| 140 | 3 | 19 |
| 下载次数 | 被引频次 | 阅读次数 |
水厂投药系统由于存在非线性、大滞后、多输入因子等特点,多年来难以实现自动控制.在充分分析自来水厂现有的大量有效历史数据和丰富实践经验的基础上,建立了基于Elman动态神经网络的智能混凝投药控制方式,并对系统的实现作了较为详细的说明.实践证明,该系统具有自学习和自适应的能力,控制精度高,克服了传统人工操作的失误隐患,并取得了较好的经济效益.
Abstract：Control system for chemical feeding in waterworks have the properties of nonlinear,large time-delay,multi-input factor.It is difficult to realize automatic control.Based on the large effective historical data and plentiful experience in waterworks,establishes intelligent coagulant chemical feeding method based on Elman dynamical neural network,then system realization is in-troduced in detail.The result of one year's application shows that the control system have the abilities of self-study and self-adaptive,high control precision.It also overcomes the mistakes of hidden danger of traditional control mode and good economic benefit has been obtained.
[1]江智军,何小斌,杨晓辉.自来水厂混凝投药控制系统的设计及应用[J].计算机测量与控制,2006,14(4):462-463.
[2]蔡利民,杨晓林.用人工神经网络设计和实现自来水投矾控制[J].江汉大学学报:自然科学版,2008,36(3):35-37.
[3]CHENG Y C,QI W M,CAI W Y.Dynamic properties of Elman and modified Elman neural network[C]//Proc of the1st Int Conf on Machine Learning and Cybernetics,Bei-jing,2002,637-640.
[4]BOGER Z.Applications of neural networks to water and wastewater treatment plant operation[J].ISA Transactions,1992,31(1):25-31.
[5]漆为民,程远楚,姬巧玲,蔡维由.PID型Elman网络及在动态系统辨识中的应用研究[J].控制与决策,2005,20(10):1197-1200.
[6]GAO X Z,GAO X M,OCASKA S J.A modified Elman neural network model with application to dynamical sys-tems identification[C]//IEEE Int Conf on Systems,Man,and Cybernetics,Beijing,1996:1376-1381.
基本信息:
中图分类号:TP273.5
引用信息:
[1]漆为民,杨晓林.改进型Elman网络在水厂智能投药控制系统中的应用[J].江汉大学学报(自然科学版),2009,37(01):66-69.
基金信息:
武汉市科技计划项目(200751699478-03)