A new fault diagnosis method based on Bayesian network model in a wastewater treatment plant of northern China

A novel diagnosing faults method is presented using a Bayesian network (BNT) model to optimize system diagnosis for a wastewater treatment plant (WTP) in northern China. The BNT model is established according to the expert knowledge based on local conditions. The historical data of the WTP are employed to implement the parameter learning of the BNT model. Some practical cases are carried out by the BNT model based on the Bayesian inference. The diagnostic results are compared with the monitoring data of that day to verify accuracy of the BNT model. Meanwhile, several fault diagnosis results and improvement measures are given in this study. The results show that the proposed method is robust enough to diagnose the faults quickly and accurately so as to optimize the operation of the WTP.