Load Identification Method Based on Load Feature Extraction and Fuzzy C Mean Clustering Method

Load identification method can identify different types of loads in power systems. Adjustable loads have great potential in providing services to help consume the curtailed wind power. However, the consumption potential of the adjustable loads is difficult to be quantified. This paper makes an in-depth analysis of the necessity of extracting load characteristic parameters, then the characteristic parameters such as maximum adjustable speed, depth of adjustment, load fluctuation coefficient, maximum adjustable time, load time-shift coefficient are proposed. In order to improve the identification precision of adjustable load, the Daubechies fast wavelet transform is used to extract energy value as a new characteristic parameter. On this basis, fuzzy C mean clustering(FCM) is used to identify the load. Finally, through two examples of simulation and method comparisons in a certain area of Hexi, Gansu Province, the proposed identification method is verified that it can effectively improve the identification accuracy and precision of adjustable load.