颗粒肥料质量流量传感器设计与试验
作者:
作者单位:

作者简介:

通讯作者:

中图分类号:

基金项目:

国家自然科学基金项目(31901408)、国家重点研发计划项目(2017YFD0700904)和吉林省科技发展计划项目(20180414074GH)


Design and Experiment of Mass Flow Sensor for Granular Fertilizer
Author:
Affiliation:

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    变量施肥具有提高肥料利用率、保护生态环境、节约农业生产成本等优点,但目前还没有得到广泛的应用,除了难以获得变量施肥的处方图之外,缺乏闭环检测也是原因之一。闭环控制是实现变量施肥的关键之一,与间接测量排肥轴的转速相比,实时检测肥料的质量流量更为准确。本文基于静电感应原理,设计了一种颗粒肥料质量流量传感器。由于颗粒肥料之间、颗粒肥料与空气、颗粒肥料与排肥管之间的摩擦和碰撞,颗粒肥料会携带一定量的电荷,因此本研究设计了环形电极来检测电荷强度,并利用电流放大电路输出感应电流。通过标定质量流量与感应电流的关系,获得了实时的肥料质量流量。搭建试验台对该颗粒肥料质量流量传感器进行检测,试验台主要包括动态信号采集系统、肥料箱、电流放大器和环形电极传感器。以大颗粒尿素(CO(NH2)2)、过磷酸钙(Ca(H2PO4)2·H2O)和氯化钾(KCl)为研究对象,其平均容重分别为0.7、1.2、1.1g/cm3。根据施肥装置的物理参数,通过调整排肥轴转速可获得近似的目标质量流量,目标质量流量的范围是3~15g/s,增量为1g/s。对于每个质量流量,进行了4次重复。每次重复30s,施肥装置与信号采集系统同时启动。利用平均感应电流和平均质量流量建立回归方程,采用插值法得到实时质量流量。随后,对每种肥料进行25次试验,从而检验本文中颗粒肥料质量流量传感器的测量精度,每次试验的目标质量流量由5个随机质量流量组成,每个质量流量下持续排肥6s,用天平称量30s内的实际质量,通过积分质量流量和时间曲线计算检测质量。采用SPSS 22.0软件对试验结果进行统计分析,分析表明,大颗粒尿素、过磷酸钙、氯化钾的检测误差分别为3.9%、5.1%、5.9%,相应的标准差分别为5.21、7.98、11.29。检测质量与实际质量无显著性差异(P>0.1),大颗粒尿素、过磷酸钙和氯化钾检测误差的数学期望值分别为3.74%、4.93%、5.22%。本文的研究结果表明,检测误差随颗粒肥料粒径的减小而增大。

    Abstract:

    Variable rate fertilization has the advantages of improving fertilizer-utilization efficiency, protecting ecological environment as well as saving agricultural production cost. But it has not been widely applied yet, besides it is hard for getting the prescription figure, lacking closed-loop detection is another major reason. Closed-loop control is one critical step towards realizing the variable rate fertilization, compared with the indirect measurement which monitors the axis speed, it is more accurate by monitoring the real-time mass flow rate. If there existed the fertilizer caking that blocked the fertilizer apparatus, it is useless for monitoring the axis speed. Based on the electrostatic induction theory, a sensor that could monitor the mass flow rate of granular fertilizer was designed. Owing to the frictions and collisions between the granular fertilizer and the air, the granular fertilizer and the fertilizer tube, as well as the frictions and collisions among the granular fertilizer themselves, therefore, the granular fertilizers would carry a certain amount of electric charges. One ring electrode was designed to detect the strength of the electric charges, subsequently, a corresponding current amplifying circuit was utilized to export the induced current. The real-time mass flow rate could be obtained by calibrating the relationship between it and the induced current. One test-bed was established in order to finish the task, the test-bed mainly included one dynamic signal acquisition system, one fertilizer box, one current amplifier and the sensor. Large granular urea (CO(NH2)2), superphosphate (Ca(H2PO4)2·H2O) as well as potassium chloride (KCl) were chosen as the research objects, their mean bulk densities were 0.7g/cm3, 1.2g/cm3 and 1.1g/cm3, respectively. According to the physical parameters of the fertilizer apparatus, the approximate target mass flow rates could be acquired by adjusting the axis speeds, and the target mass flow rates were ranged from 3g/s to 15g/s with increment of 1g/s. With respect to each mass flow rate, four replicates were conducted. Each replicate lasted for 30 s, and the fertilizer apparatus was started at the same time with the signal acquisition system. The average induced current and average mass flow rate were used to establish related regression equations, thus the real-time mass flow rate could be got by interpolation method. Subsequently, totally 25 experiments of each fertilizer were conducted to study the measurement accuracy, the targeted mass flow rates for each experiment were composited by five randomized mass flow rates, and each mass flow rate would last for 6s. The real mass during the 30s would be weighed by balance, while the detective mass was calculated by integrating the mass flow rate and time curves. The experimental results showed that there was no significant difference between the detective mass and the real mass (P>0.1), and the detective errors for large granular urea, superphosphate as well as potassium chloride were 3.9%, 5.1% and 5.9%, the corresponding standard errors were 5.21, 7.98 and 11.29. In regards to the granular fertilizer, the larger of the superficial area was, the easier of getting induced charge and saturation was. Consequently, the induced current would be larger, and the detective ring electrode was more sensitive on relative larger induced current. The mean diameters of the large granular urea, superphosphate and potassium chloride were 4.43mm, 2.77mm and 2.03mm, so the mean superficial areas should be in the same order, conclusions that generated from the research results showed that the detective error was increased along with the decrease of granular dimensions. SPSS 22.0 was used to handle further statistical analysis, the error distributions of three fertilizers were accorded with normal distribution, which meant the errors would be within ±6% under most of circumstances, the mathematical expectations of the detective errors were 3.74%,4.93% and 5.22% for large granular urea, superphosphate and potassium chloride respectively. The mass flow rate sensor that used for granular fertilizer could satisfy the requirements of real-time detection, the test-bed that designed could provide references for the research of variable rate fertilization.

    参考文献
    相似文献
    引证文献
引用本文

贾洪雷,温翔宇,王刚,刘慧力,郭慧.颗粒肥料质量流量传感器设计与试验[J].农业机械学报,2020,51(s1):130-136. JIA Honglei, WEN Xiangyu, WANG Gang, LIU Huili, GUO Hui. Design and Experiment of Mass Flow Sensor for Granular Fertilizer[J]. Transactions of the Chinese Society for Agricultural Machinery,2020,51(s1):130-136.

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2020-08-02
  • 最后修改日期:
  • 录用日期:
  • 在线发布日期: 2020-11-10
  • 出版日期: