Information Price Intelligent Prediction Model of Major Equipment in Power Grid Projects Based on the SSA-LSSVM Model

At present, the estimation for the next period information prices of equipment and materials of the power grid projects are based on those prices collected from the bid prices of equipment and materials which have been tendered recently. The source is real and reliable, but those prices are not the market prices due to time lag. During this period, if bulk raw material prices rise and fall sharply, using those prices gathered from the previous period to guide the information prices at this stage, would result in a large exceed or surplus of the project costs between two phases. That is not conducive to the control of the project costs. It is particularly important to pay close attention to the timeliness of the equipment and material prices, because of their large proportion of the project costs. In this paper, the Salp Swarm Algorithm (SSA)-the Least Squares Support Vector Machine (LSSVM) prediction model is used. On the one hand, SSA can improve the prediction accuracy of LSSVM. On the other hand, it can introduce multiple factors such as the production costs, the market environment and the policy environment to forecast the information prices and solve complex nonlinear problems. Ultimately, the prediction accuracy of the information prices can be improved and the price trends can be predicted in advance, which solve the defects of the statistical information prices from the big data sample. The empirical result shows that the SSA-LSSVM model has good applicability and practical value.