Study On Demand Response And Optimization Interaction Of Large Scale Air Conditioning Load Cluster

At present,the demand response is a research hotspot in the direction of the power system development.The air conditioning(AC)load is one of the most typical and abundant resources on the demand side,especially large-scale AC load cluster.During summer peak hours,sometimes AC loads can reach 40%of the total electricity load and keep growing.The response speed of AC load is fast,and adopting some control strategies(such as cycle control according to duty cycle)will not cause the load curve fluctuate obviously,so the distributed AC load has become the most important demand response resource in the smart grid.However,at present,there is still no detailed theoretical research method for large scale distributed AC load participating in peak shaving and interaction with distributed energy in active distribution network.This paper relies on the science and technology project of the State Grid,"study on simulation technology of automated demand response under China's electricity sales side opening(E16L00640)".Taking the distribution AC load as the research object,the aggregation model of AC load is established;Demand response of AC load participating in the system peak regulation is studied;Optimization interaction of large-scale AC load cluster and the distributed energy is studied.The main contents are as follows:Firstly,based on the equivalent thermal parameters(ETP)method,individual air conditioning load model is established,thus launches air conditioning room virtual storage model according to the heat storage capacity of the AC room;Based on the Monte Carlo sampling method,the simulation model of the large scale air conditioning load cluster is established;Based on human comfort,the constraint of large scale AC load aggregation model is proposed,which lays a theoretical foundation for subsequent demand response and optimization interaction.Secondly,the large scale AC load cluster participating in the system load shedding is studied.Considering the load shedding amount and its duration during the system peak period,a new demand response ability index is proposed.The optimization model of large scale AC load cluster to participate in the system demand response is established in MATLAB/OpenDSS.Based on the improved discrete particle swarm optimization algorithm,the maximum potential of system peak shedding is evaluated and analyzed.And the influence of different temperature conditions on demand response potential is analyzed through the example.For control strategy of each AC,the temperature sorting method is proposed and the temperature sorting method is improved based on the minimum number of AC switches.The optimal control strategy for each AC is obtained through the example,and the result of the example is analyzed.Finally,the optimization interaction of large scale AC load cluster and distributed energy in active distribution network is studied.The main evaluation indexes of active distribution network are analyzed.Considering these indexes,the multi-objective optimization model,with the minimum power cost and minimum network loss as the goal,is proposed.The optimal dispatching model of active distribution network with large scale AC load cluster and distributed energy is established in the MATLAB/OpenDS S simulation platform.Based on the improved particle swarm optimization algorithm,the improved IEEE-37 node distribution network is taken as an example.We can learn that,by coordinating the dispatching of air-conditioning load and distributed energy storage system,that the large scale air conditioning load in the active distribution network scheduling model has great scheduling potential is verified;The peak-to-valley difference and the line loss are greatly reduced;The user satisfaction is improved;The safety and economy of the active distribution network are improved.