Predicting Model of Power Engineering Cost Based on the FCM and PSO-SVM
In order to accurately estimate the cost level of the new substation project,this paper proposes a predicting model based on the combination of the fuzzy clustering method (FCM) and the support vector machine optimized by the particle swarm algorithm (PSO-SVM).Through fuzzy clustering analysis,the project samples with a high degree of similarity were classified,so that the sample rules in the same class are easier to identify.Then the PSO-SVM was used for cost forecasting of each class respectively.Compared with single forecasting model,the prediction accuracy of the seven test samples which used the PSO-SVM prediction model based on clustering analysis were reduced to less than 5％,proving the validity and accuracy of this method.
NETL
NSTL
万方数据
维普
