基于CEEMD-SSA-SVM的短期光伏发电功率预测OACSTPCD
Short-term Photovoltaic Power Prediction Based on CEEMD-SSA-SVM
针对光伏发电功率随机波动性导致预测难度大这一问题,采用改进的经验模态分解(CEEMD)对原始光伏发电功率数据进行分解,得到不同尺度的模态分量;然后引入麻雀搜索算法(SSA)对支持向量机(SVM)进行优化,建立不同尺度模态分量的预测模型;最后将各预测值叠加得到最终的光伏发电功率预测值.仿真结果表明,所提CEEMD-SSA-SVM光伏发电功率预测方法在保证原始光伏发电功率序列经CEEMD处理后具有较小重构误差的前提下,极大地提高了预测精准度.
In order to solve the problem that the prediction of photovoltaic power is difficult due to random fluctuation,an improved empirical mode decomposition(CEEMD)is proposed to analyze the original photovoltaic power data,and the modal components of different scales are obtained.Then,the sparrow search algorithm(SSA)is introduced to optimize the support vector machine(SVM),and a prediction model of the modal components of different scales is established.Finally,each predicted value is superimposed to obtain the final predicted value of photovoltaic power generation.The simulation results show that the proposed CEEMD-SSA-SVM method can greatly improve the prediction accuracy on the premise that the original photovoltaic power sequence has a small reconstruction error after CEEMD processing.
魏鹏飞;石新聪;朱咏明;何龙;李杨;巨晓敏;王清彬
国网新疆电力有限公司昌吉供电公司,新疆 昌吉 831100
动力与电气工程
光伏发电功率预测改进的经验模态分解麻雀搜索算法支持向量机
photovoltaic power generationpower predictionimproved empirical mode decompositionsparrow search algorithmsupport vector machine
《水力发电》 2024 (004)
87-94 / 8
国家自然科学基金资助项目(52266018);新疆维吾尔自治区重点研发项目(2022B01016-1)
评论