A kind of BP neural network application in Water Treatment Control System parameter tuning

摘要

In Thermal Power Generation Water Treatment process, duo to the pipeline pressure and flow changes, Water Treatment PID control system parameter unstable, causes the water treatment controller can not prompt tracking response, PH value of water changes in a larger range, increased damage to the boiler pipes. Three layers of BP neural network model can dynamically adjust the parameters of hidden layer through self-learning, hidden layer units, respectively as the proportion of PID (P) unit, integral (I) unit and differential (D) unit, so as to realize the ND parameters on-line tuning, to improve real-time of the water treatment control system, improved the stability of the system, and achieve a better control effect.