Research On Distributed Combination Optimization Technology Of Demand Response Resource

With the power market reform and the release of demand side,the relevant technology of the demand side put forward new requirements.A large number of demand response resources can participate in the optimal operation of the power system,which increases the difficulty of the dispatching center.The load aggregator is the intermediary of the dispatching center and the demand response resource.It is responsible for integrating the demand response resources into the electricity market.The response resource combination optimization problem is a multi-time scale,multi-space scale and multi-objective problem,and it is very important to establish and develop the system framework and related technology for resource aggregation optimization.In this paper,the load aggregator is used as the main body,the hierarchical distributed system framework,the hierarchical aggregation of demand response resource,and the distribution calculation of the load adjustment quantity are established.The specific research contents are as follows:(1)This paper proposes a hierarchical and distributed system framework for resource response optimization.Demand response resources and load aggregators are the main research objects of the combinatorial optimization problem.Therefore,the demand response resource response characteristics,cost characteristics and uncertainties are analyzed and summarized,and the demand response resources in the actual demand are simplified and modeled.Participate in subsequent combinatorial optimization.Secondly,it analyzes the responsibilities and tasks of the load aggregators and builds up the architecture of the market,and expounds the role of the load aggregators as the intermediary of the market and demand response resources.This paper combs the relevant research contents of demand response resource combination optimization problem.Finally,according to the characteristics of demand response resources,the hierarchical distributed architecture of demand response resource composition optimization is established,and the main contents and methods of this paper are briefly introduced.(2)Research the equivalence model of demand response resource grouping.Demand response resources are spatially large,widely distributed and diverse.Therefore,the demand response resource grouping can reduce the dimension of demand response resource space and reduce the difficulty of direct scheduling.The response model of demand response resource grouping(called load agent)shows some response characteristics and cost characteristics.The response model of demand response resource in this paper firstly evaluates the maximum response capability of each group of demand response resources,That is,the maximum equivalent output,and then interpolate between the minimum and maximum equivalent output,and use the minimization cost model to obtain the corresponding output cost,and then use the least squares method to get the load agent cost characteristics.(3)This paper studies the distributed calculation of load agent load adjustment based on consistency algorithm.When the demand response resource group is equal to the load agent,the load aggregator needs to allocate the load adjustment amount to each load agent.Since there is some communication connection in each load agent,the load agent can allocate the load adjustment of the single time period by the distributed computing based on the consistency algorithm.Due to the climbing constraint of the load agent,the climbing constraint of the load agent is processed by the Dantzig-Wolfe decomposition theory(abbreviated as DWD),and we revised the consistency algorithm to reallocate the load adjustment.The current peak regulation schedule is usually a certain deviation,using consistency algorithm to solve the problem of intra-day deviation distribution.