Aiming at the shortcomings of traditional electricity theft detection method that only uses single classifier, an electricity theft detection model based on Stacking ensemble learning is proposed, and a MAP detection indicator is built based on actual project. First, the customer’s daily electricity consumption characteristics are disassembled in multiple dimensions according to time. And then Embedding is employed in selecting the most important features, which reduces the feature dimensions. Second, XGBoost, LightGBM, and CatBoost models are used to cross-validate and predict the results by Bayesian optimization. Finally, the base classifier of Stacking, whose integrated training with logistic regression, outputs the final prediction results. Taking 2016 CCF competition data as an example, the example analysis results verify the effectiveness and feasibility of the proposed stealing model.
