Approach for prediction of cold loads considering electric vehicles during power system restoration

During the power system restoration there is a high possibility that the load demand is greater than the normal system operation for thermostatic loads. This demand increment, called Cold Load Pick-up (CLPU) has been a critical concern to utilities and will slow down the restoration process. Thus, it is imperative to predict the CLPU demand to perform restoration process more precisely. The use of new loads such as electric vehicles (EVs) has made new problems. In this paper, first the negative impact of EVs after an outage, on CLPU demand is investigated and then it is shown that the behavior of EVs during restoration service is the same as CLPU phenomenon. In doing so, the proposed approach uses Monte Carlo simulation method to predict EVs recharging demand. Since there will be some other new loads in the future that may have CLPU performance, a new approach is proposed to predict the CLPU demand of any load, without the need to have its exact model. ANFIS method is used to estimate CLPU demand based on previous outage cases in the feeder. The proposed approach is implemented in MATLAB® and simulation results confirm its ability in load prediction for restoration service.