Grid New Investment Demand Forecast Based on IBA-LSSVM under Grid-Load Investment Interface Extension

The division of investment interface and investment demand prediction of load-side industrial extension projects of power grids play an important role in lean management and power construction of power grid enterprises under the new situation. For the extension of residential load-side investment interface, the factors influencing the new investment demand of power grid are analyzed from two aspects of macroeconomics and power grid scale. The data of new investment amount and related new investment demand influencing factors are used as training set. The parameters of the least squares support vector machine (LSSVM) were optimized by using an improved bat algorithm (IBA) and a grid investment demand forecasting model with investment interface extension is constructed. The actual historical data of new investment demand in a region of China is used as the training data, and the new investment demand of the power grid in the next three years is predicted. The results show that the model has a high prediction accuracy compared with the prediction results of BP neural network and LSSVM.