Mid-long Term Load Forecasting Model Based on Principal Component Regression

Most of the factors influencing multivariate regression prediction models are macroeconomic indicators, and indicators tend to have strong correlations. The regression prediction model has strong multicollinearity, which results in distortion or inaccurate estimation of model estimates. To solve this problem, this paper first analyzes the principal components of the indicators, extracts the two principal components, eliminates the multicollinearity between the indicators, and considers the policy background of the “Electrification of Xinjiang”, and finally establishes a mid-long term load forecasting model based on principal component regression. The principal component regression prediction model has a better prediction effect. After inspection, the prediction accuracy of this model is ideal.