Research On Key Technologies Of Air Conditioning Loads Participating In Ancillary Service Of Power System

To deal with the challenges from the increasing penetration of renewables and continuous growth of electricity peak load,demand side resources have raised widespread attention due to their advantages of large quantity,low cost and fast response.The development of smart grid,advanced measurement and sensing as well as communication technologies,makes demand side resources participate in system ancillary services and correct the instability between generation and load possible.As one kind of typical flexible resources,air conditioning loads have prominent advantages in demand response: i)air conditioning loads account for a large portion of total load,and the regulation potential is enormous;ii)air conditioning loads possess thermal storage property and can be dispatched without sacrificing users’ comfort levels;iii)the peak electricity consumption period of air conditioning loads is exactly the period of system peak load,and the system power supply pressure can be relieved by air conditioning loads.Therefore,it is necessary to conduct research on air conditioning loads intensively.Based on the modeling and analysis of air conditioners,the dispatch and control strategies of air conditioning loads for various ancillary services,including peak shaving,frequency regulation and reserve allocation,are studied.The main research contents of this paper are as follows.(1)The basic models of different kinds of air conditioning loads are established,and the different control methods are analyzed and compared.Based on the working principles of fixed frequency air conditioners and inverter air conditioners,the basic models of the two kinds of air conditioning loads are studied in three perspectives: the thermodynamic model,electrothermal conversion model and power model.The control methods of different kinds of air conditioning loads are analyzed according to their operating characteristics,and the control modes for a population of air conditioning loads are summarized and compared.The modeling of air conditioning loads is the foundation for the following research on dispatch and control strategies for various system ancillary services.(2)The coordinated control strategies of fixed frequency air conditioning loads for peak shaving are proposed.The loss of load diversity problem caused by the cyclic operation of fixed frequency air conditioners is handled from the two aspects of on/off control method and setpoint shifting control method.Under the on/off control method,the response strategy based on a binary search algorithm and the recovery strategy based on the state recording are designed.Under the control method of setpoint shifting,an analytical model for the dynamic aggregated power and schedulable capacity of air conditioners is presented.On this basis,air conditioning loads are grouped and coordinately dispatched in the response and recovery stages,which not only avoids the power oscillations due to load diversity damage,but also satisfies the peak shaving target of power system.(3)The frequency regulation strategy of fixed frequency air conditioning loads based on the designed hierarchical and distributed control structure is presented.A population of air conditioning loads are constructed as a virtual generator,whose droop curve is similar to a conventional generator.In this manner,air conditioning loads can be scheduled and coordinated with the conventional generator.Then,a hierarchical and distributed control framework based on the central controller and local controllers is developed.To deal with the response power deviations caused by lock on and off constraints,a trigger value local update strategy of air conditioning loads is designed for reliable and fast frequency response on the basis of reducing communication burden.(4)The day-ahead and intra-day optimization approaches for the reserve allocation of transmission system considering various uncertainties and air conditioning loads are put forward.Faced with the uncertainties from loads,renewables and generating unit states,the generation resources and air conditioning resources are coordinated for the reserve demands of transmission system at different time scales.A day-ahead to intra-day reserve optimal scheduling framework is presented to accommodate the variations of prediction errors.Then,a day-ahead robust reserve optimal allocation model is developed considering continuous load/wind power and discrete unit state uncertainties,and an intra-day rolling stochastic optimization model is proposed to jointly schedule various reserve provisions in the generation side and load side,implementing the system supply-demand balance in the normal and uncertainty conditions.(5)The distributed coordination approach for robust reserve allocation of transmission and distribution systems is proposed,in which inverter air conditioning loads are incorporated.The aggregation of inverter air conditioning loads based on the virtual energy storage model is presented,and the reserve model of the aggregated virtual energy storage of inverter air conditioners is developed to achieve compatibility with the system reserve optimal allocation model,where the state-switch-feasibility constraints are considered.A distributed coordination framework is established to enable the synergistic but independent operations of the transmission and distribution networks,where only two kinds of boundary coupling variables,active power and reserve capacity,are exchanged for interregional energy share and reserve support.Under this framework,a distributed collaboration approach for robust reserve configuration optimization of transmission and distribution systems is proposed considering inverter air conditioning loads to facilitate the flexibility of dispatchable resources and enhance the overall reliability and cost-effectiveness.