Short-term district power load self-prediction based on improved XGBoost model

Distributed generation and diversified loads increase the uncertainty of district power prediction. Useful prediction requires a highly accurate model, and there are several challenges facing the designers of a new power system with intelligent power distribution. To solve them, we improved an XGBoost model from three aspects: model, data, and performance. This paper proposes an XGBoost model with a windowed mechanism and random grid search (WR-XGBoost model) for self-prediction of short-term district power load. Specifically, a causal sliding window with different strides is introduced into the model optimization stage to process the training and test sets separately. In data optimization, the model initially processes the data organized in forms and then uses discrete difference data as input. Finally, in optimizing the performance, a random grid search method reduces the debugging of hyperparameters. The results show that the WR-XGBoost model outperforms five comparison models in terms of predictive power and generalization, using four datasets and seven statistical indicators.

Automated summary

This paper addresses the uncertainty of district power prediction caused by distributed generation and diversified loads. By improving the XGBoost model from three aspects: model, data, and performance, a WR-XGBoost model with a windowed mechanism and random grid search is proposed for self-prediction of short-term district power load. The model introduces a causal sliding window with different strides to process the training and test sets separately, uses discrete difference data as input, and reduces the hyperparameter debugging process through random grid search. The results demonstrate that the WR-XGBoost model outperforms five comparison models in terms of predictive power and generalization using four datasets and seven statistical indicators.