Soft sensor study of effluent COD based on mutant CPSO and LSSVM
基于变异CPSO算法的LSSVM出水COD的软测量研究

On the basis of analyzing chaotic particle swarm optimization(CPSO) and least square support vector machine(SVM) theory,for oxidation ditch system in a wastewater treatment plant,applying the chaotic PSO algorithm with last out mechanism to optimize the parameters of SVM,a mutant CPSO-LSSVM soft sensor model was used to predict effluent COD of oxidation ditch system.Then compared with PSO-LSSVM and LSSVM model,the study showed that the ICPSO-LSSVM model was featured with more accuracy and better generalization ability and performance,and the samples with prediction relative error less than 10% were up to 90%,the maximal relative error was only 12.5%,and the MSE was 0.0106.Moreover,the model has high accuracy.It can be used to realizes the estimation online for effluent COD basically.