Energy Measurement Feature Extraction Based on Association Rule Mining in Integrated Energy System

Abstract

With the development of Integrated Energy System (IES) and renewable energy, the scenarios of energy measurement are more and more complex. Simulations based on single communication mode or single device cannot meet the needs of power grid. In order to build a systematic and large-scale energy measurement simulation system, it's important to study energy measurement feature extraction. To address this issue, an energy measurement feature extraction strategy based on association rule mining is proposed in this paper: Using support and confidence to evaluate the relevance between features, so as to find the suspected association in energy measurement features, and then eliminate redundant feature by nonlinear fitting and multiple correlation coefficient. The case results verify the correctness and effectiveness of the proposed method.