An AI-Based Decision-Support Pipeline for Day-Ahead Photovoltaic Forecasting

• URL: http://arxiv.org/abs/2608.02088v1
• Date: Mon, 03 Aug 2026 11:48:32 GMT
• Title: An AI-Based Decision-Support Pipeline for Day-Ahead Photovoltaic Forecasting
• Abstract summary: We develop a deployment-oriented environmental-AI pipeline for day-ahead hourly PV forecasting.<n>The pipeline corrects timestamps conventions, constructs leakage-safe solar-geometry and clearness-index features, and adds short-term atmospheric context.<n>It reduces daylight normalised RMSE by about 32% under random day-blocked evaluation and 9% under the stricter rolling-origin protocol.
• Score: 3.5367752722811794
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Reliable photovoltaic (PV) forecasts are needed for low-carbon energy systems, but newly deployed sites often have short, imperfect records. This makes standard day-ahead forecasting difficult: persistence and physical baselines can be sensitive to calibration and timestamp alignment, while single machine-learning models may capture only one structure in the data and overstate skill under non-temporal validation. We study this problem at a United Kingdom charging-station site, where PV forecast errors affect charging availability, storage scheduling, and downstream control. Using measured inverter output and publicly available meteorological inputs, we develop a deployment-oriented environmental-AI pipeline for day-ahead hourly PV forecasting. The pipeline corrects timestamp conventions, constructs leakage-safe solar-geometry and clearness-index features, adds short-term atmospheric context, and combines complementary predictors through validation-learned stacking. Against smart persistence, a clear-sky baseline that adjusts recent PV output using expected clear-sky irradiance, the best ensemble reduces daylight normalised RMSE by about 32% under random day-blocked evaluation and 9% under the stricter rolling-origin protocol. It also reduces daylight RMSE relative to the strongest individual machine-learning baseline by 6.6% and 6.4%, respectively. The results show that physics-aware stacking can support PV forecasts from limited site data, but its value depends on model class, evaluation protocol, and deployment context.

Related papers

• PARA-PV: Physics-Aware Retrieval-Augmented PV Prediction Based on Frozen Foundation Model and Distribution Shift Correction [8.475613508818208]
  PARA-PV is a Physics-Aware Retrieval-Augmented framework that embeds physical knowledge throughout the forecasting process.<n>The framework first encodes multivariate PV observations into patch-level representations.<n>It retrieves historical patches and analog trajectories consistent with the current window in temporal shape, power level, PV operating state, and intra-day period.
  arXiv  Detail & Related papers  (2026-07-09T03:15:42Z)
• EO-WM: A Physically Informed World Model for Probabilistic Earth Observation Forecasting [50.31500063307306]
  We introduce EO-WM, a video diffusion transformer for multispectral Earth Observation forecasting.<n> EO-WM incorporates a physically informed conditioning framework that represents meteorological forcing through a climatological baseline, weather anomalies, and cumulative physical stress signals.<n>Experiments show that EO-WM reduces the error in predicted Normalized Difference Vegetation Index (NDVI) decline amplitude by a relative 5.63% and improves directional hit rate by a relative 7.80%.
  arXiv  Detail & Related papers  (2026-06-25T16:53:30Z)
• Time series Foundation Models based on Physics-Informed Synthetic Histories for Cold-Start Photovoltaic Forecasting [45.753923525593045]
  The evaluation spans $440$ PV sites across four datasets and diverse climate regimes.<n>Performance is largely insensitive to the synthetic-history source, indicating that accuracy is driven more by the availability of plausible temporal context than by the specific generator.
  arXiv  Detail & Related papers  (2026-06-05T17:05:19Z)
• Tyche: One Step Flow for Efficient Probabilistic Weather Forecasting [48.263481135566316]
  We present Tyche, a one-step conditional flow model for efficient probabilistic weather forecasting.<n>Experiments on ERA5 at 1.5$circ$ and 6-hour resolution show that our Tyche, using merely a single NFE, matches or exceeds the forecast skill and calibration of state-of-the-art multi-step generative baselines.
  arXiv  Detail & Related papers  (2026-05-07T20:25:28Z)
• Intraday spatiotemporal PV power prediction at national scale using satellite-based solar forecast models [0.0]
  We present a novel framework for photovoltaic (PV) power forecasting power nowcasting models.<n>Forecasts are first validated against satellite-derived surface solar radiance (SSI)<n>We additionally provide the first visualizations of how mesoscale cloud systems shape national PV production on hourly sub-hourly.
  arXiv  Detail & Related papers  (2026-01-08T09:15:14Z)
• 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)
• Solar Forecasting with Causality: A Graph-Transformer Approach to Spatiotemporal Dependencies [5.291394959125948]
  SolarCAST is a causally informed model predicting future global horizontal radiance (GHI) at a target site using only historical GHI from site X and nearby stations S.<n>It outperforms leading time-series and multimodal baselines across diverse geographical conditions.<n>It achieves a 25.9% error reduction over the top commercial forecaster, Solcast.
  arXiv  Detail & Related papers  (2025-09-18T22:57:07Z)
• Improving day-ahead Solar Irradiance Time Series Forecasting by Leveraging Spatio-Temporal Context [46.72071291175356]
  Solar power harbors immense potential in mitigating climate change by substantially reducing CO$_2$ emissions. However, the inherent variability of solar irradiance poses a significant challenge for seamlessly integrating solar power into the electrical grid. In this paper, we put forth a deep learning architecture designed to harnesstemporal context using satellite data.
  arXiv  Detail & Related papers  (2023-06-01T19:54:39Z)
• Short-term Prediction and Filtering of Solar Power Using State-Space Gaussian Processes [11.16908638844702]
  We consider Gaussian processes (GPs) for modelling and predicting solar photovoltaic energy production in the UK. A standard application of GP regression on the PV timeseries data is infeasible due to the large data size and non-Gaussianity of PV readings. This is made possible by leveraging recent advances in scalable GP inference, in particular, by using the state-space form of GPs, combined with modern variational inference techniques.
  arXiv  Detail & Related papers  (2023-02-01T11:48:10Z)
• Data-driven soiling detection in PV modules [58.6906336996604]
  We study the problem of estimating the soiling ratio in photo-voltaic (PV) modules. A key advantage of our algorithms is that they estimate soiling, without needing to train on labelled data. Our experimental evaluation shows that we significantly outperform current state-of-the-art methods for estimating soiling ratio.
  arXiv  Detail & Related papers  (2023-01-30T14:35:47Z)
• Principal Component Density Estimation for Scenario Generation Using Normalizing Flows [62.997667081978825]
  We propose a dimensionality-reducing flow layer based on the linear principal component analysis (PCA) that sets up the normalizing flow in a lower-dimensional space. We train the resulting principal component flow (PCF) on data of PV and wind power generation as well as load demand in Germany in the years 2013 to 2015.
  arXiv  Detail & Related papers  (2021-04-21T08:42:54Z)
• Short-term prediction of photovoltaic power generation using Gaussian process regression [3.8386504037654534]
  This paper focuses on evaluating predictions of the energy generated by PV systems in the United Kingdom. The model is evaluated for short-term forecasts of 48 hours against three main factors. We also compare very short-term forecasts in terms of cloud coverage within the prediction period and only initial cloud coverage as a predictor.
  arXiv  Detail & Related papers  (2020-10-05T18:35:25Z)

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.