Research On Adjustable Capability Characterization And Response Strategy Of Flexible Load Groups For Power Peak Regulation

With the promotion of electric energy substitution strategy,the power load rises year after year,but the annual duration of peak load is very low(according to statistics,in a province is less than 80 hours),which brings great challenges to the safe and economic operation of power system.Excavating flexible load adjustable resources such as temperature control load,electric vehicle and household energy storage to participate in power peak regulation can delay power investment.It is an effective way to reduce the peak difference of power system and improve the safety and economy of power system operation.The adjustable characteristics of flexible load are related to the characteristics of load equipment,energy consumption habits of users,temperature and humidity environment and other factors.Mining its adjustable ability faces many challenges.Facing the demand of power peak regulation,aiming at the problem of the adjustable capacity characterization and the availability of adjustable power of flexible load group,this paper carries out the calculate modelings of adjustable power,the analysis of adjustable characteristics and the evaluation of user load adjustable potential to quantify the adjustable capacity of load resources,designs the market incentive mechanism and response strategy to realize the economic transfer of flexible load resources,carries out the research of control system to solve the key technical problems of application.The purpose is to transform potential load resources into real adjustable power that can participate in the response of power system,and to ensure the safe and economical operation of the new power system.The innovative work of this study is as follows.(1)Three flexible loads aggregation adjustable power calculation models are constructed.The working principle and power adjusted mode of three flexible loads,namely split fixedfrequency air conditioner,electric vehicle and energy storage,are studied in this paper.The aggregate power model of air conditioning load adjusted under orderly temperature control mode is constructed.It overcomes the disorder fluctuation of the total adjustable power caused by the diversity of the synchronous adjustment initial state of the temperature control load group,and improves the stability of the total adjustable power.Two air conditioning load adjustable power calculation models are constructed based on the equivalent thermal parameter model and BP neural network.BP neural network avoids the computational difficulties caused by the difficulty of accurate real-time identification of physical model parameters.At the same time,the adjustable power calculation models of electric vehicle load group based on Monte Carlo and energy storage load group based on area method are constructed.On this basis,this study constructs an analysis index of the adjustable characteristics of flexible load aggregation,and sets scenarios for the above three loads homogeneous and heterogeneous aggregation groups to calculate and analyze the adjustable characteristics under different adjustable modes respectively.The results of this study show that the flexible load aggregation has stable and flexible adjustable capability,and the heterogeneous load aggregation has obvious adjustable superiority over the homogeneous load aggregation.(2)A method to evaluate the adjustable potential of customer loads based on electricity consumption pattern identification is proposed.First,we use K-Means clustering algorithm to extract customer consumption patterns based on historical customer load data.Then we propose a method to calculate the adjustable potential of customer load considering the feasibility of switching between different electricity consumption patterns of customers.This method does not need to model the user load equipment and the influencing factors of the user's power consumption behavior,and avoids the difficult problems of obtaining the operating state and environmental parameters of the load equipment.The evaluation results can provide auxiliary decisions for flexible load control for customers,load aggregates and grid operators.(3)Response strategries for the participation of flexible loads in power system peak regulation is established.This paper constructs a three-layer dispatching and control architecture for flexible load aggregation considering load aggregators as the linkage.The quantity-price response strategry optimization model of customer declaration considering satisfaction and bidding success rate,the negotiated bidding optimization model of load aggregation considering customer benefits and bidding success rate,and the power peaking auxiliary services market clearing model of grid operators considering source-load cooperation are proposed respectively,and the solution algorithms of the three models are given.In this paper,the auxiliary service market of power peak regulation is taken as an example to simulate.The results verify the feasibility of the auxiliary service market mechanism of power peak regulation with the participation of flexible loads.This mechanism can stimulate the adjustment potential of flexible load with the price lever,and turn it into a real adjustable resource,so as to realize the optimal dispatching of flexible load resources.(4)A flexible load aggregation control system based on cloud edge collaboration is designed and developed.Aiming at the control problems of flexible loads,such as wide range of points and large scale,we designed a cloud-edge collaborative system architecture based on cloud platform and edge terminal.On the cloud side,we propose a performance optimization algorithm to minimize the task completion time and execution resource consumption of the cloud platform so as to improve the concurrent computing efficiency of massive tasks regulated by flexible load aggregation.In addition,we also set up an automatic power control model of flexible load to track the target power curve on the edge side,and realize accurate execution of flexible load group control instructions.The effectiveness and superiority of the developed system are verified by simulation and engineering cases.The system can realize user flexible load information collection and closed-loop control on the edge side,and realize strategy calculation and global optimization of flexible load resource call on the cloud side.