Model training method and device, energy carbon prediction method, equipment and system, and medium

By constructing a multi-level dynamic weighted energy and carbon prediction model, and combining equipment energy consumption, process parameters, and environmental data, the problem of coordinated prediction of energy consumption and carbon emissions in the painting workshop was solved, achieving coordinated optimization of energy consumption and carbon emissions and improving the company's response and control capabilities.

CN120995108APending Publication Date: 2025-11-21JIANGSU XCMG CONSTRUCTION MACHINERY RESEARCH INSTITUTE LTD
6 Cites 0 Cited by

Patent Information

Application Number
CN202511123614.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-08-12
Publication Date
2025-11-21

AI Technical Summary

Technical Problem

Existing technologies lack the ability to predict energy consumption and carbon emissions in a coordinated manner in painting workshops, making it difficult to adapt to the dynamic characteristics of multi-stage processes and thus limiting the effective implementation of refined management and control measures by enterprises.

Method used

By constructing a multi-level dynamic weighted energy and carbon prediction model, combining equipment energy consumption, process parameters and environmental data, and training it with a random forest model, a multi-level dynamic weighted mechanism is set to adjust the weights of energy consumption and carbon emissions, thereby achieving coordinated prediction and optimization of energy consumption and carbon emissions.

Benefits of technology

It enables the dual-coordinated prediction and optimization of energy consumption and carbon emissions in the painting workshop, enhances the company's response and control capabilities, and supports green transformation and energy efficiency improvement.

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

Abstract

The invention relates to a model training method and device, an energy carbon prediction method, equipment and system, and a medium. The model training method comprises the steps that multi-source data of an enterprise coating workshop are acquired, and the multi-source data comprise at least one of equipment energy consumption, process parameters and environment data; constructing an energy carbon prediction model; for the energy-carbon prediction model, a multi-level dynamic weighting mechanism is set, the energy consumption weight and the carbon emission weight of the energy-carbon prediction model are adjusted, and multiple levels comprise a process stage layer, a target weight layer and a feature weight layer; the multi-source data is adopted to train the energy-carbon prediction model after the multi-level dynamic weighting mechanism is set, and the trained energy-carbon prediction model is used for conducting collaborative prediction optimization on energy consumption and carbon emission of the coating workshop. According to the invention, double-collaborative prediction and optimization of energy consumption and carbon emission of the coating workshop can be realized, and the capability of coping with and regulating future energy consumption and carbon emission of an enterprise is improved.
Need to check novelty before this filing date? Find Prior Art