Cost Analysis of Power Engineering Based on Improved Random Forest Algorithm

Enhancing power engineering management mostly depends on raising the precision of power engineering cost forecast. Nevertheless, many variables and limited data influence the price of electric power engineering. As a result, it is crucial to forecast electric power project costs precisely in high-dimensional and small-sample scenarios. The duplicate indices are successfully removed when the data is mined using the random forest method (RF). The accuracy and stability of the cost forecast for power projects are increased when the support vector machine (SVM) model is optimized using the wolf pack algorithm (WPA), which successfully addresses the issues of SVM overfitting and readily slipping into local optimality. It has been shown via case analysis and model comparison that the suggested approach performs better regarding predictions.