A power prediction-based photovoltaic effective output optimization method and device

By combining multi-source data preprocessing and a hybrid model of CNN-LSTM attention mechanism with an adaptive particle swarm optimization algorithm, high-precision prediction of photovoltaic power and dynamic adaptation of optimal output commands are achieved, solving the problem of inaccurate photovoltaic power prediction in existing technologies and improving the stability and economic benefits of photovoltaic power generation systems.

CN122246878APending Publication Date: 2026-06-19华能澜沧江新能源有限公司
0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
华能澜沧江新能源有限公司
Filing Date
2026-03-20
Publication Date
2026-06-19

AI Technical Summary

⚠Technical Problem

Existing technologies cannot accurately predict photovoltaic power and dynamically solve for the optimal output command that is adapted to the current operating conditions, resulting in the inability to maximize the effective output of photovoltaics and affecting the stability and economic benefits of photovoltaic power generation systems.

⚗Method used

A hybrid model combining multi-source data preprocessing and CNN-LSTM attention mechanism with an adaptive particle swarm optimization algorithm is adopted to achieve high-precision prediction of photovoltaic power and dynamic adaptation solution of optimal output command.

🎯Benefits of technology

By accurately predicting photovoltaic power and dynamically solving for the optimal output command, the effective output of photovoltaic power can be maximized, thereby improving the energy utilization rate and economic benefits of photovoltaic power generation systems.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
Patent Text Reader

Abstract

This application provides a method and apparatus for optimizing effective photovoltaic (PV) output based on power prediction, relating to the field of PV technology. It addresses the technical problem in existing technologies where accurate prediction of PV power and dynamic solution of the optimal output command adapted to the current operating conditions are impossible, leading to the inability to maximize effective PV output. The method specifically includes: acquiring multi-source data; preprocessing the multi-source data to form a fused feature vector; predicting PV power based on the fused feature vector using a CNN-LSTM attention mechanism hybrid model, and outputting the predicted power value; constructing an optimization problem based on the predicted power value; and solving the optimization problem using an adaptive particle swarm optimization algorithm to output the optimal output command. This application is used for optimizing effective PV output.
Need to check novelty before this filing date? Find Prior Art