Research on Soft-sensing of Industrial Sewage Measurement Based upon Improved BPNN Model
(0) The parameter model for industrial sewage soft-sending system, which is an online forecasting mode, is established based on improved back propagation neural network (BPNN) model, and it is expected to realize the real-time feedback control of water quality. The SBR (Sequencing Batch Reactor Active Sludge process) method is proposed for sewage soft-sensing, the primary variables, which include COD and TP, are forecasted by the secondary variables integration of DO, OPR, PH and MLSS. After a series of simulation, the 4-6-2 topology structure for BPNN is designed, owing to the improved weights adjustment algorithm with the inertia items, the performance of net convergence is enhanced as well. The simulation results demonstrate that the error RMS of COD and TP never exceed o.1, the soft-sensing model based on BPNN technology is testified to be an easy one with strong robustness, which adapts to the practical applications.
soft-sensing sewage measurement BPNN SBR1 primary variable
XIA Linlin WANG Jianguo ZHANG Lihui Miao Guijuan
School of Automation Engineering, Northeast Dianli University, Jilin, Jilin, China, 132012
国际会议
第八届国际测试技术研讨会(8th International Symposium on Test and Measurement)
重庆
英文
1627-1630
2009-08-01（万方平台首次上网日期，不代表论文的发表时间）