrowing device was determined, along with the structural dimensions and kinematic parameters of four types of disc coulters. Discrete element method (DEM) simulations were conducted to model the stubble cutting process under the combined action of an active disc coulter and a runner-type opener. The results indicated that the optimal cutting performance was achieved with an Archimedean spiral disc rotating at 301. 18 r/ min and a runner-type opener width of 31.49 mm. Bench tests were subsequently constructed and executed; the relative error for both the stubble cutting rate and soil disturbance area between the bench tests and simulation results was less than 10% , confirming the reliability of the simulation model. Field performance and sowing trials demonstrated that, under heavy clay soil conditions with stubble retention, the multi-purpose planter achieved a ridge surface flatness of 15.68 mm and a seed furrow stubble cutting rate of 93.7%. The sowing depths for rapeseed, wheat, and corn were 13.2mm, 30.4mm, and 52.2mm, with depth stability coefficients of 90.6%, 86.4%, and 88.2%, respectively. These performance metrics satisfied the agronomic requirements for the cultivation of rapeseed, wheat, and corn. The research result can provide a valuable reference for the design of multi-purpose planters in the rapeseed-wheat-corn planting regions of the middle and lower reaches of the Yangtze River.

Abstract:Aiming at the prominent problems that cold waterlogged paddy fields are featured by excessive soil water content and poor aeration performance, which easily lead to surface water accumulation during seedbed preparation operations, further affecting seed germination and restricting the healthy growth of rice seedlings in the later period, a micro-ridging component dedicated to seedbed preparation in cold waterlogged paddy fields was independently developed. The developed component was mainly composed of a sliding-pushing furrow opener and a compacting and leveling part, which can synchronously realize three key functions including micro-ridging and drainage, seedbed surface leveling, as well as side deep fertilization, effectively solving the core technical bottleneck of seedbed preparation in cold waterlogged areas. Based on the sliding cutting principle, the pushing and soil-breaking surface of the sliding-pushing furrow opener was rationally designed, and through theoretical analysis and parameter optimization, the initial sliding cutting angle θA and the final sliding cutting angle θB of the curved contour were determined to be 10° and 55°, respectively. According to the actual agronomic operation requirements of cold waterlogged paddy fields, the length L0 of the sliding-pushing furrow opener was set as 150mm, and the profiling adjustment angle Δλ of the compacting and leveling part was set to 7° to ensure operational stability. Through DEM-MBD coupled simulation experiments, the optimal width of the sliding-pushing furrow opener was determined as 50 mm, and the stiffness of the profiling spring for the compacting and leveling part was calculated to be 41.42N/mm. Field experiment results showed that the bed surface evenness reached 11.01 mm, the average depth and width of the micro-ridge ditch were 53. 25 mm and 48.11 mm, respectively, and the stability coefficients of ditch depth and width were 89.56% and 90.64% , which all met the agronomic and operational requirements of mechanized ridging in cold waterlogged paddy fields. The research result can provide important technical support and theoretical reference for the design and optimization of seedbed preparation machinery in cold waterlogged paddy fields.

Abstract:Aiming to address the issues of unstable fertilization depth and indistinct stratification effects in existing maize layered fertilization systems, an inter-fertilizer soil-covering layered fertilization device was developed. Based on theoretical analysis, the fundamental structural parameters of the device were determined, and the primary factors affecting the soil-covering performance were identified as the installation angle, inclination angle, and width of the soil-covering plate. The working process of the device was analyzed by using the discrete element method (DEM). Taking the average depth of shallow- layer fertilizer and the resistance of the soil-covering plate as evaluation indices, a quadratic orthogonal rotational combination simulation experiment was conducted. A regression model was established to optimize the structural parameters. The optimal parameter combination was obtained as follows: inclination angle of the soil-covering plate was 29.93°, installation angle was 11.40°, and plate width was 31.40mm. Soil bin experiments were further carried out at different forward speeds and fertilizer application rates to evaluate the performance of the proposed device. The results showed that the fertilization depth and its coefficient of variation under various operating conditions satisfied agronomic requirements. Compared with a conventional layered fertilization device, the coefficients of variation of shallow-layer and deep-layer fertilization depths were reduced by 2.64 and 5.98 percentage points, respectively, indicating an improved stratification effect. Field experiments were conducted to further validate the practical performance of the device. The results demonstrated that the coefficients of variation of both shallow-layer and deep-layer fertilization depths were less than 10% , meeting the requirements of relevant industry standards. In summary, the proposed inter-fertilizer soil-covering layered fertilization device effectively enhanced the stability of fertilization depth and improved the layered fertilization performance, providing a reliable technical approach for precision fertilization in maize production.

