Demand Response Dispatch Of Power Grid With Wind Power Based On Multi-agent System And Multi-time Scale

With the aggravation of pollution and energy crisis,wind energy and other clean energy are developing rapidly.However,because of the stochastic and intermittent characteristics of this kind of energy,great challenges are brought to the power grid,especially the large proportion of wind power resources in the new energy,has a great problem of wind power accommodation.With the development of smart grid,demand response technology has been concerned.The load resource can be dispatched through demand responses strategy,meet needs of the corresponding system,and become one of the most important means to solve the problem of wind power accommodation.Optimization of the grid dispatch based on demand response is analyzed.The effectiveness and feasibility of demand response to the wind power accommodation is proved through the cogeneration system.Then,dispatch different types of demand response to balance the different time scales of wind power prediction error.A demand response dispatch model based on multi-agent system is proposed,considering the difficulties of power grid demand response dispatch directly.And a demand response dispatch model based on multi-agent system considering multi-time scale wind power prediction error is constructed on grid.Specifically as follows:A strategy for wind power accommodation based on demand response of the heating cogeneration system is proposed.To meet the heat load demand,the demand response to the amount of wind power accommodation is improved through the pricebased demand response and direct load control.The effectiveness of demand response for wind power accommodation is proved.The output of wind power in different scenarios are analyzed.According to the characteristics of different time scales wind power prediction error,and the characteristics of price-based demand response and incentive-based demand response,the multi-time scale demand response dispatch model considering wind power prediction error is proposed.Because the day-ahead forecast error is large,the price-based demand response is used in the initial dispatch.In the day,the pricebased demand response is dispatched again to take advantage of its low cost.According to the strong timeliness of the incentive-based demand response,it is used in real-time dispatch to decrease the influence of wind power forecast error on the system dispatch and wind power accommodation.A demand response scheduling strategy based on multi-agent system is presented in the paper.The model of agents demand response reserve reporting based on game theory considering load possibility of breach and the model of capacity distribution of system dispatch based on the agents’ trust degree are built to realize the demand response resource distribution properly,dispatch cost saving and the highest income of agents on the condition of system constrains.Study case shows that,the model can find the equilibrium solution of the game around the multi agents to realize the purpose of demand response resource distribution to the agents properly.The effects of agents dispatch task transection on the grid dispatch and agents revenue are compared.And an optimal wind power accommodation dispatch model based on agent dispatch task transection in multi-agent system is proposed based on the multi-time scale demand response dispatch of power system considering wind power prediction error and the power system dispatch model based on game-theoretic multiagent system demand response scheduling capacity reporting.The effects of this model is analyzed through the modified PJM-5Bus system.