Soft Measurement Modeling of Wastewater Treatment Process Based on IDBO-ELM
Aiming at the problem that the biochemical oxygen demand(BOD)and other water quality parameters in the sew-age treatment process are greatly affected by other environmental factors and it is difficult to establish an accurate measurement mod-el,an improved dung beetle optimizer(IDBO)extreme learning machine(ELM)is proposed.The method predicts the BOD concen-tration of effluent.Firstly,the Random Forest Algorithm(RFA)is selected to screen out factors with high correlation with BOD as input variables of the soft measurement model.Secondly,the IDBO algorithm is utilized to optimize and decide the ELM weight allo-cation to improve the prediction accuracy of the ELM network,and the Tent chaos mapping is used to increase the population diver-sity of the DBO method.Finally,the designed IDBO-ELM soft measurement model is applied to the wastewater treatment simulation platform and compared with different prediction models.The results show that the IDBO-ELM prediction model designed in this pa-per obtains higher prediction accuracy and more stable network structure.
万方数据
维普