Abstract:Aiming to address the challenges of high labor intensity, low efficiency, and poor quality in manual ridge construction caused by unsuitable soil moisture content, a bilateral paddy field ridger building machine equipped with a variable spray system was designed. It was elaborated on the hardware composition and control strategy of the variable spray system, which enabled manual and automatic flow adjustment to precisely regulate soil moisture content within the optimal range of 26%~28% . Using EDEM simulation software, the soil-engaging components rotary tillage and soil collection processes were simulated. An orthogonal experiment was conducted with forward speed, working speed, and soil-taking depth as test factors, and soil collection rate and coefficient of variation of soil collection uniformity as evaluation indicators. A regression model was established to optimize the best working parameter combination, laying the foundation for subsequent field experiments. The results showed that under the optimal ridge-building moisture content, the ridger operated at the best parameter combination: forward speed of 0.8km/h, working speed of 415r/min, and soil-taking depth of 150mm. The average firmness of the constructed ridges was increased with time, while the coefficient of variation of firmness was gradually decreased, and stabilized after 24h of air-drying. The average firmness at various measurement points on the ridge top was no less than 1 538 kPa, and on the ridge side, no less than 2612kPa, both meeting agronomic requirements. The research result can contribute to advancing high-standard farmland construction and accelerating the full mechanization of rice cultivation.

Abstract:Aiming at the problems of degradation of saline-alkali grasslands in the Songnen Plain, severe soil compaction, and hindered asexual tillering of dominant populations such as Leymus chinensis, an integrated machine for active root cutting, horizontal vibration soil loosening, and fertilization was designed. Its core components included an anti-winding root cutting disc (with a dynamic sliding cutting angle) designed based on the Archimedean spiral principle and a spring-buffered horizontal vibrating subsoiling mechanism. Through sliding cutting motion and high-frequency lateral shear force, these components achieved low-disturbance root cutting and in-situ shattering of the deep hardpan. The Box-Behnken experimental design was employed to optimize the operating parameters, indicating that the optimal parameters were as follows: a forward speed of 0.991m/s, a root cutting disc speed of 174r/min, and a vibrating motor speed of 1769r/min. Field verification with rounded parameters demonstrated that the soil compaction change rate reached 40.79% , successfully achieving “ zero-winding” cutting of lateral roots. Quantitative analysis of multi-dimensional physical indicators and surface disturbance revealed that the machine possessed extremely strong adaptability to soil heterogeneity; the average soil disturbance cross-sectional area was only 22.0cm2, the vegetation damage rate was controlled at approximately 23% , and the relative decrease in soil moisture content before and after the operation was merely 0.37%. The research can provide a theoretical basis and equipment support for the conservation tillage and restoration of degraded soda saline-alkali grasslands.

Abstract:Aiming at the problems of slow response speed and low precision of the decision-making model in the existing rapeseed direct seeder control system, which lead to decreased operation efficiency and large seeding quantity error, a control sequence decision-making and regulation system for rotational speed-working length bi-variable seeding was designed based on the air-assisted bi-variable seeding device for rapeseed. By analyzing the algorithm mechanisms of the multi-objective Ivy algorithm (MOIVY) and Lévy flight algorithm (LFA), a multi-objective fitness function involving seeding error, controller response time, and coefficient of variation for inter-row seeding quantity consistency was established, and an LFA-MOIVY control sequence decision-making model was constructed. Comparative simulation experiments were conducted to verify the superiority of the MOIVY, DE-MOIVY, and LFA-MOIVY algorithms. The results showed that when the target seeding rates were 1. 5 kg / hm2, 3. 0 kg / hm2, 4. 5 kg / hm2, and 6. 0 kg / hm2, the average Hypervolume (HV) values of the LFA MOIVY algorithm were 2. 35, 2. 63, 2. 02, and 1. 98, respectively. Its Pareto solution set exhibited comprehensive coverage and uniform distribution, and its overall performance was superior to that of the other two algorithms. Bench comparison tests indicated that compared with the traditional decision-making method, the LFA MOIVY seeding decision-making method reduced the relative seeding error by 0. 76 percentage points, the coefficient of variation by 2. 43 percentage points, and shortened the response time by 0. 14 s. Pavement tests showed that under the dual changes of different operation speeds and seeding rates of the rapeseed direct seeder, the relative seeding error was no more than 4. 5% , the coefficient of variation was no more than 6. 2% , the response time ranged from 1. 59 s to 2. 61 s, the rotational speed control accuracy was not less than 97. 6% , and the working length control accuracy was not less than 98. 6% , all of which met the requirements of national and relevant industry standards. Field tests demonstrated that when the seeding rate was 4. 5 kg / hm2 and the operation speed was 3 ~ 12 km/ h, the rotational speed adjustment range was 18. 8 ~ 47. 0 r/ min, the working length adjustment range was 19. 0 ~ 30. 0 mm, the seeding accuracy was increased by 3. 41 percentage points, the coefficient of variation was decreased by 2. 24 percentage points, and the response time ranged from 2. 10 s to 2. 86 s. The research results can provide a reference for the precision seeding regulation technology of small-seeded crops.

