Cost Forecasting of Power Engineering based on RF and SVM optimized by WPA

The improvement of the accuracy of power engineering cost prediction plays a key role in improving the management of power engineering. However, there are many factors that affect the cost of electric power engineering, and the amount of data is relatively small. Therefore, it is very important to accurately predict the cost of electric power projects under high dimensional and small sample conditions. The random forest algorithm (RF) is used to mine the data, and the redundant indexes are eliminated effectively. The optimization of support vector machine (SVM) model by using wolf pack algorithm (WPA) effectively solves the problems of SVM overfitting and easily falling into local optimality, which improves the accuracy and stability of the power project cost prediction. By example analysis and comparison with other models, it is proved that the proposed method has better prediction performance.