Answers are grounded in this paper and cite their sources.
This model improves forecasting accuracy of hybrid renewable energy, highlighting nonlinear coupling features.
Accurate forecasting of hybrid renewable energy outputs is critical for reliable and efficient power grid operations. Existing forecasting models often overlook the nonlinear coupling characteristics inherent in hybrid renewable systems, limiting their predictive accuracy. To address this issue, this paper proposes an enhanced forecasting approach that extends the Kolmogorov-Arnold Network (KAN) by introducing Chebyshev polynomial basis functions, combined with Bidirectional Gated Recurrent Units (BiGRU). Unlike conventional KAN approaches employing B-spline basis functions, the proposed Chebyshev-KAN adapts the existing framework to explicitly capture the complex nonlinear interactions within hybrid renewable generation, providing improved numerical stability and global approximation capability. Additionally, the BiGRU module models temporal dependencies effectively, enhancing the coupling representation between solar and wind energy generation. Comparative experiments on real-world hybrid renewable datasets demonstrate that the proposed Chebyshev-KAN+BiGRU model outperforms traditional benchmarks, including B-spline KAN+CNN, Transformer, and BiGRU models, particularly under rapidly fluctuating environmental conditions.
No takes yet. Share an insight, caveat, or question.
Tang et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: