Real-time Pricing Of Smart Grid Based On Demand Side Management

With the world energy crisis and environmental pollution becoming more and more serious,smart grids and energy Internet of Things have developed rapidly due to their advantages in energy saving and emission reduction.As an important mechanism of the smart grid,the purpose of demand side management(DSM)is to reasonably optimize the allocation of electricity supply and demand.It can guide the demand side to adjust the electricity consumption pattern,and use energy scientifically and rationally in order to reduce waste.Ultimately,the installation of new production and distribution facilities is avoided.Demand response(DR)is one of the important means of DSM,including pricebased DR and incentive-based DR.The real-time pricing(RTP)based on DR is an ideal pricing mechanism for smart grid,which can maintain the balance of electricity between the supply and demand sides and promote the energy dispatch behavior of DSM.Therefore,it is very significant for accelerating the establishment of the RTP mechanism to guide electricity production,promote energy conservation,emission reduction,and sustainable social development.This paper studies complex smart grids involving renewable energy,storage equipment,and multiple loads,etc.From the perspective of deterministic problems and stochastic programming,different models have been established using single level optimization and bilevel programming for DSM.The specific pricing algorithm for the RTP has been obtained.Through analysis and comparison of simulation cases,the rationality of the model and the effectiveness of the algorithm are fully verified.The main work of this paper is summarized as follows:1.For a smart grid with multiple microgrids connected to the grid,including renewable energy,storage equipment,multiple loads,etc.,a bilevel programming model is established according to the master-slave hierarchical relationship between the supplier and users.By analyzing the characteristics of the model,the existence of the optimal solution is proved,and a distributed hybrid genetic-branch and bound algorithm(GABBA)is designed to solve the model.This method not only formulates the specific algorithm of the RTP,but also obtains the optimal energy dispatching strategy within each microgrid.At the same time,the algorithm realizes parallel computing to improve computing efficiency and avoid the disclosure of user’s utility privacy.Compared with the fixed electricity price,the numerical results show that the RTP can effectively reduce the peak-to-average ratio and peak-to-valley difference of consumption,and avoid the installation of new generation equipment.In addition,it also improves the social welfare of the entire system.2.In the novel smart grid environment,the RTP is studied for regional integrated energy systems,including electricity,natural gas,solar,wind and other energy sources.A bilevel programming model is established and transformed into a single level mixed integer programming model using the Karush-Kuhn-Tucker conditions.Then,the step of the RTP is obtained,at the same time,the optimal energy dispatch and consumption strategy of regional integrated energy system(RIES)are designed based on the integrated DR for electrical and thermal loads,according to the mutual conversion and supplement between multiple energy sources.Comparing three different demand response scenarios,the results show that the integrated DR not only effectively reduces carbon emissions,but also significantly improves the total social welfare.3.Considering the volatility of renewable energy power generation and the uncertainty of electricity load,a stochastic bilevel model is established for smart grid that coexist with traditional energy,photovoltaic,and wind power.Using mathematical skills transform it into a deterministic model,and a distributed hybrid particle swarm optimization-branch and bound algorithm(PSO-BBA)is proposed by combining classic and intelligent algorithms.Furthermore,the RTP schemes and algorithms are obtained.Numerical results are compared in three ways for a actual cases in China.First,the results and speed are compared according to the interval division of different time periods.Second,compared with the deterministic model,although some social welfare is damaged,it can maintain the stability of the grid to avoid large-scale blackouts.Finally,compared with the operation cost minimization model,it shows that considering the user’s utility is more practical in the proposed social welfare maximization model.4.The RTP issue is studied based on DR with large-scale wind power participation.According to the inherent uncertainty of wind power,a stochastic programming model is established with chance constraints based on the maximization of social welfare.The stochastic programming model is transformed into a deterministic convex optimization by probability method.Then the model under single time period is studied due to the independence between multiple time slots.In order to protect users’ privacy and reduce large-scale computation,the consensus-based distributed alternating direction multiplier method(ADMM)is used to solve the model,then the shadow price of the RTP is obtained.Simulation results show that large-scale wind power participation in DR is conducive to the adjustment and supplement of traditional energy generation in a timely manner.At the same time,it can reduce energy costs and the wind abandonment rate.