期刊文献+

基于极端学习机的光伏发电功率短期预测 被引量:3

Short-Term Forecasting of PV Capacity Based on Extreme Learning Machine in Similar Days
下载PDF
导出
摘要 为了进一步提高光伏发电功率的预测准确度,该文首次将极端学习机方法(ELM)和相似日方法结合并引入光伏发电功率短期预测领域。通过分析影响光伏发电功率的各个因素,分时段预测光伏发电功率。该方法在不同时间段中利用相似日评价函数选取历史相似日,结合预测日的天气因素,采用极端学习机对预测日对应时段的发电功率进行预测。通过对预测效果进行比较和分析,结果表明该方法比传统的神经网络预测算法有更好的预测效果。 To improve the prediction accuracy of PV capacity, the extreme learning machine (ELM) method combined with the method of similar days is introduced into the domain of short-term forecasting of PV capacity for the first time. This paper analyzes various factors which may influence the photovoltaic capacity, and forecasts photovoltaic capacity according to different periods. The evaluation function is used to choose similar days from history data based on different periods of the predicted day. This method employs the similar day evaluation function to select history similar day in different periods, eombined with the weather of the forecasting day, then utilizes extreme learning machine for predicting the power output of the corresponding periods in forecasting day. According to comparison and analysis, the results show that the proposed method is better than the method with general version of neural network.
出处 《控制工程》 CSCD 北大核心 2013年第3期386-390,共5页 Control Engineering of China
基金 浙江省科技厅重大专项重点工业项目(2009C11020) 国家自然科学基金项目(51007015)
关键词 光伏发电系统 相似日 极端学习机 发电功率预测 photovoltaic system similar day extreme learning machine generated power forecasting
  • 相关文献

同被引文献29

引证文献3

二级引证文献20

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部