# US 12,112,391 B2 · Methods and systems for machine-learning for prediction of grid carbon emissions > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,112,391 B2 - Document ID: US12112391B2 - Status: Granted - Kind: B2 - Application number: 18/198,626 - Filed: 2023-05-17 - Published: 2023-10-19 - Granted: 2024-10-08 - Earliest claimed: 2019-05-20 - Assignees: Singularity Energy, Inc. - Applicants: Singularity Energy, Inc. - Inventors: Wenbo Shi - Web page: https://patentgrep.com/patent/US12112391B2 ## CPC - G06Q 50/06: Energy or water supply - G06N 3/0442: characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU] - G06N 3/0464: Convolutional networks [CNN, ConvNet] - G06N 3/08: Learning methods - G06N 3/09: Supervised learning - G06N 5/01: Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound - G06N 20/00: Machine learning - G06Q: INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS ## Abstract A method for calculating consumption-based carbon emissions in power grids is disclosed. The method includes receiving, from a first grid monitoring device monitoring a first grid, a plurality of first power output quantities from a plurality of power generators in the first grid; receiving, from a second grid monitoring device monitoring a second grid, a plurality of second power output quantities from a plurality of power generators in the second grid, wherein the first grid is connected to the second grid and can send and/or receive power to and/or from the second grid; computing and solving a linear system of equations based on variables representative of each the first grid and the second grid; and displaying, on a user interface, a consumption-based carbon emissions calculation describing carbon emissions consumed per megawatt-hours (MWh) produced by the first grid and the second grid based on solving the linear system of equations. ## Claims ### Claim 1 A method of using machine-learning for calculating consumption-based carbon emissions in multiple interconnected power grids, the method comprising: receiving, from a first local grid monitoring device, a plurality of first power output quantities, wherein each first power output quantity of the plurality of first power output quantities represents power output generated in a first local grid; receiving, from a second local grid: at least a second power output quantity; and a carbon intensity of the second local grid; training, by a computing device, an emission estimation machine-learning process using training data entries, wherein the training data entries comprise correlations between past power output quantities and reported carbon emission data; generating, using the emission estimation machine-learning process, a first plurality of projected carbon emission rates, wherein each of the first plurality of projected carbon emission rates is generated as a function of the first plurality of power output quantities and the at least a second power output quantity; and displaying, on a user interface, a consumption-based carbon emissions calculation describing carbon emissions consumed per megawatt-hours (MWh) produced by the first grid and the second grid based on solving a linear system of equations. ### Claim 2 (depends on 1) The method of claim 1, further comprising: tuning, by the computing device, matrices of coefficients using the emission estimation machine-learning process and historical data. ### Claim 3 (depends on 1) The method of claim 1, further comprising: tuning, by the computing device, matrices of coefficients using the emission estimation machine-learning process and a regression algorithm. ### Claim 4 (depends on 1) The method of claim 1, further comprising: measuring, by the computing device, power sent from the first local grid to the second local grid by subtracting a corresponding amount of first power output quantities from the plurality of first power output quantities. ### Claim 5 (depends on 1) The method of claim 1, further comprising: receiving, from a second local grid monitoring device, a plurality of second power output quantities, wherein each second power output quantity of the plurality of second power output quantities represents power output generated in the second local grid. ### Claim 6 (depends on 5) The method of claim 5, further comprising: measuring, by the computing device, power sent from the second local grid to the first local grid by subtracting a corresponding amount of second power output quantities from the plurality of second power output quantities. ### Claim 7 (depends on 6) The method of claim 6, further comprising: exchanging power between the first local grid and the second local grid, wherein exchanged power output quantities are reported by a respective local power grid monitoring device; and determining, by the computing device, that the second local grid is a power source based on receipt of at least some of the plurality of second power output quantities by at least the first local grid monitoring device. ### Claim 8 (depends on 1) The method of claim 1, further comprising: calculating, by the computing device, a plurality of values representative of at least the second local grid; and modeling the second local grid as a power generator based on the plurality of values. ### Claim 9 (depends on 1) The method of claim 1, further comprising: identifying at least a power generator at the second grid; and obtaining a power output quantity of the identified at least a power generator. ### Claim 10 (depends on 1) The method of claim 1, further comprising: generating, by the computing device, data describing at least a recommended course of action by the emission estimation machine-learning process; and determining, by the computing device, predicted grid carbon emissions based on at least the recommended course of action and at least the first plurality of power output quantities. ### Claim 11 A system for using machine-learning for calculating consumption-based carbon emissions in multiple interconnected power grids, the system comprising a computing device configured to: receive, from a first local grid monitoring device, a plurality of first power output quantities, wherein each first power output quantity of the plurality of first power output quantities represents power output generated in a first local grid; receive, from a second local grid: at least a second power output quantity; and a carbon intensity of the second local grid; train, by a computing device, an emission estimation machine-learning process using training data entries, wherein the training data entries comprise correlations between past power output quantities and reported carbon emission data; generate, using the emission estimation machine-learning process, a first plurality of projected carbon emission rates, wherein each of the first plurality of projected carbon emission rates is generated as a function of the first plurality of power output quantities and the at least a second power output quantity; and display, on a user interface, a consumption-based carbon emissions calculation describing carbon emissions consumed per megawatt-hours (MWh) produced by the first grid and the second grid based on solving a linear system of equations. ### Claim 12 (depends on 11) The system of claim 11, wherein the computing device is further configured to: tune, by the computing device, matrices of coefficients using the emission estimation machine-learning process and historical data. ### Claim 13 (depends on 11) The system of claim 11, wherein the computing device is further configured to: tune, by the computing device, matrices of coefficients using the emission estimation machine-learning process and a regression algorithm. ### Claim 14 (depends on 11) The system of claim 11 wherein the computing device is further configured to: measure power sent from the first local grid to the second local grid by subtracting a corresponding amount of first power output quantities from the plurality of first power output quantities. ### Claim 15 (depends on 11) The system of claim 11, wherein the computing device is further configured to: receive, from a second local grid monitoring device, a plurality of second power output quantities, wherein each second power output quantity of the plurality of second power output quantities represents power output generated in the second local grid. ### Claim 16 (depends on 15) The system of claim 15, wherein the computing device is further configured to: measure power sent from the second grid to the first grid by subtracting a corresponding amount of second power output quantities from the plurality of second power output quantities. ### Claim 17 (depends on 15) The system of claim 15, wherein the computing device is further configured to: exchange power between the first local grid and the second local grid, wherein exchanged power output quantities are reported by a respective local power grid monitoring device; and determine, by the computing device, that the second local grid is a power source based on receipt of at least some of the plurality of second power output quantities by at least the first local grid monitoring device. ### Claim 18 (depends on 11) The system of claim 11, wherein the computing device is further configured to: calculate, by the computing device, a plurality of values representative of at least the second local grid; and model, by the computing device, the second local grid as a power generator based on the plurality of values. ### Claim 19 (depends on 11) The system of claim 11, wherein the computing device is further configured to: identify, by the computing device, at least a power generator at the second grid; and obtain, by the computing device, a power output quantity of the identified at least a power generator. ### Claim 20 (depends on 11) The system of claim 11, wherein the computing device is further configured to: generate, by the computing device, data describing at least a recommended course of action by the emission estimation machine-learning process; and determine, by the computing device, predicted grid carbon emissions based on at least the recommended course of action and at least the first plurality of power output quantities.