基于GBDT和SVM的光伏发电出力预测研究

梯度提升决策树（GBDT）和支持向量机（SVM）是预测光伏出力的两种常用算法。分析了光伏发电出力的影响因素，介绍了GBDT算法和SVM算法的原理，以及基于两种算法的光伏出力预测模型的流程，并进行对比分析。实验结果表明，基于GBDT算法的光伏出力预测模型的平均绝对相对误差和均方根误差都较小，预测效果更好。

Research on Photovoltaic Output Prediction Based on GBDT and SVM

Gradient boosting decision tree(GBDT) and support vector machine(SVM) are two common algorithms for forecasting photovoltaic output.This paper analyzes the influencing factors of photovoltaic power generation output, introduces the GBDT algorithm and SVM algorithm mechanism and the flow of photovaltaic output prediction model based on these two algorithms and makes comparative analysis.The test results show that the average absolute relative error and root mean square error of gradient boosting decision tree algorithm are smaller, and the forecasting effect is better.