Optimal load dispatch for electric heating users based on CNN-BiLSTM model

Adopting electric energy to replace traditional coal-fired and gas-fired heating methods is conducive to realizing sustainable development. However, electric energy alternative heating loads are characterized by distribution and randomness, which makes it difficult for dispatch centers to regulate them directly. To address this problem, we propose a load-optimized dispatch strategy based on the CNN-BiLSTM model. Firstly, we propose various load characteristic indexes and introduce a CNN-BiLSTM model, which fuses the user load characteristics to forecast their loads. Secondly, we propose a dispatch strategy to maximize the economic efficiency of the dispatch layer while ensuring that the users get the best thermal comfort. Finally, we employ an enhanced particle swarm algorithm to solve the optimal dispatch results. The simulation results illustrate that our proposed optimal dispatch scheme better incorporates the characteristics of the user loads and thus maximizes the revenue of the dispatch layer.