Research On Modeling And Optimization Of Ancillary Peak- Shaving And Ancillary Frequency Regulation With Demand Response In Electricity Market

Under the reform of the electricity market,most of the common resources on the demand side,such as air conditioners,refrigerators,and water heaters,have become the ideal system reserve due to their large number and the possibility of instantaneous cut-off,which can effectively participate in the power market through a variety of polymerization management methods,such as auxiliary peak shaving and auxiliary frequency modulation.In the demand side,the proportion of resident users is the largest,and the electricity consumption habits of different users are quite different.Therefore,multi-household users have greater potential for interactive scheduling optimization.For this reason,a model of demand response participating in auxiliary peak shaving based on household energy management is proposed in this paper.Firstly,it establishes the model of household photovoltaic power supply,energy storage system and load model respectively,and proposes a strategy of household energy management based on battery operation and photovoltaic self-use.Secondly,under the condition of satisfying the constraints of users and power grid,a multi-objective optimization model considering user comfort,economy and load curve optimization of power grid is established,which is solved by a multi-objective optimization algorithm based on Pareto tribe evolution.At the same time,the Nash equilibrium game is used to make multi-objective compromise decision after obtaining the Pareto curve of multiple objectives on the grid side and the user side.The simulation results of 100 residential user systems show that the proposed model and algorithm can effectively improve user comfort and economy,and help to peak load regulation of power grid.In order to model the demand-side resource participation in ancillary frequency regulation,this paper proposes a demand response model of generation-consumption coordinated frequency control in an islanded microgrid.By introducing load aggregator to aggregate largescale household users,the problem of dimension disaster in optimization is solved.The load aggregators can continuously assess their reserve capacity to participate in auxiliary frequency modulation according to the operation status of temperature control equipment in each household.At the same time,in order to meet the requirement of on-line optimization of auxiliary frequency modulation,this paper also proposes an ensemble learning algorithm based on collective intelligence,which is composed of multiple sub-optimizers and a learning concentrator.The sub-optimizer exerts the collective intelligence ability to provide exploration and development samples for the learning concentrator,while the learning concentrator based on reinforcement learning is mainly used for knowledge learning and transfer.The simulation example of islanded microgrid shows that the periodic and quality requirements of generationconsumption collaborative frequency control can be effectively met by on-line optimization of demand response participation and ensemble learning.