Integrated short-term power load prediction method, system and equipment based on signal decomposition and medium

The invention discloses an integrated short-term power load prediction method, system and device based on signal decomposition, and a medium, and the method comprises the steps: inputting a prediction result of a WPD-CatBoost combination model, a prediction result of a CEEMDAN-CatBoost combination model, a prediction result of a VMD-CatBoost combination model, a prediction result of an LMD-CatBoost combination model, and a prediction result of an ITD-CatBoost combination model into a trained SVM model; according to the method, the system, the equipment and the medium, the accuracy of power load prediction is relatively high. 本发明公开了一种基于信号分解的集成短期电力负荷预测方法、系统、设备及介质，包括：将WPD-CatBoost组合模型的预测结果、CEEMDAN-CatBoost组合模型的预测结果、VMD-CatBoost组合模型的预测结果、LMD-CatBoost组合模型的预测结果、ITD-CatBoost组合模型的预测结果输入训练后的SVM模型中，预测得到待预测时刻的集成电力负荷预测结果，该方法、系统、设备及介质对电力负荷预测的准确性较高。