Deep Reinforcement Learning-based Low-carbon Economic Dispatch of Park Integrated Energy System
To reduce the park-integrated energy system's operating costs and carbon emissions while solving the random fluctuations caused by the system uncertainty,a low-carbon economic dispatch model of the park-integrated energy system considering ladder-type carbon trading is proposed and solved by the deep reinforcement learning method.Firstly,the ladder-type carbon trading model is proposed,and the low carbon economic dispatch problem of the park-integrated energy system is mathematically described by taking carbon trading costs into account;secondly,the dispatch problem is formulated as a Markov decision process framework,defining the observation state,dispatch action and reward function of the system;then the proximal policy optimization algorithm is used to make low carbon economic dispatch decisions.The proposed method does not need to predict load or model the uncertainty,and the network is trained to respond to the system state in real-time.Finally,the simulation is based on multiple scenarios and algorithms,and the results show that the proposed method improves the system operation economy while reducing the system's carbon emissions.
park integrated energy systemladder-type carbon tradingdeep reinforcement learningproximal policy optimizationlow-carbon economic dispatch
万方数据
维普
