• EI
    • CSA
    • CABI
    • 卓越期刊
    • CA
    • Scopus
    • CSCD
    • 核心期刊

基于广义回归神经网络的灌溉系统首部多用户配水快速PID控制

张千, 魏正英, 张育斌, 张磊, 贾维兵, 魏浩然

张千, 魏正英, 张育斌, 张磊, 贾维兵, 魏浩然. 基于广义回归神经网络的灌溉系统首部多用户配水快速PID控制[J]. 农业工程学报, 2020, 36(11): 103-109. DOI: 10.11975/j.issn.1002-6819.2020.11.012
引用本文: 张千, 魏正英, 张育斌, 张磊, 贾维兵, 魏浩然. 基于广义回归神经网络的灌溉系统首部多用户配水快速PID控制[J]. 农业工程学报, 2020, 36(11): 103-109. DOI: 10.11975/j.issn.1002-6819.2020.11.012
Zhang Qian, Wei Zhengying, Zhang Yubin, Zhang Lei, Jia Weibing, Wei Haoran. Rapid-response PID control technology based on generalized regression neural network for multi-user water distribution of irrigation system head[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2020, 36(11): 103-109. DOI: 10.11975/j.issn.1002-6819.2020.11.012
Citation: Zhang Qian, Wei Zhengying, Zhang Yubin, Zhang Lei, Jia Weibing, Wei Haoran. Rapid-response PID control technology based on generalized regression neural network for multi-user water distribution of irrigation system head[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2020, 36(11): 103-109. DOI: 10.11975/j.issn.1002-6819.2020.11.012

基于广义回归神经网络的灌溉系统首部多用户配水快速PID控制

基金项目: 十三五国家重点研发计划项目(2017YFD0201504);国家科技支撑计划项目(2015BAD24B00)

Rapid-response PID control technology based on generalized regression neural network for multi-user water distribution of irrigation system head

