摘要
高比分布式光伏并网带来诸多问题,准确、可靠的光伏出力预测是提升高比分布式光伏渗透率配网调整和控制能力的基础。为此,该文从工程易用性出发,选择较为容易实现的极限学习机(Extreme Learning Machine,ELM)开展光伏出力预测,为缓解ELM权重和偏置随机性导致预测准确率不高的问题,该文使用麻雀智能算法优化ELM权重和偏置,试验结果表明,相较于原有的极限学习机,该文所使用的预测模型误差降低一半,有效提升光伏出力预测准确性。
The high aspect ratio distributed photovoltaic grid connection has brought about many problems.Accurate and reliable photovoltaic output prediction is the foundation for improving the penetration rate of high aspect ratio distributed photovoltaic distribution network adjustment and control capabilities.Therefore,starting from engineering usability,this paper selects the relatively easy to implement Extreme Learning Machine(ELM)to carry out photovoltaic output prediction.In order to alleviate the problem of low prediction accuracy caused by the randomness of ELM weights and biases,this paper uses Sparrow Intelligent Algorithm to optimize ELM weights and biases.The experimental results show that compared with the original ELM,the error of the prediction model used in this paper is reduced by half,effectively improving the accuracy of photovoltaic output prediction.
出处
《科技创新与应用》
2025年第25期58-61,共4页
Technology Innovation and Application
关键词
分布式光伏
配电网
极限学习机
麻雀智能算法
准确性
distributed photovoltaic
distribution network
Extreme Learning Machine(ELM)
Sparrow Search Algorithm(SSA)
accuracy
