Learning Long-Term Temporal Dependencies in Photovoltaic Power Output Prediction Through Multi-Horizon Forecasting

• URL: http://arxiv.org/abs/2605.19074v1
• Date: Mon, 18 May 2026 19:56:34 GMT
• Title: Learning Long-Term Temporal Dependencies in Photovoltaic Power Output Prediction Through Multi-Horizon Forecasting
• Abstract summary: This paper proposes a transition from traditional single-horizon estimation toward a multi-horizon forecasting framework.<n>We experimentally demonstrate that joint optimization over a sequence of future values allows deep neural networks to better capture latent inter-step temporal dependencies.<n>We evaluate the models' abilities to predict power output across multiple discrete future time steps simultaneously.
• Score: 9.023507281431252
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: The rapid global expansion of solar photovoltaic (PV) capacity-reaching a record 597 GW in 2024-highlights the urgent need for robust forecasting models to mitigate the grid instability caused by the intermittent nature of solar irradiance. While deep learning-based direct forecasting using ground-based sky images (GSI) has emerged as a dominant approach, existing literature is often constrained by single-architecture evaluations and an exclusive focus on single-horizon (point) prediction. This paper proposes a transition from traditional single-horizon estimation toward a multi-horizon forecasting framework, leading to an architecture-independent improvement in accuracy. We hypothesize and demonstrate experimentally that joint optimization over a sequence of future values allows deep neural networks to better capture latent inter-step temporal dependencies by avoiding precocious convergence of the network in terms of both weight gradients and filter diversity. Leveraging this architecture-independent improvement that integrates sequential sky imagery with historical PV generation data, we evaluate the models' abilities to predict power output across multiple discrete future time steps simultaneously. Our methodology is validated through a comparative analysis across diverse deep learning architectures. The results demonstrate that this multi-horizon approach significantly enhances predictive accuracy and robustness across the entire forecast horizon while maintaining computational parsimony. By achieving superior performance with negligible overhead compared to single-horizon models, this work provides a scalable and efficient solution to improve the resilience of modern power grids.

