IDS-Net: A novel framework for few-shot photovoltaic power prediction with interpretable dynamic selection and feature information fusion

2025-07-16

AI Summary

To address low photovoltaic (PV) power forecasting accuracy in newly commissioned power plants due to scarce historical data, this paper proposes an interpretable few-shot cross-domain forecasting framework. Methodologically, it innovatively integrates dynamic source-domain selection, a dual-weighted channel mechanism, and interpretable feature fusion, while incorporating Maximum Mean Discrepancy (MMD)-based distribution alignment, ReliefF-based feature selection, Hampel outlier correction, and weighted multi-submodel ensemble—forming an end-to-end adaptive transfer learning pipeline. Experiments on real-world operational data from two newly built PV plants in Hebei Province demonstrate that the framework significantly improves forecasting accuracy under few-shot settings (≤7 days of historical data), reducing mean absolute error (MAE) by 21.3%. It further exhibits strong generalizability, decision interpretability, and engineering practicality, providing a reliable technical foundation for early-stage operation optimization of new energy power stations.

Technology Category

• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Computer Vision: Learning & Optimization for CV
• Intelligent Robots: State Estimation

Application Category

• Web Mining and Content Analysis: Large pretrained models with web data
• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

With the growing demand for renewable energy, countries are accelerating the construction of photovoltaic (PV) power stations. However, accurately forecasting power data for newly constructed PV stations is extremely challenging due to limited data availability. To this end, we propose a novel interpretable dynamic selection network (IDS-Net) based on feature information fusion to achieve accurate few-shot prediction. This transfer learning framework primarily consists of two parts. In the first stage, we pre-train on the large dataset, utilizing Maximum Mean Discrepancy (MMD) to select the source domain dataset most similar to the target domain data distribution. Subsequently, the ReliefF algorithm is utilized for feature selection, reducing the influence of feature redundancy. Then, the Hampel Identifier (HI) is used for training dataset outlier correction. In the IDS-Net model, we first obtain the initial extracted features from a pool of predictive models. Following this, two separate weighting channels are utilized to determine the interpretable weights for each sub-model and the adaptive selection outcomes, respectively. Subsequently, the extracted feature results from each sub-model are multiplied by their corresponding weights and then summed to obtain the weighted extracted features. Then, we perform cross-embedding on the additional features and fuse them with the extracted weighted features. This fused information is then passed through the MLP (Multi-Layer Perceptron) layer to obtain predictions. In the second stage, we design an end-to-end adaptive transfer learning strategy to obtain the final prediction results on the target dataset. We validate the transfer learning process using two PV power datasets from Hebei province, China, to demonstrate the effectiveness and generalization of our framework and transfer learning strategy.

Problem

• Accurate few-shot prediction for new PV stations with limited data
• Interpretable dynamic selection and feature fusion for PV power forecasting
• Adaptive transfer learning strategy for cross-domain PV data prediction

Innovation

• Uses MMD for similar source domain selection
• Applies ReliefF algorithm for feature selection
• Implements interpretable dynamic weighting in IDS-Net