基于神经网络PID算法的优化饲料配制系统

doi:10.6048/j.issn.1001-4330.2023.04.026

新疆农业科学 ›› 2023, Vol. 60 ›› Issue (4): 1003-1010.DOI: 10.6048/j.issn.1001-4330.2023.04.026

• 草业·植物保护·畜牧兽医·农业经济 • 上一篇下一篇

基于神经网络PID算法的优化饲料配制系统

方杰¹(), 张杰², 马娟², 田翔², 于秀针², 冯斌²()

1.新疆农业大学机电工程学院,乌鲁木齐 830000
2.新疆农业科学院农业机械化研究所,乌鲁木齐 830000

收稿日期:2022-08-11 出版日期:2023-04-20 发布日期:2023-05-06
通信作者: 冯斌(1968-),男,陕西人,研究员,硕士生导师,研究方向为智能化养殖,(E-mail)xjwsfb@sina.com
作者简介:方杰(1996-),男,安徽人,硕士研究生,研究方向为智能化养殖,(E-mail)fang5431_hf@qq.com
基金资助:
财政部和农业农村部:国家现代农业产业技术体系(CARS-37)

Research and optimization of feed preparation system based on neural network PID algorithm

FANG Jie¹(), ZHANG Jie², MA Juan², TIAN Xiang², YU Xiuzhen², FENG Bin²()

1. College of Mechanical and Electrical Engineering, Xinjiang Agricultural University, Urumqi 830052, China
2. Institute of Agricultural Mechanization, Xinjiang Academy of Agricultural Sciences, Urumqi 830091, China

Received:2022-08-11 Online:2023-04-20 Published:2023-05-06
Correspondence author: FENG Bin (1968-), male, born in Shaanxi, professor, research direction: Beef cattle breeding facilities, equipment and intelligent technology, (E-mail)xjwsfb@sina.com
Supported by:
National Beef Cattle and Yak Industrial Technology System Project(CARS-37)

摘要/Abstract

摘要：

【目的】设计饲料配制控制系统,并采用神经网络PID优化算法实现对配料精度的提高。【方法】以西门子S-200 smart型PLC为主控设计饲料配制控制系统,针对现有常规PID算法的控制策略存在超调大、收敛慢等缺陷和BP神经网络梯度下降过程容易出现局部最小化问题,提出以附加动量项的BP神经网络PID算法实现称重误差的降低。【结果】基于动量项的梯度下降法建立的BP神经网络PID算法模型解决了参数自学习整定问题,在响应速度上该算法与PID算法对比为3:1,试验后平均精度99.6%。并在收敛速度和改善超调现象具有更高效的表现。【结论】配料系统经算法优化后误差得到有效控制。

关键词: 自动配料; PLC控制; 动量因子; BP神经网络PID算法

Abstract:

【Objective】To design a feed preparation control system in view of the operation requirements of feed weighing and batching in small and medium-sized pastures and the weighing error of feed dynamic weighing by using the neural network PID optimization algorithm to improve the batching accuracy. 【Methods】Siemens S-200 smart PLC was taken as the main control, the feed preparation control system was designed, which led to summary that the control strategy of the existing conventional PID algorithm had the defects of large overshoot and slow convergence, and the local minimization problem was easy to occur in the gradient descent process of BP neural network. A BP neural network PID algorithm with additional momentum term was proposed to reduce the weighing error. 【Results】The BP neural network PID algorithm model based on the gradient descent method of momentum term solved the problem of parameter self-learning tuning, and had more efficient performance in convergence speed and improving overshoot.【Conclusion】The actual verification shows that the batching error is effectively controlled.

Key words: automatic batching; PLC control; momentum factor; BP neural network PID algorithm

中图分类号:

方杰, 张杰, 马娟, 田翔, 于秀针, 冯斌. 基于神经网络PID算法的优化饲料配制系统[J]. 新疆农业科学, 2023, 60(4): 1003-1010.

FANG Jie, ZHANG Jie, MA Juan, TIAN Xiang, YU Xiuzhen, FENG Bin. Research and optimization of feed preparation system based on neural network PID algorithm[J]. Xinjiang Agricultural Sciences, 2023, 60(4): 1003-1010.