Abstract:Aiming to improve the efficiency and quality of clamp-type rice pot seedling transplanting, a method was introduced. This method combined a multi-arm rotary system for high-speed seedling picking with a pair of counter-rollers for accelerated seedling projection. A second-order oval gear, known for its stable transmission, was used in the planetary gear mechanism. This mechanism featured four symmetrically arranged clamping arms to increase the number of seedlings picked per cycle. A kinematic model and a computer-aided design program were developed to optimize the clamping trajectory and posture. This process led to the identification of an optimal “peach-shaped” trajectory and the completion of the gear system design, which was verified by simulation. To ensure stable performance at high speeds, a spatial swing cam mechanism was designed based on a fifth-order polynomial motion law. This design achieved low pressure angles and ensured smooth movement. Furthermore, single-factor tests were conducted to analyze how roller diameter, speed, and position affect seedling projection. The results enabled precise acceleration and directional projection of seedlings, which was crucial for achieving upright posture and consistent planting depth. An orthogonal experiment identified soil moisture content as the most significant factor affecting the planting success rate. Single-factor tests showed that at a speed of 110 r/ min, the seedling picking success rate averaged 94. 46% . With a projection speed of 5 m/ s, seedlings achieved an ideal planting depth of 2. 5 ~ 3. 0 cm. Combined tests showed that the overall planting success rate was 91. 6% , with an operational efficiency of 435 plants per minute and a damage rate was 2. 1% . In summary, this research adopted a separated seedling-retrieving and seedling-planting collaborative method, offering a technical solution for high-speed, high-quality mechanized rice transplanting.

Abstract:Aiming to improve the seedling planting and establishment performance of the integrated seedling taking and planting pot seedling transplanting mechanism, a secondary extension pot-protecting transplanting mechanism was proposed. Through the functional decoupling design of the integrated seedling taking and planting arm, and with the cooperative operation of the two progressive extensions of the seedling pushing rod and the clamping jaws, the mechanism sequentially achieved pot seedling clamping, pot protection, and planting. The mechanism adopted planar cam and end cam as the core driving units: the planar cam drove the seedling pushing rod to perform two progressive extensions, sequentially completing seedling protection and planting actions; the end cam controlled the opening and closing of the clamping jaws to realize the clamping and releasing of pot seedlings. To achieve the design goals of low impact and low wear for the cam mechanism, combined with the technical requirements of transplanting movement and the structural parameters of the transplanting arm, a kinematic model of the cam mechanism was established, and the cam profile design was completed based on the curve envelope theory. For the planar cam mechanism, by constructing an elastodynamic equation and formulating a contact mechanics optimization strategy, the contact form between the cam and the fork was optimized into surface contact, which effectively reduced kinematic impact. For the end cam mechanism, a profile optimization function based on curvature constraint was introduced to optimize its peak pressure angle characteristics. The design scheme was double-verified through virtual prototype kinematic simulation and physical prototype field tests. The results showed that the transplanting arm can significantly reduce the sensitivity of pot seedling planting uprightness to the rotational speed of the planetary gear train. At a working speed of 250 plants/ (min · row), the qualified rate of uprightness reached 87. 50% , which was significantly higher than that of the original mechanism, making it suitable for high-speed and high- quality pot seedling transplanting scenarios.

Abstract:Fertilizer spreading is a crucial part of agricultural production. Improving spreading uniformity can ensure crop yield and quality while reducing environmental pollution. To address the poor uniformity of fertilizer distribution when spread by the single-channel linear type spreading blades, a single-disc centrifugal variable rate fertilizer spreader based on double-channel inclined spreading was designed. A double-variable fertilizer spreading control system was developed by using the fuzzy PID algorithm, which can adjust the fertilizer application rate and spread width simultaneously according to real time information. A complete kinematic model was constructed by analyzing the fertilizer particles at the front and rear of the blades and in the air, and optimum blade parameters were determined to achieve the most uniform fertilizer distribution. A fertilizer spreading simulation model was established based on the discrete element method, and single-factor tests, orthogonal tests, and response surface analysis were carried out. The test results showed that when the fertilizer outlet opening was 8°, the blade deflection angle was 0. 42°, the blade inclination angle was 16. 22°, the disc rotation speed was 450 r/ min, and the forward speed was 1. 1 m/ s, the fertilizer distribution reached the best state. At this time, the coefficient of variation of fertilizer distribution was 18. 7% , and the effective spreading width was 10. 96 m. Bench tests and field tests were conducted to verify the simulation test results and the performance of the double- variable fertilizer spreading control system. The results showed that the average error in the coefficient of variation for fertiliser distribution uniformity between the bench test and the simulation test was 0. 24% , indicating that the discrete element method can accurately simulate the fertilizer spreading process. The coefficient of variation of fertilizer distribution uniformity was 24. 5% , the fertilizer application control accuracy was 90. 7% , the total fertilizer discharge stability was 94. 6% , and the error of fertilizer application per unit area was 19. 8% , meeting the agronomic requirements. The research results can provide an important reference for improving fertilizer spreading uniformity and studying precise variable rate fertilization technology.

