SPECTRA：用于概率能源预测的状态空间外部上下文和时频分辨率架构
SPECTRA: State-Space Exogenous Context and Temporal-Frequency Resolution Architecture for Probabilistic Energy Forecasting
- Shanghai Advanced Research Institute, Chinese Academy of Sciences（中国科学院上海先进研究院）
- University of Chinese Academy of Sciences（中国科学院大学）
- School of Physical Science and Technology, Shanghaitech University（上海科技大学物理科学与技术学院）
- State Power Investment Corporation Limited（国家电力投资集团有限公司）
- China University of Petroleum（中国石油大学）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
针对现代电力系统概率能源预测问题，提出状态空间外部上下文和时频分辨率架构，通过自适应分离、对齐及建模等操作，实现预测。实验表明该方法在多种预测中表现出色，并支持确定性 - 随机分离为有效设计原则。
AI中文摘要：
现代电力系统在面对可再生能源间歇性、灵活需求、市场波动和天气相关发电等相互作用的不确定性时，越来越需要概率预测。然而，现有方法往往将多尺度分解、外部变量对齐和概率输出视为单独步骤，模糊了可预测结构和承载不确定性的波动如何共同塑造预测分布。本文提出了一种用于一般概率能源预测的状态空间外部上下文和时频分辨率架构。其核心前提是趋势周期分量主要决定基线轨迹，而高频残差和外部扰动控制预测不确定性的范围和不对称性。相应地，该架构自适应分离确定性和残差流，将外部上下文与两者对齐，通过多分辨率谱 - 时间状态空间建模细化确定性主干，并从其互补表示中估计有序分位数边界。在负荷、价格、太阳能和风能预测实验中，在18种设置中的14种中取得了最佳连续排名概率得分，比最强基线平均降低CRPS 5.74%，降低上尾分位数风险7.27%。这些结果支持确定性 - 随机分离作为一般概率能源预测的有效设计原则。
英文摘要：
Modern power systems increasingly require probabilistic forecasts amid interacting uncertainties from renewable intermittency, flexible demand, market volatility, and weather-dependent generation. However, existing methods often treat multi-scale decomposition, exogenous-variable alignment, and probabilistic output as separate steps, obscuring how predictable structures and uncertainty-bearing fluctuations jointly shape the forecast distribution. This paper proposes a state-space exogenous-context and temporal-frequency resolution architecture for general probabilistic energy forecasting. Its central premise is that trend-periodic components primarily determine the baseline trajectory, whereas high-frequency residuals and external perturbations govern the spread and asymmetry of forecast uncertainty. Accordingly, the architecture adaptively separates deterministic and residual streams, aligns exogenous context with both, refines the deterministic backbone through multi-resolution spectral-temporal state-space modeling, and estimates ordered quantile boundaries from their complementary representations. Experiments on load, price, solar, and wind forecasting achieve the best continuous ranked probability score in 14 of 18 settings, reducing average CRPS by 5.74\% and upper-tail quantile risk by 7.27\% over the strongest baselines. These results support deterministic-stochastic separation as an effective design principle for general probabilistic energy forecasting.
