Optimal Electric Heating Load Dispatch Strategy with CNN-BiLSTM-Attention and IGDT Interactive Approach

Adopting electric energy to replace traditional coalfired and gas-fired heating methods is conducive to energy utilization and safety. In order to reduce the influence of the uncertainty of distribution and randomness output on the electric heating loads, an optimized electric heating load dispatch strategy is proposed to address this issue based on the interaction between the CNN-BiLSTM-Attention load forecast and the IGDT decision approach. Firstly, a CNN-BiLSTM-Attention model is employed to forecast distributed electric heating loads, integrating different load characteristics. Secondly, a dispatch strategy is proposed to maximize economic efficiency to ensure optimal thermal comfort for electric energy heating load users. Considering the uncertainty of the user behaviors, the IGDT decision model is introduced under the risky speculative strategy to find the maximized benefit. Then, through the data-driven forecast and decision-making interactions, an enhanced particle swarm algorithm is employed to derive the optimal dispatch results. Finally, the simulation strategy results illustrate that the proposed approach incorporates the characteristics of electric heating loads and improves the revenue of the load users.