Abstract:Aiming to achieve high-precision monitoring of cleaning loss during grain harvesting, it is essential not only to develop cleaning loss monitoring sensors with low error and high stability but also to establish an accurate mathematical mapping between the sensor readings and the actual cleaning loss. Focusing on a cleaning loss monitoring sensor previously developed and systematically investigated methods to improve monitoring accuracy, three-dimensional particle models representing components of the threshing mixture, such as grains, short straw, and impurities, were constructed for typical crops, including rice, wheat, and rapeseed, based on their physical and compositional characteristics. Multiple three-dimensional flow channel models of cleaning devices were then established according to actual operating conditions. Using a coupled CFD-DEM simulation approach, numerical simulations of gas- solid two-phase flow were conducted under both single-factor and multi-factor conditions, including crop type, feeding rate, sieve opening, and fan inlet area. The relationship and variation patterns between sensor readings and actual cleaning loss were analyzed in detail under different operating scenarios. Field catching tests were subsequently performed to validate the simulation results. The findings indicated that crop type, sieve opening, and fan inlet area were the primary factors influencing the proportion of cleaning loss detected by the sensor. In contrast, feeding rate had an insignificant effect within a certain range. The monitoring proportion values of cleaning loss obtained from the simulation were larger than that of the actual field test values, with a standard deviation of 0. 97% in the extent of overestimation, which verified the effectiveness of the simulation analysis. Subsequently, several representative nonlinear fitting models were established based on the simulated sample data. Through comparative analysis of the fitting results and considering practical engineering applicability, a quadratic polynomial regression model was selected as the compensation model for cleaning loss monitoring ratio, achieving an adjusted R2 greater than 0. 991 and a root mean square error lower than 0. 216% . Finally, the model was integrated with an embedded system to realize online compensation of the cleaning loss monitoring ratio, with the deviation between the compensated values and the measured values not exceeding 1. 08% . This research significantly improved the accuracy and adaptability of cleaning loss monitoring, reduced reliance on manual catching box tests, and provided robust technical support for intelligent operational parameter control and comprehensive performance evaluation of combine harvesters.

Abstract:Aiming to address the issues of low separation efficiency of soil blocks and high loss rate of fallen fruits during peanut harvesting, a shovel-chain type peanut harvester was designed. Through analysis of the force conditions on peanut plants and soil particles, the motion laws of the peanut plant-soil composite body during the process of transportation without soil, flipping and laying were explored. The structural parameters of the transportation separation device and the flipping and laying device were optimized. Based on the DEM-MBD coupling method, a simulation model of the peanut plant-soil-loop transportation chain was established, and simulation experiments on the peanut plant transportation without soil were carried out to explore the influence laws of the loop transportation chain speed, loop transportation chain inclination angle, and vibration wheel speed on the force conditions of the peanut plants. Field experiments for peanut digging and harvesting were conducted, with the inclusion rate of soil and loss rate of fallen fruits as test indicators. The optimal parameter combination was solved. The field experiments showed that when the loop transportation chain speed was 1. 5 m/ s, the loop transportation chain inclination angle was 25°, and the vibration wheel speed was 150 r/ min, the loss rate of fallen peanuts was 2. 9% , the inclusion rate of peanuts was 6. 4% , and all performance indicators met the requirements of relevant standards.

Abstract:Canola is one of the most important oilseed crops worldwide, and accurate prediction of its yield and total production is of great significance for food and edible oil security assessment, agricultural production management, and bioenergy potential analysis. To address the limited sample size in canola yield and production prediction, as well as the lack of systematic evaluation of production modeling and multi-source information synergy, county-level canola yield and total production data were collected, and multi-source remote sensing and meteorological big data were integrated to develop a collaborative prediction framework for canola yield and total production. On this basis, six tree-based machine learning models were employed to systematically evaluate the performance differences among models and data- source combinations in predicting canola yield and production. The results demonstrated that the joint use of remote sensing and meteorological data consistently outperformed single-source inputs for both yield and production prediction, with this advantage showing strong consistency and robustness across different tree- based model structures. For yield prediction, CatBoost achieved the best performance ( R2 = 0. 679, RMSE = 0. 269 t/ hm2, MAE = 0. 194 t/ hm2), whereas for production prediction, Cubist performed best (R2 = 0. 952, RMSE = 9. 42 × 103 t, MAE = 5. 00 × 103 t). The spatial patterns of the predicted yield and production were generally consistent with the observations. Overall, the results indicated that the integration of multi-source remote sensing and meteorological information with tree-based machine learning methods can effectively improve the accuracy of regional-scale canola yield and production prediction. The research result can provide reliable technical support for canola yield monitoring, precision agricultural management, and harvest scheduling decisions for agricultural machinery.

