Load Forecasting Based on Data Mining and Improved Stacking Ensemble Learning Under Load Aggregator

In the process of demand response, in order to manage resources in the load side better and solve the problems of large amount of data, unclear data characteristics and many invalid data, the load aggregator proposed a load forecasting model based on data mining and improved stacking ensemble learning. Firstly, analyze the energy consumption behavior of loads agented by load aggregators, and models for various loads participating in demand response are established; Then, the data processing feature engineering based on data mining model is established to extract the features of the original data and form the load forecasting feature data set; Finally, through improving stacking ensemble learning model, all kinds of loads under the load aggregator are predicted. To verify the effectiveness of the model, the article conducts experiments using real load data from a certain location. And compare it with other algorithms. The experiment shows that the model proposed in the article improves the prediction speed and accuracy, providing a reliable basis for load aggregators to participate in the demand response market.