A roof photovoltaic power generation prediction method fusing multi-source data

By integrating multi-source data to predict rooftop photovoltaic power generation, a power generation efficiency scoring model and a two-level prediction model were constructed. This addressed the shortcomings in pollution impact assessment and cleaning decisions during the operation and maintenance of photovoltaic systems, enabling precise photovoltaic panel management and intelligent cleaning strategies, thereby improving power generation efficiency and operational benefits.

CN121562882BActive Publication Date: 2026-09-15HUANENG (CHANGSHA) NEW ENERGY CO LTD
2 Cites 0 Cited by

Patent Information

Application Number
CN202511618705.4
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2025-11-06
Publication Date
2026-09-15
Estimated Expiration
2045-11-06

AI Technical Summary

Technical Problem

Existing operation and maintenance methods for rooftop photovoltaic systems cannot accurately assess the type and density of pollution and their differential impact on light transmittance, resulting in decreased power generation efficiency. Furthermore, cleaning decisions are out of sync with the system's power generation needs, making it impossible to proactively and accurately plan cleaning tasks to compensate for the power generation shortfall.

Method used

A rooftop photovoltaic power generation prediction method that integrates multi-source data collects photovoltaic panel status parameters, constructs a power generation efficiency scoring model, establishes a two-level prediction model including power generation prediction and power generation improvement prediction, and formulates intelligent cleaning strategies to fill power generation gaps.

Benefits of technology

It enables precise analysis and hierarchical management of the health status of photovoltaic panels, breaking through the limitations of traditional single-data judgment, improving the pertinence and efficiency of operation and maintenance work, and enhancing the energy output efficiency and intelligent operation level of photovoltaic systems.

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

Abstract

The application belongs to the technical field of photovoltaic systems, and discloses a roof photovoltaic power generation prediction method fusing multiple source data; comprising the following steps: S1: collecting state parameters influencing power generation efficiency of photovoltaic panels in real time through a data acquisition module.The application realizes accurate analysis and hierarchical management of the health state of roof photovoltaic panels by setting a power generation efficiency score, breaks through the limitation of traditional dependence on single power data or simple visual judgment, carries out normalization processing and weighted fusion on four physical parameters, i.e., a solar incident angle, a dirt coverage rate, a dirt type index and a photovoltaic panel aging coefficient, forms a comprehensive power generation efficiency score, and classifies photovoltaic panels into high, medium and low efficiency levels based on the score, thereby providing intuitive and accurate management basis for operation and maintenance personnel, which enables resources to be preferentially invested in photovoltaic panels with the lowest efficiency and the greatest improvement potential, and greatly improves the pertinence and efficiency of operation and maintenance work.
Need to check novelty before this filing date? Find Prior Art