Research On Aggregated Modeling And Control Strategy Of Constant Frequency Air Conditioning Load

Air conditioning loads account for a large proportion in China’s summer electricity peak load,having the characteristics of short duration throughout the year and costly peak regulation.The short time start-stop of air conditioners does not have a great impact on user’s comfort.Besides,air conditioning loads has the ability to participate in demand response quickly.Load-shedding of air conditioning by means of clusters has a significance in mitigating the pressure of peak regulation during peak periods,making up for the system power shortage caused by error of renewable energy output prediction.Since the majority of public air conditioners are constant frequency air conditioners,this paper regards constant frequency air conditioning load as the research object,conducting research in the following aspects:First of all,aiming at the problem that parameters differences in air-conditioning group are neglected in traditional aggregated model,grouping and clustering preprocessing is used to improve the simulative precision of the electric power.At the same time,the constraints of minimum outage time and integer setting temperature during operation of constant frequency air conditioners are taken into account,adding blocking layers to the state space model and eliminating the variable of setting temperature gradient.The simulation results show that improved aggregated model can simulate the electric power of air conditioning group with high accuracy,which can reflect the actual operation conditions of air conditioning group and provide the basis for follow-up participation in demand side response.Secondly,the system structure of power demand response in Jiangsu Province is summarized.For the peak clipping which is the main application,the reason of the severe load fluctuation caused by the traditional temperature regulation method after peak clipping period is analyzed in mechanism.On the basis of this,the improved peak clipping strategy is put forward.The example verifies this strategy can reduce the degree of load fluctuation after peak clipping period,which reduce the impact on power system.Based on the improved clipping strategy,the paper evaluates the response capability of controlled groups under different durations of control,and analyzes the factors influencing the response capability.Finally,combined with the load reduction capability of air conditioning groups,air conditioning loads are used as "virtual units" to make up for the lack of generating power due to the day-ahead wind power forecast errors.The paper proposes an intra-day bi-level rolling scheduling mode that includes the response of air conditioning loads.The upper level is managed by the energy management center to consider the scheduling cost of generation-consumption resources,and allocate the system power shortage to units and load aggregators.In the lower level,based on the reference power trajectory,the load aggregator adjusts the setting temperature of each air conditioning group during the regulation period through predictive control,so that total electric power of controlled groups track reference power trajectory.The simulation example shows that the introduction of demand side resources in intraday scheduling is equivalent to raise the modification upper limits of unit output plan,which is beneficial to balance intraday electric power,and reduce subsequent pressure of frequency regulation.It is analyzed that when the unit cost of load reduction is lower than the critical price,the load aggregator can allocate a considerable amount of load reduction in the bidding with the thermal power unit,which will help increase the revenue.When this condition is satisfied,the economy of the overall system operation is improved.