Related papers

• FarSky: Task-Aware Latent-Space Coupling for Generative Intra-Hour Solar Forecasting [0.0]
  FarSky is a generative forecasting framework that learns task-aware representations of sky images.<n>It achieves the best overall deterministic and probabilistic forecasting performance, improving forecast skill by up to 11 percentage points.<n>It substantially improves ramp event detection over existing methods, achieving F1-scores above 60%.
  arXiv  Detail & Related papers  (2026-08-06T08:32:56Z)
• Inpainting-Style Conditional Diffusion for Multivariable Time Series Forecasting [15.992262494632321]
  We propose a conditional diffusion-based framework for multitemporal time-series solar forecasting power.<n>The proposed method reformulates temporal PV data as structured two-dimensional representations using a sliding-window patch construction.<n>Experiments conducted on benchmark PV dataset, including GEFCom2014, demonstrate that the approach achieves high forecasting accuracy.
  arXiv  Detail & Related papers  (2026-05-27T11:25:28Z)
• LeapTS: Rethinking Time Series Forecasting as Adaptive Multi-Horizon Scheduling [74.94985663101906]
  We present LeapTS, a novel framework that reformulates time series forecasting as a dynamic scheduling process over the prediction horizon.<n>We show that LeapTS improves overall forecasting performance by at least 7.4% while achieving a 2.6$times$ to 5.3$times$ inference speedup over representative Transformer-based models.
  arXiv  Detail & Related papers  (2026-05-11T09:54:02Z)
• ThinkJEPA: Empowering Latent World Models with Large Vision-Language Reasoning Model [53.15040805435013]
  Vision-language models (VLMs) provide strong semantic grounding and general knowledge by reasoning over uniformly sampled frames.<n>We propose a VLM-guided JEPA-style latent world modeling framework that combines dense-frame dynamics modeling with long-horizon semantic guidance.
  arXiv  Detail & Related papers  (2026-03-23T17:59:42Z)
• Insight-V++: Towards Advanced Long-Chain Visual Reasoning with Multimodal Large Language Models [65.4947731385794]
  We present a unified multi-agent visual reasoning framework that evolves from our foundational image-centric model, Insight-V.<n>We introduce two novel algorithms, ST-GRPO and J-GRPO, which enhance spatial-temporal reasoning and improve evaluative robustness.<n>Experiments on base models like LLaVA-NeXT and Qwen2.5-VL demonstrate significant performance gains across challenging image and video reasoning benchmarks.
  arXiv  Detail & Related papers  (2026-03-18T15:28:07Z)
• FaST: Efficient and Effective Long-Horizon Forecasting for Large-Scale Spatial-Temporal Graphs via Mixture-of-Experts [49.9321870703948]
  Existing models predominantly focus on short-horizon predictions and suffer from notorious computational costs and memory consumption.<n>We present FaST, an effective and efficient framework based on Mixture-of-Experts (MoEs) for long-horizon and large-scale STG forecasting.<n>FaST is underpinned by two key innovations. First, an adaptive graph agent attention mechanism is proposed to alleviate the computational burden.<n>Second, we propose a new parallel MoE module that replaces traditional feed-forward networks with Gated Linear Units (GLUs)
  arXiv  Detail & Related papers  (2026-01-08T18:00:58Z)
• Ultra-short-term solar power forecasting by deep learning and data reconstruction [60.200987006598524]
  We propose a deep-learning based ultra-short-term solar power prediction with data reconstruction.<n>We employ deep-learning models to capture long- and short-term dependencies towards the target prediction period.
  arXiv  Detail & Related papers  (2025-09-21T14:22:35Z)
• Enhanced Photovoltaic Power Forecasting: An iTransformer and LSTM-Based Model Integrating Temporal and Covariate Interactions [16.705621552594643]
  Existing models often struggle with capturing the complex relationships between target variables and covariates.<n>We propose a novel model architecture that leverages the iTransformer for feature extraction from target variables.<n>A cross-attention mechanism is integrated to fuse the outputs of both models, followed by a Kolmogorov-Arnold network mapping.<n>Results demonstrate that the proposed model effectively capture seasonal variations in PV power generation and improve forecasting accuracy.
  arXiv  Detail & Related papers  (2024-12-03T09:16:13Z)
• Learning Global and Local Features of Power Load Series Through Transformer and 2D-CNN: An Image-based Multi-step Forecasting Approach Incorporating Phase Space Reconstruction [1.9458156037869137]
  This study proposes a novel multi-step forecasting approach by delicately integrating the PSR with neural networks to establish an end-to-end learning system. A novel deep learning model, namely PSR-GALIEN, is designed, in which the Transformer and 2D-CNN are employed for the extraction of the global and local patterns in the image. The results show that, compared with six state-of-the-art deep learning models, the forecasting performance of PSR-GALIEN consistently surpasses these baselines.
  arXiv  Detail & Related papers  (2024-07-16T09:59:13Z)
• Long-term Wind Power Forecasting with Hierarchical Spatial-Temporal Transformer [112.12271800369741]
  Wind power is attracting increasing attention around the world due to its renewable, pollution-free, and other advantages. Accurate wind power forecasting (WPF) can effectively reduce power fluctuations in power system operations. Existing methods are mainly designed for short-term predictions and lack effective spatial-temporal feature augmentation.
  arXiv  Detail & Related papers  (2023-05-30T04:03:15Z)
• Forecasting Photovoltaic Power Production using a Deep Learning Sequence to Sequence Model with Attention [0.0]
  We propose a supervised deep learning model for end-to-end forecasting of PV power production. The proposed model is based on two seminal concepts that led to significant performance improvements in other sequence-related fields. The results show that the new design performs at or above the current state of the art of PV power forecasting.
  arXiv  Detail & Related papers  (2020-08-06T17:20:08Z)
• An Integrated Multi-Time-Scale Modeling for Solar Irradiance Forecasting Using Deep Learning [1.52292571922932]
  Short-term solar irradiance forecasting is difficult due to the non-stationary characteristic of solar power. We propose a unified architecture for multi-time-scale predictions for intra-day solar irradiance. Our proposed method results in a 71.5% reduction in the mean RMSE averaged across all the test sites.
  arXiv  Detail & Related papers  (2019-05-07T14:40:32Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.