Carbon emission calculating system based on artificial intelligence
Patent Information
- Application Number
- TW115110094
- Authority / Receiving Office
- TW · TW
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2026-03-18
- Publication Date
- 2026-11-11
- Estimated Expiration
- 2046-03-17
Smart Images

Figure TWG2TB001921124_001 
Figure TWG2TB001921124_002 
Figure TWG2TB001921124_003
Abstract
Claims
1. An artificial intelligence-based carbon emission management system, comprising: an input module configured to receive integrated data associated with an enterprise, the integrated data including financial data and operating capacity data; a carbon emission management knowledge base storing multiple carbon emission comparison tables corresponding to multiple industries; and an artificial intelligence module connected to the input module and the carbon emission management knowledge base, configured to perform: receiving the integrated data from the input module; parsing the integrated data to determine a corresponding industry for the enterprise; searching the carbon emission management knowledge base to obtain a carbon emission comparison table for the corresponding industry; calculating a predicted carbon emission data based on the integrated data and the carbon emission comparison table; and using a predicted energy consumption generation module, combining the time characteristics or equipment characteristics of the integrated data, allocating the predicted carbon emission data to a time series or equipment series.
2. The carbon emission management system as described in claim 1, wherein the integrated data further includes physical device sensing data, which is sensing data including timestamps obtained by sensing a physical device, and the artificial intelligence module is configured to extract physical operating characteristics from the physical device sensing data using a feature extraction function and substitute them into an energy consumption equation corresponding to the device to calculate the actual energy usage data corresponding to the physical device and having timestamps.
3. The carbon emission management system as described in claim 2, wherein the physical device sensing data corresponds to a plurality of physical devices, the physical devices include two or more physical device types, the physical device sensing data is sensing data including timestamps obtained by sensing the physical devices, the artificial intelligence module is configured to identify the type of each device, extract physical operating characteristics from the physical device sensing data by the feature extraction function, and substitute them into the energy consumption equation corresponding to each physical device type to calculate the actual energy usage data corresponding to the physical device and having timestamps.
4. The carbon emission management system as described in claim 3 further includes a generation module that generates a comprehensive carbon emission report based on the actual energy usage and including carbon emission trends and abnormal equipment alerts.
5. The carbon emission management system as described in claim 3 further includes a comparison and correction module connected to the artificial intelligence module. First, it calculates the predicted energy consumption corresponding to the predicted carbon emission data. Then, it determines whether the predicted energy consumption needs to be corrected. For example, it uses a statistical error index to determine the error between the predicted energy consumption and the actual energy consumption in a specific time series. When the statistical error index exceeds a preset threshold, it determines that the predicted energy consumption needs to be corrected.
6. The carbon emission management system as described in claim 5, after determining that the predicted energy consumption needs to be corrected, triggers the artificial intelligence module to perform fine-tuning.
7. The carbon emission management system as claimed in claim 1, wherein when the AI module retrieves the integrated data, if it infers that the retrieved data is incorrect or cannot be retrieved, and an external source is required to correct the error or obtain the data, the AI module searches the Internet to obtain the external source.
8. The carbon emission management system as described in claim 1 further includes a category mapping module connected to the artificial intelligence module; the category mapping module extracts multiple reference fields and a footprint symbol from the carbon emission comparison table into data to be calculated, and maps it to a first category, a second category, or a third category, the data to be calculated including multiple categories such as energy consumption category, raw material consumption category, and transportation and logistics category, and multiple fields to be calculated corresponding to each category.
9. The carbon emission management system as described in claim 8 further includes a carbon emission calculation module connected to the category mapping module and configured to perform: calculating an energy consumption carbon emission based on an energy consumption calculation rule and a plurality of the fields to be calculated for the energy consumption category; calculating a raw material consumption carbon emission based on a raw material consumption calculation rule and a plurality of the fields to be calculated for the raw material consumption category; and calculating a transportation and logistics carbon emission based on a transportation and logistics calculation rule and a plurality of the fields to be calculated for the transportation and logistics category.
10. The carbon emission management system as claimed in claim 9 further includes a generation module connected to the carbon emission calculation module, the generation module generating a comprehensive carbon emission report based on the carbon emissions from energy consumption, the carbon emissions from raw material consumption, and the carbon emissions from transportation and logistics.
Citation Information
Patent Citations
Urban carbon emission accounting method based on night light remote sensing and improved STIRPAT
CN114461697B
A method for predicting long-term carbon emissions in energy-consuming areas
CN117151732B
Training method, device, equipment and storage medium for carbon emission prediction model
CN118822295B
Power storage system selection method and apparatus, electronic device and storage medium
WO2026000911A1
