Evaluation Of Electric Power Demand Response Potential And Its Application

Because traditional power grid is faced with sustainable development bottlenecks,smart grid gradually leads a boom to new round of research and application.Electric power demand response(DR)is an important means to realize the smart grid,whose potential characterizes the available margin of DR.Expanding demand response potential,not only to ease the power grid operating pressure,reduce system operating costs,but also effectively consume intermittent energy,benefit to energy saving.So,quantitative assessment research of demand response potential can help to guide the development of scientific and reasonable price and incentive policies,which has urgent and practical significance in the situation of power system reform and stimulating users to interact with power grid.However,DR’s randomness and uncertainty are greatly increasing the difficulty of potential assessment.And in the applications of DR,a series of technical problems caused by its randomness and uncertainty are needed to overcome.Therefore,the main works are as follows:1)Demand response potential quantitative evaluation approach considering load statistical characteristics is proposed.Demand price-elasticity coefficients are achieved based on the principle of econometrics and power demand is solved under the tariff and incentive policy.Combined with historical statistics,load characteristic statistical models are established.Besides,demand response range of industrial,commercial and residential customers are respectively analyzed.The intersection of the above two ranges forms the demand response envelope to obtain the quantified DR potential.Based on the data of the 2014 electric Statistical Yearbook,amount of DR in Ningxia Autonomous Region and Zhejiang Province was calculated,and also the DR potential of Ningxia Autonomous Region was evaluated.2)Because DR is a random behavior,so in the DR application to power system,quantitative analysis for probability of demand response and factors influencing DR probability is done.On the basis of studying on household load response model,demand response probability optimization model targeting at maximum household load’s DR engagement is established.When the number of users participating in demand response is sufficient enough,the proposed probability optimization model is equivalent to a series of 0-1 integer programming problems,which greatly reduces the difficulties of solving problem.Physical influence factors on DR probability are quantified,and the influence trend and extent is analyzed.The feasibility and correctness of the model and algorithm are verified by numerical case study with 6000 households in the residential area.3)The robust optimal scheduling model of wind power system with consideration of the uncertainty of demand response is established and the solution is given.The model is established targeting at minimum system total cost,which expresse the uncertainties of DR power and wind power forecast value as the form of uncertain sets.Based on the Benders decomposition algorithm,the two-stage problem of the model is decomposed into main problem and sub-problem.The main problem solves the set of 0-1 discrete variables and the sub-problem solves the set of continuous variables.The optimal solution of model is obtained by iteratively solving main problem and sub-problem which exchange the best solution information.The results of an example with 10 conventional power units,2 wind power plants and 3 DR nodes show that DR applications can reduce power system total costs and abandoned wind power.