A Power Prediction Method and System for New Energy Power Plants Based on Energy Storage Collaborative Optimization

By constructing a power prediction method for new energy power plants that integrates energy storage and optimization, and utilizing variational mode decomposition and combined prediction models, along with real-time data from the energy storage system, the accuracy and stability issues of power prediction for new energy power plants are resolved, achieving efficient power prediction and economic optimization of the energy storage system.

CN121584562BActive Publication Date: 2026-05-26XIAN THERMAL POWER RES INST CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
XIAN THERMAL POWER RES INST CO LTD
Filing Date
2026-01-26
Publication Date
2026-05-26

AI Technical Summary

Technical Problem

Existing power prediction technologies for new energy power plants suffer from problems such as insufficient processing capability for non-stationary sequences, limited feature dimensions and correlations, disconnect between prediction and energy storage regulation, and insufficient model generalization ability, resulting in low prediction accuracy.

Method used

By collecting real-time data from new energy power plants, a real-time feature set containing the operating characteristics of energy storage systems is constructed. Variational mode decomposition and sparrow search algorithms are used to optimize the number of modes and penalty factors. A combined prediction model of temporal convolutional network and bidirectional gated recurrent unit is used to predict power. Based on the predicted values, a hierarchical energy storage charging and discharging strategy is implemented to smooth out fluctuations.

Benefits of technology

It significantly improves the accuracy and stability of ultra-short-term power prediction for new energy power plants, realizes deep synergistic optimization between energy storage systems and power prediction, and improves the stability of grid operation and the efficiency of new energy absorption.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN121584562B_ABST
    Figure CN121584562B_ABST
Patent Text Reader

Abstract

This disclosure provides a power prediction method and system for new energy power plants based on energy storage collaborative optimization. By simultaneously collecting power, meteorological, and energy storage system operation data from the new energy power plant, and forcibly ensuring a certain proportion of energy storage features in the feature set, power prediction is deeply coupled with energy storage status. Through adaptive decomposition of the power sequence, each component is fused with a feature set containing energy storage information, providing a hierarchical and information-rich model input. The combined prediction model composed of a temporal convolutional network and a bidirectional gated recurrent unit achieves targeted prediction of different frequency fluctuation components. This disclosure provides support for stable grid operation and efficient new energy consumption, solving the problems of disconnect between the prediction model and energy storage system operation, and limited prediction accuracy.
Need to check novelty before this filing date? Find Prior Art