Abstract:Timely and accurate acquisition of the distribution of maize planting is of great significance to agricultural production management. Aiming at the problems of data loss and sample dependence in maize identification in cloudy and rainy areas, an unsupervised synergistic maize mapping index (SMMI) was proposed to achieve maize extraction in Shouyang County, Shanxi Province. The GF-6 WFV and Sentinel-1 SAR images of the key growth period of spring maize from May to August 2021 were selected as the data sources. By coupling the two red-edge bands of GF-6, SAVI, TVI, RDVI and the VH features of Sentinel-1, SMMI was constructed for threshold classification to extract the distribution of spring maize. The results showed that the overall accuracy of extracting spring maize by using SMMI reached 90. 72% , and the user accuracy and producer accuracy of maize identification reached 88. 07% and 87. 8% , respectively. Compared with the index without using radar features and the multi-feature set-based random forest (RF) algorithm, the overall accuracy was increased by 1. 86, 0. 95 percentage points, respectively. It was indicated that SMMI was significantly superior to the two methods in the recognition accuracy of spring maize. It revealed the importance of the red-edge band and VH polarization characteristics in the identification of spring maize. Combining optical and radar features simultaneously can achieve a better classification effect than a single data source. This research can provide a reference method for mapping the distribution of maize in multi-cloud and rainy areas and had certain potential for agricultural application.

Abstract:Real-time monitoring of nitrogen status is essential for improving fertilizer use efficiency and crop yield and quality. Nitrogen monitoring plays a key role in precision agriculture; however, traditional soil and leaf sampling methods cannot meet the requirements for large-scale, real-time monitoring. Multi- source data from RGB and multi-spectral sensors were integrated to develop a model for estimating leaf nitrogen content ( LNC) in winter wheat, with the aim of improving the accuracy and reliability of nitrogen monitoring. Multi-source features were first extracted from UAV-based RGB and multi-spectral imagery acquired at key growth stages. Key variables were then identified by using a two-stage feature selection strategy that combined Pearson correlation analysis and variable importance in projection (VIP). Color space, texture, and spectral features were subsequently integrated to construct LNC estimation models. The results showed that the fused models consistently outperformed single-sensor models across different growth stages. Among the tested methods, the support vector regression (SVR) model achieved the best performance, with R2 increased by 0. 03 ~ 0. 16, RMSE decreased by 0. 02 ~ 0. 10 percentage points, and NRMSE reduced by 0. 61 ~ 3. 90 percentage points. These findings demonstrated that multi-source remote sensing data fusion can effectively improve LNC estimation accuracy and model robustness. The research result showed that multi-source remote sensing information fusion can effectively improve the accuracy of LNC estimation, overcome the limitations of a single sensor, and provide technical approaches and theoretical support for precision agriculture.

Abstract:China is the largest rice-growing country in the world, and accurate recognition of different growth stages of rice is essential for achieving intelligent rice field management. Targeting the classification of rice growth stages, including initial heading stage, full heading stage, milk ripening stage, and yellow ripening stage, a recognition model for rice growth stages was proposed based on multi- feature fusion. The specific methodology included multi-feature extraction module, utilizing high- resolution remote sensing images obtained from drones as the data source, multi-scale features and deep semantic features of the images were extracted by using the Swin Transformer model and ResNet18 model. An improved feature fusion module was designed, which included an adaptive feature fusion ( AFF) module and a global context modeling ( GC) module. The AFF module achieved adaptive fusion of different features by learning the weight relationships between features. The GC module incorporated global context information to enhance the model’s ability to perceive global features, thereby improving adaptability in complex scenarios. Model optimization, utilizing the Lion optimizer and a dynamic learning rate strategy to accelerate the model??s convergence speed. Additionally, a Dropout layer was introduced to prevent overfitting and improve the model??s generalization ability. Comparative experiments demonstrated that the proposed model achieved a 97. 14% accuracy rate in recognizing rice at the heading and maturity stages, which was 3. 29, 2. 53, and 1. 09 percentage points higher than that of other mainstream models such as MobileNetV2, EfficientNet and Swin Transformer, respectively. Notably, while maintaining high accuracy, the model had relatively fewer parameters and lower computational costs, showcasing lightweight characteristics suitable for practical applications.

