Uncertain Interval Forecasting for Combined Electricity-Heat-Cooling-Gas Loads in the Integrated Energy System Based on Multi-Task Learning and Multi-Kernel Extreme Learning Machine

The accurate prediction of electricity-heat-cooling-gas loads on the demand side in integrated energy system (IES) can provide significant reference for multiple planning and stable operation IES. This paper combines multi-task learning (MTL) method, Bootstrap improved Salp Swarm Algorithm (ISSA) multi-kernel extreme machine (MKELM) method to establish uncertain interval model loads. ISSA introduces dynamic inertia weight chaotic local searching mechanism into basic SSA improve speed avoid falling optimum. MKELM is established by combining RBF kernel function Poly integrate superior ability generalization two functions. Based model, weather, calendar information, social–economic factors, historical load are selected as input variables. Through empirical analysis comparison discussion, we obtain: (1) results workday better than those holiday. (2) Bootstrap-ISSA-MKELM based MTL has performance that STL method. (3) comparing discover combined prediction.