中文说明:污水处理过程中关键水质参数无法在线监测的问题,提出基于PCA GABP神经网络的污水水质软测量方法。该方法由两部分组成:主元分析PCA和GABP神经网络。其中,GABP算法采用局部改进遗传算法优化神经网络权值,并采用自适应学习速率动量梯度下降算法对神经网络进行训练,建立软测量模型。仿真结果表明该软测量模型稳定性好、精度高,可用于污水处理厂对BOD进行在线预测。
English Description:
Key water quality parameters in the wastewater treatment process beyond the on-line monitoring of problem of sewage quality based on PCA GABP neural network soft sensor methods. This method consists of two parts: principal component analysis PCA and GABP neural networks. Among them, the GABP algorithm using an improved genetic algorithm optimizing neural network weights, and adaptive learning rate momentum gradient descent algorithm for training neural networks, build model. Simulation results show that the model has good stability, high accuracy, can be used for online prediction of the sewage treatment plant on the BOD.