Abstract:Rapeseed is China’ s dominant oil crop and the mean seed number per silique is the key predictor of yield; however, manual dissection is destructive, subjective and labor-intensive. A pose- aware, non-destructive counting pipeline that integrated digital imaging with a customized YOLO detector named RAPESEED YOLO was proposed. The network employed a lightweight VanillaNet backbone, a P2 micro-object head and multi-scale dilated attention (MSDA) to enhance tiny-seed features, while a Dynamic Head with DySample up-sampler and a C3k2 _ MDConv2 fusion module refined multi-scale context. Training was regularized by Shape-IoU loss to improve bounding-box regression for ellipsoidal seeds. Evaluated on 15 field plots ( three cultivars, two densities), RAPESEED-YOLO achieved 98. 6% precision, 98. 7% recall and 98. 9% mAP; the overall pipeline delivered 95. 6% counting accuracy with less than 10% relative error across 6 800 siliques. Operating on a 15 W edge device, the system processed three plants per hour, demonstrating robustness to uneven illumination, occlusion and cultivar variation. Pose-specific multiplication factors ( upright 1 × , semi-lateral 2 × strong-side, full- lateral 2 × single-side) convert detected seeds into total silique counts without physical contact. With only 2. 3 × 106 parameters and 5. 1 × 109 FLOPs, the method offered a lightweight, accurate and deployable solution for high-throughput rapeseed yield estimation.

Abstract:Aiming to address the issue of high damage rates caused by a single harvesting mode for apples of varying maturities, an apple maturity identification model, YOLO 11n-LDA, was proposed based on YOLO 11n, combined with a flexible gripper to achieve low-damage harvesting for apples at different maturity stages. Firstly, the large separable kernel attention ( LSKA) module was integrated into the SPPF module of the YOLO 11n network to expand the model??s receptive field. Subsequently, dynamic snake convolution (DySnakeConv) was added to the C3k2 module to enhance the model??s perception of multi-scale features. Furthermore, the asymptotic feature pyramid network (AFPN) replaced the neck network to bridge the semantic gap and improve feature fusion stability. Finally, the maximum safe picking force for apples of different maturities was determined via ANSYS simulation, and a comprehensive harvesting test platform, comprising a collaborative robotic arm, a flexible fin gripper, a camera, and an industrial control computer was constructed for experimental testing. Experimental results on the PApple_RGB-D-Size dataset showed that, with maturity classified into eating-ripe, harvest-ripe, and unripe, the improved model achieved an overall mean average precision ( mAP @ 0. 5) of 87. 7% . The AP @ 0. 5 for eating-ripe, harvest-ripe, and unripe apples were 86. 6% , 89. 0% , and 87. 5% , respectively; the overall mAP@ 0. 5 and the AP@ 0. 5 for the three maturity categories was increased by 3. 9, 8. 0, 1. 6, and 2. 1 percentage points compared with that of YOLO 11n. In the orchard apple maturity dataset, the improved model achieved an mAP@ 0. 5 of 95. 5% , representing an increase of 0. 9 percentage points compared with that of YOLO 11n. In harvesting experiments, the detection rate, harvesting success rate, and damage rate were 99. 0% , 84. 0% , and 4. 8% , respectively. The enhancement of the vision model improved detection accuracy, while the visual-tactile fusion method provided a reference scheme for low-damage harvesting.

Abstract:Aiming to address the problem of low classification accuracy in strawberry ripeness recognition caused by branch and leaf occlusion and uneven lighting under natural conditions, a cascade detection method that integrated an improved YOLO v8 instance segmentation with a lightweight CNN classification network was proposed. This method firstly used the GhostNet lightweight structure and an attention mechanism to improve the YOLO v8n-seg model for fruit localization and segmentation, achieving accurate target segmentation in complex backgrounds with a relatively low model parameter count (2. 6 × 106). Nextly, fruit target areas were cropped from the original images by using the segmentation masks and input into a CNN classification network built on EfficientNet-Lite. This network, with mobile inverted bottleneck convolution (MBConv) modules and a Softmax classifier, focusing on capturing subtle color and texture feature changes on the fruit surface, effectively improved the model's ability to recognize fruits in transitional ripening stages. Experimental results showed that the proposed decoupled spatial localization and attribute recognition coordination strategy significantly enhanced system robustness. The average ripeness recognition accuracy reached 88. 1% , the overall system parameter count was 7. 3 × 106 (EfficientNet-Lite parameters accounted for 4. 7 × 106 ), and the inference speed was 71. 9 f/ s, achieving accurate grading of unripe, turning, and ripe strawberries while ensuring real-time processing. The research result can provide technical support for the vision system of intelligent strawberry-picking robots.

