Research On Demand Side Distributed User Response Model And Differentiated Demand Response Strategy

With the rapid development of society,the demand for energy is rising rapidly,but traditional energy will bring serious environmental pollution problems.Therefore,accelerating the development of renewable energy,gradually reducing traditional fossil energy,and forming a green and sustainable energy supply system have become the common consensus of all countries in the world.However,with the massive penetration of renewable energy,the flexibility of regulation on the supply side of the grid is rapidly decreasing and caused problems such as low utilization of grid capacity resources and waste of photovoltaic power and wind power.Therefore,there is an urgent need to exploit the adjustable potential on the demand side to form an energy supply system with interactive adjustment of supply and demand.Due to the limitation of load capacity,a large number of small and medium-sized users,mainly residential users,cannot directly participate in the centralized bidding and adjustment of the electricity wholesale market.They can only participate in the electricity wholesale market through electricity sales.Their total energy consumption is large and they have strong adjustable potential.However,the load characteristics and adjustable elasticity of users vary greatly,and are easily affected by external environmental factors.It is difficult to fully release their adjustable potential.Based on this,this paper studied and discussed the demand response guidance mechanism and key technologies for the distributed users on the demand side.The main research contents and conclusions are as follows:(1)Clarify the demand for load-side regulation of different types of demand response projects in the electricity wholesale market,and sort out the adjustment mechanism in demand response.First of all,from the perspective of the electricity wholesale market,the demand for accuracy and timeliness of load regulation in the energy market,the capacity market and the auxiliary service market are analyzed separately.Then through the combing of price-based and incentive-based demand response business mechanisms,the applicability of different adjustment mechanisms to distributed users on the demand side is analyzed.Finally,the key technologies are refined,and the advantages and disadvantages of different technologies are compared and analyzed.(2)Aiming at the diversified characteristics of distributed users on the demand side,build a robust user response behavior model to achieve accurate prediction of user load regulation behavior.First,the decision-making mechanism of user participation in demand response is analyzed from an economic perspective.Then,the main influencing factors of user load regulation decision-making are analyzed.The LSTM network is introduced to construct the user response behavior model.Finally,simulation experiments verified that the proposed model has good accuracy in predicting user group load regulation behavior,and at the same time,it has good performance under different demand response mechanisms.(3)In the initial stage of the electricity market,distributed users are unfamiliar with demand response services and have low acceptance.In response to this situation,designed incentive differentiation mechanism in demand response for demand side distributed users,and equal gradient iterative learning incentive strategy.This paper mathematically proved that under this mechanism,when the cost gradient of different user responses is the same,it has the lowest incentive cost.Based on this,an equal gradient iterative learning incentive strategy is constructed to optimize the amount of incentives given to different users in each demand response.Through simulation experiments,it is verified that the proposed strategy can guide users to accurately complete the demand response target,and at the same time effectively reduce the incentive cost of electricity sellers.(4)In order to fully explore the differences in user adjustment flexibility and further reduce incentive costs,settlement differentiation incentive mechanism and LSTM-PSO optimization strategy are designed.The concept of incentive factors is introduced to increase the user's adjustment revenue per unit with the increase of the response.Then,the LSTM-PSO optimization strategy is constructed to guide users to complete the demand response target with the lowest possible incentive cost.This paper mathematically proved the superiority of the proposed differential incentive mechanism,and at the same time verified that the proposed settlement differentiation incentive mechanism and LSTM-PSO optimization strategy can reduce the incentive cost to near the theoretical optimal value on the premise of ensuring the completion of demand response goals through simulation experiments.(5)Facing a large number of prosumer groups with dual attributes of load and power in the future,price differentiation mechanism in demand response considering photovoltaic consumption and edge cloud collaborative iterative optimization strategy are designed.By optimizing the aforementioned incentive factors,a more flexible electricity price settlement discount factor is formed,so that prosumer who make greater contribution to the supply-demand balance can obtain lower electricity price discounts.Based on the particle swarm optimization algorithm,the edge cloud collaborative optimization iterative algorithm is designed,and the fitness function is optimized to accelerate the convergence speed of the algorithm to save the computational cost of electricity sellers.Through simulation experiments,it is verified that the proposed mechanism and strategy can effectively reduce the unbalanced load of the prosumer group and improve the efficiency of the electricity sellers and prosumer group.In summary,this paper studied the implementation methods and key technologies of electricity sellers aggregate demand-side distributed users to participate in demand response projects in the wholesale electricity market.Accurate prediction of user load regulation behavior is achieved by establishing user response behavior models,and the short-term,medium-term and long-term incentive mechanisms and optimization strategies for demand response are designed.It is expected that the research in this paper can provide a reference for the implementation of demand side distributed users participating in demand response adjustment in the electricity wholesale market.