图/表 10

参考文献 20

[1]	苏开华. PLC技术在饲料自动化生产上的应用研究进展[J]. 饲料研究, 2021, 44(1):100-103.
	SU Kaihua. Application of PLC technology in feed automation production Progress[J]. Feed Research, 2021, 44(1): 100-103.
[2]	薛蕾. 不同饲料称量系统对饲料生产效率及经济效益的影响[J]. 饲料研究, 2020, 43(12):133-135.
	XUE Lei. Effects of different feed weighing systems on feed production efficiency and economic efficiency Effects of benefits[J]. Feed Research, 2020, 43(12): 133-135.
[3]	车璐, 荣迪, 马丽芳, 等. 饲料配料精度的影响因素分析[J]. 饲料博览, 2018,(9):12-15.
	CHE Lu, RONG Di, MA Lifang, et al. Factors affecting the accuracy of feed ingredients Analysis[J]. Feed Expo, 2018,(9): 12-15.
[4]	何武林. 基于PLC的饲料生产过程自动控制系统设计[J]. 中国饲料, 2019,(2):90-93.
	HE Wulin. Design of automatic control system for feed production process based on PLC[J]. Chinese Feed, 2019,(2):90-93.
[5]	刘文秀, 郭伟. PLC配料称重系统的模糊自适应控制[J]. 沈阳工业大学学报, 2018, 40(6):676-681. DOI
	LIU Wenxiu, GUO Wei. Fuzzy adaptive control of PLC batching and weighing system[J]. Journal of Shenyang University of Technology, 2018, 40(6): 676-681. DOI
[6]	刘超, 程国华. 基于BP神经网络饲料配料误差补偿模型[J]. 制造业自动化, 2011, 33(13):44-47.
	LIU Chao, CHENG Guohua. Feed proportioning error compensation based on BP neural network Model[J]. Manufacturing Automation, 2011, 33(13): 44-47.
[7]	苏和, 韩治平, 浩红军. 牧畜饲料加工过程中的自动化控制与管理[J]. 内蒙古师范大学学报(自然科学汉文版), 2014, 43(5):568-572.
	SU He, HAN Zhiping, HAO Hongjun. Automation in animal feed processing Control and management[J]. Journal of Inner Mongolia Normal University (Natural Science Chinese Ed.), 2014, 43(5): 568-572.
[8]	门洪, 英宇翔, 王忠. 基于PLC的智能水泥自动配料系统设计[J]. 制造业自动化, 2013, 35(1):128-130, 136.
	MEN Hong, YING Yuxiang, WANG Zhong. Design of intelligent cement automatic batching system based on PLC[J]. Manufacturing Automation, 2013, 35(1): 128-130, 136.
[9]	曹应明, 顾菊芬. 基于PLC和Wincc的饲料厂智能监控系统设计[J]. 饲料工业, 2012, 33(1):11-13.
	CAO Yingming, GU Jufen. Design of intelligent monitoring system for feed plant based on PLC and WinCC[J]. Feed Industry, 2012, 33(1): 11-13.
[10]	李伟, 朱钒, 邵新宇. 基于组态软件WINCC的配料监控系统开发与设计[J]. 仪表技术与传感器, 2010,(4):41-44.
	LI Wei, ZHU vanadium, SHAO Xinyu. Development and design of batching monitoring system based on Configuration Software WinCC[J]. Instrument Technology and Sensor, 2010,(4): 41-44.
[11]	韩治国, 陈能祥, 许锦, 等. 基于BP神经网络的倒立摆智能控制研究[J]. 实验技术与管理, 2021, 38(6):101-106.
	HAN Zhiguo, CHEN Nengxiang, XU Jin, et al. Research on intelligent control of inverted pendulum based on BP neural network[J]. Experimental Technology and Management, 2021, 38(6): 101-106.
[12]	包志家, 李奇. 基于MATLAB神经网络工具箱的BP神经网络的应用研究[J]. 信息与电脑(理论版), 2021, 33(2):181-183.
	BAO Zhijia, LI Qi. Application Research of BP neural network based on MATLAB neural network toolbox[J]. Information and Computer (Theoretical Ed.), 2021, 33(2): 181-183.
[13]	胡黄水, 赵思远, 刘清雪, 等. 基于动量因子优化学习率的BP神经网络PID参数整定算法[J]. 吉林大学学报(理学版), 2020, 58(6):1415-1420.
	HU Huangshui, ZHAO Shiyuan, LIU Qingxue, et al. BP neural network PID parameter tuning algorithm based on momentum factor optimized learning rate[J]. Journal of Jilin University (Science Edition), 2020, 58(6): 1415-1420.
[14]	辛梅. 基于神经网络的车辆动态称重系统设计和应用[J]. 微型电脑应用, 2020, 36(5):142-144.
	XIN Mei. Design and application of vehicle dynamic weighing system based on neural network[J]. Microcomputer Application, 2020, 36(5): 142-144.
[15]	郭骅祥, 杨延超. 基于OPC协议的数据采集实际应用分析[J]. 工业控制计算机, 2019, 32(9):40-41, 43.
	GUO Huaxiang, YANG Yanchao. Practical application analysis of data acquisition based on OPC protocol[J]. Industrial Control Computer, 2019, 32(9): 40-41, 43.
[16]	张震, 赵文豪, 张晨宇, 等. 基于MATLAB的PID参数整定的仿真与分析[J]. 装备制造技术, 2019,(3):79-81, 85.
	ZHANG Zhen, ZHAO Wenhao, ZHANG Chenyu, et al. Simulation and analysis of PID parameter setting based on MATLAB[J]. Equipment Manufacturing Technology, 2019,(3): 79-81, 85.
[17]	庄育锋, 胡晓瑾, 翟宇. 基于BP神经网络的微量药品动态称重系统非线性补偿[J]. 仪器仪表学报, 2014, 35(8):1914-1920.
	ZHUANG Yufeng, HU Xiaojin, ZHAI Yu. Nonlinear compensation of micro drug dynamic weighing system based on BP neural network[J]. Journal of Instrumentation, 2014, 35(8): 1914-1920.
[18]	宋立业, 程英, 李志富, 等. 基于BP神经网络的加药控制系统研究[J]. 计算机测量与控制, 2012, 20(2):389-391.
	SONG Liye, CHENG Ying, LI Zhifu, et al. Research on dosing control system based on BP neural network[J]. Computer Measurement and Control, 2012, 20(2): 389-391.
[19]	庞强, 苑明哲, 吴星刚, 等. 基于神经网络的水泥生料配料多目标优化设计方法[J]. 东南大学学报(自然科学版), 2009, 39(S1):76-81.
	PANG Qiang, YUAN Minzhe, WU Xingang, et al. Multi objective optimization design method of cement raw meal batching based on neural network[J]. Journal of Southeast University (Natural Science Ed.), 2009, 39(S1): 76-81.
[20]	第一机械部机械研究院. 农业机械设计手册[M]. 机械工业出版社, 1973.
	Machinery Research Institute of the First Ministry of Machinery. Design Manual of Agricultural Machinery[M]. Beijing: Machinery Industry Press, 1973.