  • 摘要: 针对多用户配水状态下灌区流量、压力需求变化范围大,传统流量、压力控制响应速度慢等问题,建立适用于多用户灌区配水的灌溉系统首部控制技术。该研究通过分析供水系统流量、压力调节方式,提出了流量、压力PID(Proportion Integration Differentiation)耦合调节方法,建立以电动阀开度为流量控制量、变频器频率为压力控制量对流量和压力进行调控的灌溉首部控制系统。为了减少系统的调节时间,提高系统的运行效率,采用广义回归神经网络(Generalized Regression Neural Network,GRNN)建立流量、压力和电动阀控制模拟量、变频器控制模拟量间的预测模型,形成神经网络PID控制模型(GRNN_PID),并进行模型精度和控制精度验证。GRNN训练结果显示,变频器控制模拟量的相对误差为0.11%~3.86%,电动阀控制模拟量相对误差为0.09%~5.74%,模型精度较高。使用3个调节过程模拟3个用户的需水行为对模型进行验证,结果表明,GRNN_PID模型3个过程的调节时间分别为11.6 、10.7 和7.2 s,PID模型3个过程的调节时间分别为31.7 、29.6 和16.9 s,GRNN_PID模型大幅减少了系统的调节时间,提高了系统的运行效率;分别计算了2种模型的控制精度,GRNN_PID调节方法和传统PID调节方法的稳态流量和压力误差都在1%以内,最大超调量为8%,控制精度较高但相差不大,表明GRNN是从策略上加速系统调节速度,其本身并没有对PID的参数进行调整,因此对系统的控制精度影响不大。研究可为灌溉系统流量压力快速控制提供参考。
    Abstract: Abstract: Flow rate and pressure are two important control parameters of the irrigation system and they determine if the irrigation process can meet the irrigation demand or not. In the same pipeline system, flow rate and pressure affect each other, and it is difficult to control them separately. This study explored the relationship between flow rate and pressure and their control parameters, established a control technology suitable for the head of water distribution of the multi-user irrigation system. By analyzing the water supply characteristic curve of the pipeline water supply system, a flow pressure coupling adjustment method was proposed. This method allowed system controller output voltage analog and current analog to control the frequency of the inverter and the valve opening of the electric valve. At the same time, a proportion integration differentiation (PID) controller was used to realize the coupling control of the flow rate and the pressure. In addition, a method that use generalized regression neural network (GRNN) to accelerate the speed of PID (GRNN_PID) was proposed in order to speed up the response of irrigation control system, improve the operating efficiency and ensure the safety of the system. The flow, pressure and corresponding control quantities data were obtained through experiments, and the GRNN method was used to fit the relationship between them. Afterwards, the control quantities required for the target flow and pressure were quickly calculated by the fitting model and were used directly to control the corresponding actuator (pump and electric valve) so as to fine-tune the flow and pressure through PID. The GRNN training results showed that the relative error of the analog quantity used to control the frequency of the frequency converter was between 0.11% and 3.86%, and the relative error of the analog quantity used to control the electric valve was between 0.09% and 5.74%, indicating that the GRNN model has a high fitting accuracy. Three adjustment processes was used to simulate the water demand behavior of three users to verify the control model, and the tests showed that the adjustment times of the three processes of the GRNN_PID model were 11.6, 10.7 and 7.2 s, respectively, and the adjustment times of the three processes of the PID model were 31.7, 29.6 and 16.9 s, respectively. The GRNN_PID model greatly reduced the adjustment time of the system and improved the operating efficiency of the system. The control accuracies of GRNN_PID and traditional PID adjustment method were compared, and the results showed that the steady-state error of the GRNN_PID adjustment method and the traditional PID adjustment method were both within 1%, and the maximum overshoot was below 8%, which means that the control accuracy is high but the difference is not very big. The reason that leads to this result above is that GRNN accelerates the system regulation speed from the strategy, and it does not change the parameters of PID, and thus it has little impact on the control accuracy of the system. This research can provide tools for the rapid control of flow and pressure in irrigation system.
  • [1] 崔昊杰,郭萍,李茉. 基于不确定性的区间分式规划灌区优化配水模型[J]. 中国农业大学学报,2018,23(3):111-121.Cui Haojie, Guo Ping, Li Mo. Interval fractional programming optimization model for irrigation water allocation under uncertainty[J]. Journal of China Agricultural University, 2018, 23(3): 111-121. (in Chinese with English abstract)
    [2] 张志政,王毅,矫亚涛. 管道灌溉流量控制模型研究[J]. 节水灌溉,2009(1):41-43.Zhang Zhizheng, Wang Yi, Jiao Yatao. Study on pipe flux control model[J]. Water Saving Irrigation, 2009(1): 41-43. (in Chinese with English abstract)
    [3] 张颖丽,刘元英,姜海. 大中型灌区信息化技术研究与应用[J]. 江苏水利,2019(10):61-64.Zhang Yingli, Liu Yuanying, Jiang Hai. Research and application of information in large and medium-sized irrigation areas[J]. Jiangsu Water Resources, 2019(10): 61-64. (in Chinese with English abstract)
