Electricity User Schedulable Load Classification Method Based on MFNN Algorithm

Abstract

To solve the problems of users' electricity consumption analysis under the background of smart grid, this paper proposes a schedulable load classification method based on Multi-Layer Feed-Forward Neural Networks (MFNN) to classify the schedulable load curve-data for the resident users. Firstly, this paper selected the optimal feature set of the load curve-data by the feature extraction strategy, in which the extracted features are used as the input parameters in MFNN. Secondly, it depicted the proposed classification method including the related MFNN training operation and the implementation course to achieve the classification results of the schedulable load curve-data, which is effective to analyze the user's electricity behavior. Using the Irish electricity user curve-data as the data source, the experiment results illustrate that the proposed MFNN-based method may achieve better classification performance to find the pattern mode of user's electricity consumption status.