光伏预测中的集成复杂度
Ensemble Complexity in Photovoltaic Forecasting
浏览论文内容
中文总结 AI 辅助
本研究通过匹配比较和消融实验评估光伏预测中集成组件的贡献，发现静态融合可降低误差但收益有限，天气门控无增量优势，并强调组件级评估与测试集重用限制的重要性。
中文摘要 AI 辅助
集成方法可以改善光伏预测，但同时也会增加贡献甚微或增加计算成本的组件。我们通过固定异构预测器库的匹配比较和消融实验来评估这些影响。每小时实验使用GEFCom2014和另外三个公开数据集，采用时间顺序划分和三个随机种子。在回顾性ERA5辅助下，静态融合相对于匹配的提升方法，在PVDAQ、OPSD和Ausgrid上将缩放平均绝对误差分别降低了1.11%、4.41%和1.63%；经过多重比较校正后，仅OPSD仍具有显著性。天气门控未提供一致的增量收益。探索性成员移除显示组级依赖性和个体冗余。一个单独的、先前检查过的十五分钟案例将一个神经成员替换为树预测器：归一化误差下降1.72%，但实测推理速度更慢。这些发现支持在明确限制天气可用性和测试集重用的情况下进行组件级评估。
英文摘要
An ensemble can improve photovoltaic forecasts while adding components that contribute little or increase computation. We assess these effects through matched comparisons and ablations of a fixed heterogeneous predictor bank. Hourly experiments use GEFCom2014 and three additional public datasets, with chronological partitions and three seeds. Under retrospective ERA5 assistance, static fusion reduces scaled mean absolute error against matched boosting by 1.11%, 4.41%, and 1.63% on PVDAQ, OPSD, and Ausgrid; only OPSD remains supported after multiple-comparison correction. Weather gating offers no consistent incremental benefit. Exploratory member removals show group-level dependence alongside individual redundancy. A separate, previously inspected fifteen-minute case replaces one neural member with a tree predictor: normalized error falls by 1.72%, but measured inference is slower. These findings support component-wise evaluation with explicit limits on weather availability and test-set reuse.
发表机构
- Huaibei Normal University（淮北师范大学）
机构由 AI 辅助整理，请以论文原文为准。
