Evaluation Of Thermostatically Controled Loads Demand Response Potential

To cope with the dual challenges of environmental pollution and shortage of fossil resources,the proportion of renewable energy generation in the power system is increasing rapidly,and the replacement of electric energy is promoted in many fields such as residential heating and transportation.It leads to a partial and periodic imbalance between supply and demand in the power system,which is characterized by frequent fluctuations,increase of the gap between the valley and peak load,and reserve capacity insufficient.Under the new energy structure,it is difficult to satisfy the reliable,economical,and efficient operation of the power system only depending on the output regulation of the controllable generator set.Thermostatically controlled loads(TCL)are important flexible resources in the power grid.Through demand response,they can suppress the random fluctuation of power flow,alleviate the contradiction between supply and demand,and improve the operating efficiency of the power system.On the other hand,the characteristics of TCL,such as large number of individuals,scattered location and rapid state change,make their management and development more difficult.To facilitate the demand side management organization to obtain the TCL response potential and optimize the response control and scheduling in realtime,the response potential evaluation method and control strategy of TCL are studied in this paper.The main contributions are given as follows.(1)Firstly,a TCL response potential dynamic evaluation method is proposed.A model parameter identification method and operation state estimation algorithm are proposed to settle the difficulty in obtaining information related to the response potential of a large amount of TCL.To evaluate the response potential of the TCL cluster from the response power and the duration,a response potential aggregation algorithm based on state grouping was proposed.A dynamic evaluation process of TCL response potential under the short time scale is designed to achieve accurate calculation at the users’ end and simplified estimation at the load management organization,and to eliminate the cumulative error through periodic synchronization.At the users’ end,the model parameters are solved and the response potential index is calculated by collecting real-time operating state data.In the load management organization,the operating state estimation and periodic synchronization are used to obtain the response capacity of each load unit,and the response capacity index of the cluster is calculated.The simulation results show that the evaluation results of the proposed method can characterize the demand response potential of TCL.(2)Then,a method to obtain TCL cluster status statistics is given.The temperature state of TCL changes rapidly,and the working mode and temperature control interval of different users and different types of equipment are different,so it is difficult to make overall statistics and analysis.Therefore,a general temperature state normalization index for cooling and heating equipment is established.To quantitatively describe the distribution of temperature state of TCL cluster,the expression of the probability density function of normalized temperature state of TCL was constructed,and the parameter estimation method based on partial sampling was given.The calculation methods of TCL cluster switching state and power statistics are given.Through the simulation example,the effectiveness of the proposed method is verified and the factors affecting the accuracy of parameter estimation are analyzed.In view of the shortcomings of the method,some improvements are made.(3)Finally,a unit selection model for TCL demand response is constructed.To reduce the communication and calculation pressure caused by TCL response control,the unit selection decision was made based on the cluster state distribution information,and the control instructions were issued by signal broadcasting in the model.Two unit selection algorithms are proposed.The sequential selection algorithm takes the temperature state as the response priority reference index and implements unit selection by issuing threshold temperature.The random selection algorithm takes into account the user’s wishes and implements unit selection by issuing a response index.Simulation analysis shows that the proposed model can simplify the communication process between aggregator and TCL unit,and the regulation accuracy is higher when applied to large-scale TCL clusters.The random selection algorithm can satisfy the difference of users’ willingness in the process of unit selection.