In order to solve the problems of poor generalization ability and low prediction accuracy of traditional water pH prediction model,an improved MEA optimized method for improving ELMwater pH prediction is proposed.On the basis of MEA,chaos theory,Levy flight and Cauchy and Gauss mixed mutation strategy were introduced,which effectively solved the shortcomings of MEA that was premature and easy to get into local optimum,and improved the search performance.The superposition of Morlet wavelet and inverse hyperbolic sine was taken as ELM's excitation function,and regularization term was introduced to improve ELMdynamic approximation performance and generalization ability.The model was applied to the pH prediction of Taihu Lake.The results show that compared with other models,the model has better generalization ability and higher prediction accuracy,which can provide a basis for management and decision making.
关键词
莱维飞行/思维进化算法/极限学习机/pH值
Key words
Levy Flight/Mind evolutionary algorithm/Extreme learning machine/pH
维普
万方数据