The research on modeling and simulation of crude oil output prediction based on KPCA-DE-SVM

The accurate prediction of crude oil output plays an important role in the deployment of oilfield development and ensuring stable production. Crude oil output forecast is the premise and the core project management system of the whole oil production, while crude oil output is a dynamic system affected by multivariate variables. To accurately predict crude oil output, this paper presents a method to predict crude oil output in combination of kernel principal component analysis (KPCA) and differential evolution algorithm optimized support vector machine (DE-SVM). Firstly, the influence factors of crude oil output are extracted by using nuclear principal component analysis. Secondly, the parameters of support vector machine are optimized by using differential evolution algorithm. Finally, the prediction model of support vector machine based on differential evolution is constructed. The results show that the prediction accuracy is high and the error is stable at 1.5%, which has good application value.