    [4] 张勇刚. 水利灌区管理工作中的问题及对策[J]. 科技创新与应用,2016(29):218.
    [5] Singh A. Irrigation planning and management through optimization modelling[J]. Water Resources Management, 2014, 28(1): 1-14.
    [6] 付玉娟,蔡焕杰. 基于机会约束规划模型的灌溉管网不确定优化研究[J]. 西北农林科技大学学报:自然科学版,2008,36(5):219-224.Fu Yujuan, Cai Huanjie. Uncertain optimal design of irrigation pipe network with chance-constrained programming[J]. Journal of Northwest A&F University: Natural Science Edition, 2008, 36(5): 219-224. (in Chinese with English abstract)224.
    [7] 马建琴,陈哲,刘蕾. 农业多水源灌溉实时优化配置[J]. 江苏农业科学,2018,46(7):211-214.Ma Jianqin, Chen Zhe, Liu Lei. Real-time optimization of agricultural multi-water irrigation[J]. Jiangsu Agricultural Sciences, 2018, 46(7): 211-214. (in Chinese with English abstract)
    [8] 田云,史洁,金东琦. PLC变频调速节能灌溉系统的设计[J]. 农机使用与维修,2014(5):13-16.Tian Yun, Shi Jie, Jin Dongqi. Design of PLC frequency conversion and speed adjusting energy-saving in irrigation system[J]. Agricultural Mechanization Using & Maintenance, 2014(5): 13-16. (in Chinese with English abstract)
    [9] Caba S, Lepper M, Liu S. Nonlinear controller and estimator design for multi-Pump systems[C]: IEEE Conference on Control Technology and Applications: IEEE, 2018.
    [10] Fernando. Pump systems performance impacts multiple bottom lines: Engineering[J]. Engineering and Mining Journal, 2010: 56-61.
    [11] 李宝. 水流量标准装置变频调速稳压系统研究[D]. 天津:天津大学,2009.Li Bao. Research on Pressure Stabilization System of Water Flow Standard Facility Based on Frequency Conversion[D]. Tianjin: Tianjin University, 2009. (in Chinese with English abstract)
    [12] 刘永,谷立臣,杨彬,等. 液压系统流量、压力闭环控制实验研究[J]. 机床与液压,2017,45(7):23-25.Liu Yong, Gu Lichen, Yang Bin, et al. Experimental study on closed loop control of flow and pressure of hydraulic system[J]. Machine Tool & Hydraulics, 2017, 45(7): 23-25. (in Chinese with English abstract)
    [13] 刘汉忠,官元红. 模糊PID自适应算法在流量压力控制系统中的应用[J]. 化工自动化及仪表,2011,38(5):567-569.Liu Hanzhong, Guan Yuanhong. Application of fuzzy PID adaptive algorithm in flow-Pressure control system[J]. Control and Instruments in Chemical Industry, 2011, 38(5): 567-569. (in Chinese with English abstract)
    [14] 刘艳雄,李杨康,华林,等. 基于遗传算法精冲机快速缸液压伺服系统设计及PID控制优化[J]. 武汉理工大学学报:交通科学与工程版,2017,41(1):52-56.Liu Yanxiong, Li Yangkang, Hua Lin, et al. Rapidly cylinder hydraulic servo system design and optimization of PID control based on genetic algorithm[J]. Journal of Wuhan University of Technology:Transportation Science & Engineering, 2017, 41(1): 52-56. (in Chinese with English abstract)
    [15] 刘心漪. EAST快速控制电源灰色PID预测研究[D]. 合肥:合肥工业大学,2018.Liu Xinyi. The Study of EAST Fast Control Power Supply Grey Prediction and PID Control[D]. Hefei: Hefei University of Technology, 2018. (in Chinese with English abstract)
    [16] Mohammadi A, Ryu J. Neural network-based PID compensation for nonlinear systems: Ball-on-plate example[J]. International Journal of Dynamics and Control, 2020, 8(1): 178-188.
    [17] Yadav A K, Gaur P. An optimized and improved STF-PID speed control of throttle controlled HEV[J]. Arabian Journal for Science and Engineering, 2016, 41(9): 3749-3760.
    [18] 祁增慧. 基于PLC控制的城市恒压供水系统[D]. 天津:天津大学,2008.Qi Zenghui. Pressure Constant City Water Supply System Based on PLC Control[D]. Tianjin: Tianjin University, 2008. (in Chinese with English abstract)
    [19] 汤健,陈玮. 多台水泵并联的最优化方法[J]. 信息技术,2015(6):42-46.Tang Jian, Chen Wei. The optimal control method of muti-pumps in parallel[J]. Information Technology, 2015(6): 42-46. (in Chinese with English abstract)
    [20] Abdelsalam Ahmed,Basma Moharam,Essam Rashad. Power saving of multi pump-motor systems using variable speed drives[C]. Twentieth International Middle East Power Systems Conference. Egypt: Cairo University, 2018.
    [21] 黄双成,李志伟. OPC技术下MATLAB与PLC的通讯实现[J]. 机械工程与自动化,2014(3):192-193.Huang Shuangcheng, Li Zhiwei. Communication between MATLAB and PLC by OPC technology[J]. Mechanical Engineering & Automation, 2014(3):192-193.
    [22] Cao J, Ye Q, Li P. Resistance furnace temperature control system based on OPC and MATLAB[J]. Measurement and Control, 2015, 48(2): 60-64.
  • 期刊类型引用(5)

    1. 夏延秋,王春丽,冯欣,蔡美荣. 基于灰狼算法优化GRNN的润滑油摩擦磨损性能预测. 摩擦学学报. 2023(08): 947-955 . 百度学术
    2. 孙兆光. 基于支持向量机的农田水利灌溉分流机械运行异常检测. 机械与电子. 2022(07): 43-47 . 百度学术
    3. 朱爱华,戴光鑫. 基于贝叶斯神经网络的农田分区灌溉需水量模拟. 农业工程. 2022(07): 78-83 . 百度学术
    4. 宋占华,邢书仑,王征,田富洋,王锋德,李法德. 苜蓿调制试验台测控系统设计与试验. 农业机械学报. 2021(02): 122-134 . 百度学术
    5. 田敏,白金斌,李江全. 基于遗传算法的液肥变量施肥控制系统. 农业工程学报. 2021(17): 21-30 . 本站查看

    其他类型引用(1)

计量
  • 文章访问数:  836
  • HTML全文浏览量:  1
  • PDF下载量:  435
  • 被引次数: 6
出版历程
  • 收稿日期:  2020-01-14
  • 修回日期:  2020-05-02
  • 发布日期:  2020-05-31

目录

    /

    返回文章
    返回