Abstract:Cow tracking is an essential technique for obtaining positional and behavioral data of cows, playing a vital role in intelligent farm management. To tackle challenges such as missed and false detections, as well as track fragmentation caused by small object sizes, occlusion, and overlap in farm videos, the YOLOX-BT algorithm was proposed. This method enhanced both detection and tracking performance. Specifically, the YOLOX-s model was improved by replacing the original CSP2 _ 1 structure with the CFP_EVCBlock, boosting small object detection capability. Within CFP_EVCBlock, a feature refinement module ( FRM ) was designed by integrating multiple feature enhancement mechanisms, enabling prioritized extraction of important features and effectively mitigating occlusion and overlap issues. In the tracking phase, appearance similarity was introduced into the data association process, and the weights between appearance and IoU matching were dynamically adjusted, improving cow identity matching accuracy. Experimental results showed that the improved detector achieves an average precision of 94. 9% , precision of 94. 2% , recall of 92. 7% , and an inference speed of 22. 9 f/ s, representing gains of 1. 2, 0. 8, 2. 3 percentage points, and 29. 4% , respectively, over the baseline. Additionally, the number of parameters and FLOPs were reduced by 9. 3% and 16. 7% , respectively. In the cow multi-object tracking task, YOLOX-BT achieved a MOTA of 93. 2% , IDF1 of 95. 2% , 13 ID switches, and a MOTP of 5. 3% . Compared with commonly used methods, the proposed algorithm showed significant improvements in both tracking accuracy and identity consistency, providing robust technical support for intelligent cow management.

Abstract:Water quality is a core element related to human health and food security. To address the issue of agricultural water pollution caused by agricultural practices,it is urgent to establish a systematic evaluation of agricultural water quality and reveal its spatiotemporal differentiation characteristics and driving mechanisms. Using the wild horse optimizer-random forest (WHO-RF) model to evaluate the surface water and groundwater quality of 15 farms under the jurisdiction of Jiansanjiang Branch of Heilongjiang Beidahuang Agricultural Reclamation Group Co. ,Ltd. , their spatiotemporal evolution laws were analyzed,and the causes of water quality changes were explored. The results indicated that the water quality of Jiansanjiang Branch met the requirements for agricultural water use. From 2018 to 2020,the surface water quality firstly decreased and then increased,while the groundwater quality firstly increased and then decreased. In 2020,the difference in water quality between the two was the largest. From 2021 to 2022, the surface water and groundwater quality both increased, and the agricultural water quality showed a significant improvement trend. Compared with the central farms, the surface water and groundwater quality of farms near the riverbank underwent significant changes while the spatial heterogeneity of regional water quality was weakened. The OOB index was used to determine the content of TP and NH3-N in surface water,as well as the content of Cl - ,CODMn,and NH3-N in groundwater, which were five key indicators affecting regional agricultural water quality. The average nitrogen fertilizer application rate had the most significant impact on regional agricultural water quality. To test the comprehensive performance of the WHO-RF model,traditional random forest (RF) model and dragonfly algorith-random forest (DA-RF) model were selected for comparative analysis. By comparing four evaluation indicators based on mean absolute percentage error, coefficient of determination, root mean square error,and mean square error,the superiority and reliability of the model in regional water quality evaluation were verified. The results can provide an approach for water quality evaluation, and also provide guidance for regional water quality risk response.

Abstract:Biochar is regarded as a promising material for improving soil hydraulic properties due to its porous structure and abundant surface functional groups. Cotton stalk, characterized by its rigid texture and high lignin content, is difficult to degrade effectively through conventional biochemical pathways. Converting it into biochar for soil amendment represents a viable strategy for synergizing agricultural waste recycling with ecological restoration. Biochar was produced from cotton stalks at different pyrolysis temperatures ( 300℃, 400℃, 500℃, 600℃and 700℃). The structural properties of biochar, including pore architecture, specific surface area, functional group evolution, and crystalline structure, were systematically characterized. Laboratory-based one-dimensional soil column experiments were conducted to quantitatively evaluate the effects of biochar produced at different temperatures and application rates (1. 0% , 1. 5% , and 2. 0% ) on soil wetting front migration, water infiltration, and evaporation processes. The Philip and Kostiakov models were applied to describe the relationship between infiltration rate and time, thereby simulating water infiltration behavior. Similarly, the Rose and Gardner models were used to numerically simulate soil water evaporation patterns. The results indicated that biochar produced at 500℃exhibited the highest specific surface area (1. 67 m2 / g) and total pore volume (0. 003 2 cm3 / g ), with well-developed micro- and mesoporous structures and a high degree of aromatization. Soil column experiments demonstrated that biochar application significantly delayed wetting front movement, reduced infiltration rate, suppressed evaporation, and enhanced water retention capacity, primarily through capillary blocking and pore structure reorganization. Among all treatments, the application of 2. 0% biochar produced at 500℃yielded the most favorable outcomes, exhibiting a 78. 6% prolongation of infiltration time, a 12. 9% increase in cumulative infiltration, and a 34. 0% reduction in evaporation compared with that of the control group. Model fitting showed that the Kostiakov model described the infiltration process with high accuracy (R2 > 0. 99), outperforming the Philip model, while the Rose model simulated evaporation more accurately (R2 > 0. 99) than the Gardner model. In conclusion, cotton stalk biochar produced at appropriate pyrolysis temperatures and application rates can significantly enhance soil water movement and retention. However, practical application for improving soil water retention should consider the native physicochemical properties of soil to select biochar produced under suitable pyrolysis conditions.

