A method for constructing a double-carbon digital system model of carbon metering of a power system

By combining offline learning with online decision-making, a multi-stage scenario set of new energy output is generated and a deterministic optimization problem is solved. A state-value function dataset is constructed, which solves the problem of dynamic measurement and tracking of carbon emissions under the uncertainty of new energy in the power system, and realizes the real-time and accuracy of carbon-energy collaborative optimization of the power system.

CN122197384APending Publication Date: 2026-06-12INNER MONGOLIA LANGRUN ENERGY TECHNOLOGY CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
INNER MONGOLIA LANGRUN ENERGY TECHNOLOGY CO LTD
Filing Date
2026-04-15
Publication Date
2026-06-12

AI Technical Summary

⚠Technical Problem

Existing technologies struggle to achieve dynamic and accurate measurement and carbon flow tracking of carbon emissions in the power system while considering the uncertainties of new energy sources. They also struggle to support multi-stage collaborative optimization. Traditional methods have low measurement accuracy and cannot track the sources of carbon emissions on the electricity consumption side.

⚗Method used

We employ a method that combines offline learning with online decision-making. We generate a multi-stage set of scenarios for new energy output through a deep generative model, solve a multi-stage deterministic optimization problem, construct a state-value function dataset, train a deep generative model of value function to output the probability distribution of future cost values, and make real-time decisions in conjunction with carbon flow physics equations.

🎯Benefits of technology

It achieves real-time and accurate carbon-energy collaborative decision-making in the power system under the uncertainty of new energy, improves the accuracy of carbon emission measurement and the robustness of decision-making, and supports the forward-looking and risk quantification capabilities of multi-stage optimization.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122197384A_ABST
    Figure CN122197384A_ABST
Patent Text Reader

Abstract

The application provides a double-carbon digital system model construction method for carbon measurement of a power system, belongs to the technical field of carbon measurement of a power system, and is used for solving the problem of difficulty in dynamic measurement and collaborative optimization of carbon emission under new energy uncertainty in the related art. The application generates a new energy multi-stage scene set by using a diffusion probability model through receiving multi-source historical data; a multi-stage deterministic optimization problem considering a carbon flow physical equation is solved for each scene to construct a state-value function dataset; a conditional diffusion probability model is trained as a value function deep generation model; in online operation, a measured state variable is input into the model to obtain a future cost probability distribution estimation, and then a single-stage optimization problem is constructed and solved to obtain a current decision and execute it. The application realizes the fusion of offline learning and online decision making, can accurately measure carbon emission and track carbon flow, effectively handles new energy uncertainty, and supports real-time rolling optimization decision making under the double-carbon target.
Need to check novelty before this filing date? Find Prior Art