Combination load forecasting method for CCHP system based on IOWA operator

Abstract

Load forecasting is the basis of the design and implementation of the control strategy of the combined cooling heating and power (CCHP) system, and the precision affects the comprehensive energy efficiency of the system directly. In this paper, the gray relational analysis method is used to indicate the strong coupling relationship among the loads of heating, cooling and electricity in the system. Furthermore, a load forecasting method with the least squares support vector regression (LS-SVR) prediction and the radial basis function neural network (RBF Neural Network) prediction combined based on induced ordered weighted averaging (IOWA) operator is proposed by establishing the optimal model based on the minimum sum of error squares. The simulation results based on the historical load data of a CCHP system show that the accuracy of the multivariate combination forecasting method proposed in this paper is higher than that of single variable prediction method and the single prediction method, and the feasibility and effectiveness of the combination load forecasting method based on IOWA operator are verified.