Advanced control framework of regenerative electric heating with renewable energy based on multi-agent cooperation

Heating, ventilation, and air conditioning (HVAC) systems have accounted for a significant proportion of building energy growth. It is crucial to develop efficient HVAC systems based on renewable energy (RE) utilization to reduce energy costs (EC) and carbon emissions. However, HVAC system regulation is meet-ing a huge challenge owing to the randomness of RE and fluctuation of demand-side load. Aiming to tackle this problem, a multi-agent cooperative optimization control framework based on deep reinforce-ment learning was proposed to achieve the optimal match between the supply and demand sides of the regenerative electric heating system. Then, the dueling double deep Q-network and deep deterministic policy gradient method were applied to achieve the discrete and continuous control of independent agents, respectively, and value-decomposition network was exploited to realize the above agents' coop-erative optimization. Meanwhile, to further improve the stability, efficiency, and flexibility of the opti-mization control framework, the feasible action screening mechanism, prioritized experience replay mechanism, and regulation mechanism based on occupant behavior were introduced. According to sim-ulation results, in contrast to the baseline model, the reduction of thermal discomfort duration, uncon-sumed RE, and EC under the control of the multi-agent cooperative optimization framework can reach 84.86%, 70.79%, and 16.08%, respectively.(c) 2023 Elsevier B.V. All rights reserved.

Automated summary

In order to reduce energy costs and carbon emissions, HVAC systems need to be based on renewable energy utilization to improve efficiency. This study proposes a multi-agent cooperative optimization control framework based on deep reinforcement learning, which achieves control of the regenerative electric heating system by optimizing the match between supply and demand. By introducing feasible action screening mechanism, prioritized experience replay mechanism, and regulation mechanism based on occupant behavior, the stability, efficiency, and flexibility of the optimization control framework are further improved. Simulation results show that compared to the baseline model, the multi-agent cooperative optimization framework reduces thermal discomfort duration by 84.86%, unconsumed renewable energy by 70.79%, and energy costs by 16.08%.