Scholarlink

<p>Accurate photovoltaic (PV) power forecasting is essential for grid operation but remains difficult due to nonlinear multi-scale dynamics and seasonal distribution shifts. This work presents <b>MKAN-iTransformer</b>, a cascaded framework that integrates two existing components—the Multi-Scale Kolmogorov–Arnold Network (MKAN) for scale-aware temporal representation learning and iTransformer for variable-wise attention and inter-variable dependency modeling—under a <b>15-minute single-step</b> setting. Experiments on a real-world 30&#xa0;MW PV plant dataset from the Chinese State Grid Renewable Energy Generation Forecasting Competition use chronological splits within each season. MKAN-iTransformer achieves the best overall performance in <b>spring</b>, <b>autumn</b>, and <b>winter</b>. In spring, it reaches MSE=2.892, RMSE=1.701, MAE=0.864, and <InlineEquation ID="IEq1"> <EquationSource Format="TEX">\({R^{2}}=0.947\)</EquationSource> </InlineEquation>, improving over LSTM by 23.5%/12.5%/20.5% (MSE/RMSE/MAE). In autumn, it attains MSE=2.884, RMSE=1.698, MAE=0.774, and <InlineEquation ID="IEq2"> <EquationSource Format="TEX">\({R^{2}}=0.962\)</EquationSource> </InlineEquation>, reducing errors vs. iTransformer by 16.5%/8.7%/12.4%. In winter, it achieves MSE=1.721, RMSE=1.312, MAE=0.443, and <InlineEquation ID="IEq3"> <EquationSource Format="TEX">\({R^{2}}=0.969\)</EquationSource> </InlineEquation>, yielding 81.6%/57.1%/71.9% error reductions vs. Transformer. Ablation further confirms the complementarity between MKAN and iTransformer and shows that direct KAN integration can be unstable under winter shifts (KAN-iTransformer: MSE=7.082, <InlineEquation ID="IEq4"> <EquationSource Format="TEX">\({R^{2}}=0.872\)</EquationSource> </InlineEquation>).</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Interpretable ultra-short-term photovoltaic power forecasting with multi-scale temporal modeling and variable-wise attention

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年2月24日

查看全文
  • Linjie Liu,
  • Min Liu,
  • Zhuangchou Han,
  • HaiQiang Zhao

摘要

Accurate photovoltaic (PV) power forecasting is essential for grid operation but remains difficult due to nonlinear multi-scale dynamics and seasonal distribution shifts. This work presents MKAN-iTransformer, a cascaded framework that integrates two existing components—the Multi-Scale Kolmogorov–Arnold Network (MKAN) for scale-aware temporal representation learning and iTransformer for variable-wise attention and inter-variable dependency modeling—under a 15-minute single-step setting. Experiments on a real-world 30 MW PV plant dataset from the Chinese State Grid Renewable Energy Generation Forecasting Competition use chronological splits within each season. MKAN-iTransformer achieves the best overall performance in spring, autumn, and winter. In spring, it reaches MSE=2.892, RMSE=1.701, MAE=0.864, and \({R^{2}}=0.947\) , improving over LSTM by 23.5%/12.5%/20.5% (MSE/RMSE/MAE). In autumn, it attains MSE=2.884, RMSE=1.698, MAE=0.774, and \({R^{2}}=0.962\) , reducing errors vs. iTransformer by 16.5%/8.7%/12.4%. In winter, it achieves MSE=1.721, RMSE=1.312, MAE=0.443, and \({R^{2}}=0.969\) , yielding 81.6%/57.1%/71.9% error reductions vs. Transformer. Ablation further confirms the complementarity between MKAN and iTransformer and shows that direct KAN integration can be unstable under winter shifts (KAN-iTransformer: MSE=7.082, \({R^{2}}=0.872\) ).