| Power system maintains active power balance between generation and load by setting frequency modulation reserve.Traditionally,Frequency control reserve is mainly provided by the power generation side.With the increasingly complex power supply structure,the cost of frequency control reserve on the power generation side increases day by day.At the same time,the demand for frequency control reserve capacity increases significantly after large-scale renewable energy is connected to the grid.Therefore,on the basis of analyzing the characteristics of temperature control load,the research is carried out from two aspects: the polymerization modeling and clustering analysis of temperature control load,and the participation of temperature control load in the frequency modulation control strategy of the system.Firstly,physical models of air conditioning and electric water heater load are introduced from literatures and their validity is verified.Based on monte carlo method,the polymerization model of temperature controlled load was constructed,and the influence of different number of temperature controlled load on the polymerization model was considered.On this basis,K-Means clustering algorithm is used to carry out clustering and equivalent replacement of temperature control load,and the influence of characteristic parameters of temperature control load on clustering is analyzed,the error caused by equivalent replacement is calculated,and the feasibility of temperature control load clustering and equivalent replacement is verified by simulation.Secondly,considering the frequency events of the system may lead to heavy load or even over-limit of power of a branch of the distribution network.Therefore,with the constraint condition that the distribution lines do not overload and the optimization objective that the total network loss and voltage deviation are the minimum,an optimal scheduling strategy of temperature control load frequency modulation considering the distribution network constraints is proposed.Genetic tabu hybrid search algorithm is used to determine the position and capacity of N optimal nodes participating in frequency control.Simulation analysis based on IEEE33 node distribution network proves the effectiveness of optimal scheduling strategy.Thirdly,in view of the secondary frequency modulation problem of multi-group temperature control load participating in the system,a method is proposed to distinguish the working state and adjust the variation of load temperature setting value.The simulation shows that this method can solve the "secondary peak load" problem.On this basis,the time-sharing group frequency modulation control strategy is designed,and the decision-making model of different temperature control load groups participating in frequency modulation is established.The influence of different frequency modulation tasks,temperature control load groups and target sizes on the frequency modulation control strategy is analyzed by numerical examples,which provides strategic support for the temperature control load participating in the secondary frequency modulation of the system.a recovery control strategy considering the load limit controlled time is put forward,and the time and corresponding temperature change of each recovery set value of load groups participating in frequency modulation are determined with the constraint conditions of the load limit controlled time and the change of temperature set value in the process of frequency modulation.Finally,a distributed dynamic frequency modulation control strategy is proposed.This strategy reasonably controls the scattered load with the consideration of user comfort,and makes use of the dynamic relationship between the state change of the temperature control load and the trigger frequency to make the power of the temperature control load change with the change of the system frequency.Based on PSAT software platform,the effectiveness of control strategy is verified when different frequency events occur. |
PDF Full Text Request