| 不可测过程参数的智能化软计算方法及应用 |
| |
| 引用本文： | 刘载文,王小艺,崔莉凤,苏维均. 不可测过程参数的智能化软计算方法及应用[J]. 清华大学学报(自然科学版), 2007, 47(Z2): 1742-1746 |
| |
| 作者姓名： | 刘载文 王小艺 崔莉凤 苏维均 |
| |
| 作者单位： | 1. 北京工商大学,信息工程学院,北京,100037
2. 北京工商大学,化学与环境工程学院,北京,100037 |
| |
| 基金项目： | 北京市自然科学基金;北京市优秀人才培养基金;北京市高校人才强校计划项目 |
| |
| 摘 要： | 为解决控制系统中存在的时变过程数学建模困难,有些过程参数难以在线实时检测的问题,研究基于过程神经元网络的不可测过程参数软计算方法。分析正交基函数展开的学习算法收敛速度慢的问题,将BP网络的改进算法引入到过程神经元网络的训练中,增加基函数展开系数的规一化处理、权函数动量项调整项,以及学习率自适应调整方法,改进算法明显消除了误差收敛曲线振荡和网络收敛速度慢的问题。该方法在污水处理过程出水水质BOD预测中网络训练速度快,取得较好的预测结果,是一种不可测过程参数智能化软计算的有效方法。
|
| 关 键 词： | 软计算 智能化 神经网络 算法 污水处理 |
| 文章编号： | 1000-0054(2007)S2-1742-05 |
| 修稿时间： | 2007-04-12 |
Computing method for unmeasurable process parameters |
| |
| LIU Zaiwen,WANG Xiaoyi,CUI Lifeng,SU Weijun. Computing method for unmeasurable process parameters[J]. Journal of Tsinghua University(Science and Technology), 2007, 47(Z2): 1742-1746 |
| |
| Authors: | LIU Zaiwen WANG Xiaoyi CUI Lifeng SU Weijun |
| |
| Abstract: | A soft computing method was developed to compute unmeasurable process parameters using a process neural network.The system simplifies the mathematic modeling to determine process parameters that are difficult to measure online in time-depandent control systems.An improved BP algorithm was developed for the training process to improve the slow convergence of the learning algorithm based on orthogonal basis functions.The orthogonal basis function,momentum adjustment term,and automatic learning rate adjustment method are normalized in the analysis.The oscillation of the error convergence curve and the slow convergence are eliminated by the modified algorithm.The method is used to predict the effluent BOD from a wastewater treatment process.The results show that the method can efficiently compute unmeasurable process parameters. |
| |
| Keywords: | soft computing intelligence neural network algorithm wastewater treatment |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
