A Hierarchical Dispatch Framework for Coordinated Energy Management of Electric Vehicles and Thermostatically Controlled Loads

To mitigate the grid challenges posed by the largescale integration of electric vehicles (EVs), In this paper, a hierarchical energy management scheme is put forward to utilize the demand response potentials of EVs and thermostatically controlled loads (TCLs).Under this framework, incentivized clusters of EVs and TCLs can rapidly follow the dispatch commands from a load aggregator, effectively minimizing the grid impact caused by numerous flexible loads. Leveraging load forecasts from a hybrid convolutional neural network (CNN) and long short-term memory (LSTM) model, the framework assumes a load aggregator manages and deploys flexible loads to minimize prediction errors. Based on the resulting dispatch strategy, peer-to-peer (P2P) power trading is conducted among participants, utilizing a distributed optimization method to maximize mutual benefits. Following the conclusion of P2P energy swaps, a sophisticated multi-objective optimization scheme is put in place. This scheme is designed to trim the fat off system costs, reduce the carbon footprint, and curtail unnecessary wind energy generation. A top-of-the-line Non-dominated Sorting Genetic Algorithm II (NSGA-II) is then deployed to crack the whip on this model and pinpoint the Pareto Front of solutions. To back the framework, a robust case study was conducted on the IEEE 30 -node power grid. The simulations show that our power dispatch strategy hits the nail on the head, providing the EVs and TCLs with the juice they need and bestowing considerable fiscal and environmental perks upon the power network.