序号 Order number	精料重量 Weight of concentrate (kg)		粗料重量 Weight of coarse material (kg)
序号 Order number	优化前 Before optimization	优化后 After optimization	优化前 Before optimization	优化后 After optimization
1	401.66	401.60	400.01	400.60
2	402.38	400.48	401.90	400.17
3	402.46	400.42	401.70	399.79
4	401.96	401.64	399.05	400.75
5	405.45	401.62	401.84	400.45
6	400.68	400.41	401.08	399.71
7	405.69	401.00	402.96	399.86
8	401.68	401.76	402.62	401.09
9	401.66	401.60	400.71	400.01

序号 Order number	精料重量 Weight of concentrate (kg)		粗料重量 Weight of coarse material (kg)
序号 Order number	优化前 Before optimization	优化后 After optimization	优化前 Before optimization	优化后 After optimization
1	401.66	401.60	400.01	400.60
2	402.38	400.48	401.90	400.17
3	402.46	400.42	401.70	399.79
4	401.96	401.64	399.05	400.75
5	405.45	401.62	401.84	400.45
6	400.68	400.41	401.08	399.71
7	405.69	401.00	402.96	399.86
8	401.68	401.76	402.62	401.09
9	401.66	401.60	400.71	400.01

序号 Order number	精料 Spermatic dapartment (kg)	青储 Ensilage (kg)	苜蓿秸秆 Alfalfa (kg)	总计 Total (kg)	总偏差 Total departure (kg)
1	201.43	401.27	600.78	1 203.48	4.94
2	200.98	402.55	599.11	1 201.94	4.42
3	200.46	403.01	600.14	1 203.61	5.08

序号 Order number	精料 Spermatic dapartment (kg)	青储 Ensilage (kg)	苜蓿秸秆 Alfalfa (kg)	总计 Total (kg)	总偏差 Total departure (kg)
1	201.43	401.27	600.78	1 203.48	4.94
2	200.98	402.55	599.11	1 201.94	4.42
3	200.46	403.01	600.14	1 203.61	5.08

基于神经网络PID算法的优化饲料配制系统

Research and optimization of feed preparation system based on neural network PID algorithm

在线阅读

PDF下载

可视化

摘要/Abstract

引用本文

使用本文

图/表 10

参考文献 20

相关文章 0

编辑推荐

Metrics