Ultra-short-term photovoltaic power generation power prediction method based on feature enhancement and multi-source data fusion and related device

By adding feature enhancement modules to the YOLOv5s model to extract cloud cluster characteristics, combined with the principal component analysis method, dimensionality reduction meteorological data, a multi-source data set was constructed, and ultra-short-term photovoltaic power prediction was used to predict ultra-short-term photovoltaic power generation, the problem of poor fusion of multi-source data in the existing technology was solved and the accuracy of prediction was improved.

CN119742769APending Publication Date: 2025-04-01DATANG HYDROPOWER SCI & TECH RES INST CO LTD +2
0 Cites 0 Cited by

Patent Information

Application Number
CN202411830850.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-12-12
Publication Date
2025-04-01

AI Technical Summary

Technical Problem

The existing technology is difficult to achieve the effective fusion of multi-source meteorological characteristic data and cloud cluster characteristic data, resulting in low accuracy of ultra-short-term photovoltaic power generation power prediction.

Method used

The YOLOv5s model was used to extract cloud cluster features, and the meteorological data was dimensionally reduced in combination with the principal component analysis method, the training set and test set were constructed, and ultra-short-term photovoltaic power prediction was used to predict ultra-short-term photovoltaic power generation power.

Benefits of technology

Through multi-source data fusion, the accuracy of photovoltaic power generation prediction is improved, and the problem of inaccurate prediction results under single factor consideration is solved.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention provides an ultra-short-term photovoltaic power generation power prediction method based on feature enhancement and multi-source data fusion and a related device, and belongs to the technical field of photovoltaic power generation power prediction. The method comprises the following steps: constructing a training set and a test set by using obtained cloud cluster features and historical meteorological data of a photovoltaic power station after dimension reduction processing; constructing an LSTM model, training the LSTM model based on the constructed training set, and testing an LSTM model training result based on the constructed test set to obtain a trained LSTM model; and inputting to-be-predicted photovoltaic power station meteorological data and overhead cloud picture data into the trained LSTM model to carry out ultra-short-term photovoltaic power generation power prediction so as to obtain an ultra-short-term photovoltaic power generation power prediction result. The problems that effective fusion of multi-source meteorological feature data and cloud cluster feature data cannot be achieved and the accuracy of ultra-short-term photovoltaic power generation power prediction is not high in the prior art are solved.
Need to check novelty before this filing date? Find Prior Art