Abstract:A detection method that combined hyperspectral imaging and multiple optimization strategies was proposed to effectively detect early decay of oranges, addressing the issue of low detection accuracy in traditional methods. Traditional detection approaches, such as visual inspection and manual sorting, were highly subjective and fail to identify subtle changes in early decayed oranges, leading to substantial post-harvest losses in the citrus industry. To solve this problem, the hyperspectral data in the 450 ~ 1 050 nm wavelength range were collected by using a professional hyperspectral imaging system, covering the visible and near-infrared regions closely related to fruit internal quality. Then, enhanced data were synthesized by the Borderline-SMOTE algorithm with Kullback-Leibler (KL) divergence of 0. 02 and Wasserstein distance of 3. 4, which effectively alleviated the sample imbalance problem between healthy and early decayed oranges. Subsequently, totally 24 key characteristic wavelengths were screened out by the ReliefF algorithm to eliminate redundant information and reduce computational complexity. Machine learning based classification models and CNN model were constructed in combination with Bayesian optimization, which optimized key hyper parameters to improve model performance. A systematic evaluation was carried out on their classification performance and computational efficiency. After Bayesian optimization and feature selection, the classification error of the CNN model was reduced to 0. 008. The running time was significantly decreased from 910. 4 s to 177. 9 s, representing a reduction of 80. 5% . The accuracy rate on the test set reached 99. 0% . The research result can not only provide a reliable technical solution for early decay detection of oranges, but also lay a theoretical foundation for the development of rapid and intelligent detection equipment, which was of great significance for promoting the high-quality development of the citrus industry.

Abstract:The lack of online detection methods for spatial moisture content distribution during the curing process in bulk curing barns leads to delayed process control responses and frequent issues such as greenish and spotted tobacco leaves. A novel online detection method was proposed based on a sparse weighing network and machine learning. This method established a nonlinear mapping from local moisture content measurements to spatial distribution, using dynamic features including tobacco weight, instantaneous dehydration rate, and ambient temperature and humidity from the front section of the barn, combined with the spatial coordinates of target points. On this basis, Bayesian optimization (BO) was employed to globally optimize the hyperparameters of the constructed XGBoost model, resulting in a predictive model adapted to the stage-varying temperature and humidity characteristics. An online detection system integrated with the weighing sensor network was developed and validated. Results demonstrated that the proposed prediction model delivered excellent prediction performance on the independent test set, with a coefficient of determination (R2) of 0. 996, a mean absolute error (MAE) of 0. 76% and a root mean square error (RMSE) of 1. 14% for the original wet basis moisture content of tobacco leaves, significantly outperforming the conventional SVR and MLP models. Online validation tests showed that the average R2 between the predicted and measured original moisture content was 0. 978, with mean absolute error of 2. 77% and average RMSE of 3. 66% , demonstrating the system??s high reliability. The coefficients of variation for spatial moisture content distribution during the yellowing (38℃), color-fixing (48℃), and stem-drying (65℃) stages were 1. 73% , 8. 27% , and 20. 02% , respectively, clearly indicating a gradual decline in moisture uniformity as curing progressed and confirming the system??s capability to effectively quantify the spatial distribution uniformity. The findings can provide technical support for the precise control of tobacco curing process parameters.

Abstract:Aiming to address the conflict between hydrodynamic performance and functional scalability in small remotely operated vehicles (ROVs) operating in shallow waters, an optimized hybrid configuration design integrating streamlined main body and open-frame structure was proposed. The propulsion support plate was designed with Myring profiles, with emphasis on chamfering the leading edge to reduce hydrodynamic drag. Motion stability was analyzed through a six-degree-of-freedom (6-DOF) dynamic model, while computational fluid dynamics ( CFD) simulations and towing tests in a water tank were conducted to measure drag coefficients and added mass. Systematic evaluation of drag characteristics and flow field structure before and after optimization was performed. Results demonstrated that the optimized design significantly improved flow field performance, reducing flow separation area by 51. 0% and decreasing peak turbulent kinetic energy by 62. 7% . At speed of 0. 01 m/ s, the overall drag was decreased by 4. 99% . Straight-line and oblique navigation simulations further confirmed the optimized ROV exhibited superior hydrodynamic coefficients. Finally, pool tests were carried out to verifiy its motion stability and maneuverability, which confrimed the reliability of the simulation results. The research result can provide an effective theoretical framework and practical reference for designing small ROVs with low drag and high maneuverability.