# US 11,907,865 B2 · Determination of security-constrained optimal power flow > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 11,907,865 B2 - Document ID: US11907865B2 - Status: Granted - Kind: B2 - Application number: 16/821,060 - Filed: 2020-03-17 - Published: 2020-09-24 - Granted: 2024-02-20 - Earliest claimed: 2019-03-18 - Assignees: Hitachi Energy Ltd - Applicants: Hitachi Energy Ltd - Inventors: Marco Giuntoli; Veronica Biagini - Web page: https://patentgrep.com/patent/US11907865B2 ## CPC - G06F 30/20: Design optimisation, verification or simulation (optimisation, verification or simulation of circuit designs G06F30/30) - G06N 7/01: Probabilistic graphical models, e.g. probabilistic networks - G06N 20/00: Machine learning - G06Q 10/04: Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem" (market predictions or forecasting for commercial activities G06Q30/0202) - G06Q 50/06: Energy or water supply - H02J 3/381: {Dispersed generators} - G06F 2111/06: Multi-objective optimisation, e.g. Pareto optimisation using simulated annealing [SA], ant colony algorithms or genetic algorithms [GA] - G06F 2119/06: Power analysis or power optimisation - H02J 3/003: {Load forecast, e.g. methods or systems for forecasting future load demand} - H02J 2103/30: Simulating, planning, modelling, reliability check or computer assisted design [CAD] of electric power networks - Y02E 40/70: Smart grids as climate change mitigation technology in the energy generation sector - Y02E 60/00: Enabling technologies; Technologies with a potential or indirect contribution to GHG emissions mitigation - Y04S 10/50: Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications - Y04S 40/20: Information technology specific aspects, e.g. CAD, simulation, modelling, system security - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS - 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 - H02J: ELECTRIC POWER NETWORKS; CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY - Y02E: REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION - Y04S: SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS ## Abstract A method can be used to determine a security-constrained optimal power flow (SC-OPF) of a power grid. Input data associated with a power grid scenario is obtained. The input data defines a security-constrained optimization problem (SC-OPF problem). The power grid scenario includes a power grid structure, a power demand and a generator capability and/or cost. Power flows in branches of the power grid that are equal to or greater than limits for a contingency associated with the power grid scenario are estimated based on the obtained input data by an AI. A modified optimization problem which is smaller than the SC-OPF problem is solved based on the estimated power flows. ## Claims ### Claim 1 A method of operating a power grid, the method comprising, by at least one processor executing at least one program element: collecting input data, from a central and/or decentralized monitoring device, measuring device, or computer model, associated with a power grid scenario, the input data defining a security-constrained optimization problem (SC-OPF problem), wherein the power grid scenario comprises a power grid structure, a power demand and a generator capability and/or cost; estimating, based on the obtained input data, power flows in branches of the power grid that are equal to or greater than limits for a contingency associated with the power grid scenario, the estimating performed by an artificial intelligence (AI) module; reducing, by the AI module, the SC-OPF problem to a modified optimization problem, the modified optimization problem being smaller than the SC-OPF problem; determining, based on the estimated power flows, a solution to the modified optimization problem by a solver module; and operating the power grid based on the solution of the modified optimization problem. ### Claim 2 (depends on 1) The method according to claim 1, wherein determining the solution to the modified optimization problem is performed iteratively until a termination criterion is reached. ### Claim 3 (depends on 1) The method according to claim 1, wherein determining the solution to the modified optimization problem is performed iteratively until no violation is identified or violations are eliminated. ### Claim 4 (depends on 1) The method according to claim 1, wherein determining the solution to the modified optimization problem comprises adding possible new flow violations iteratively. ### Claim 5 (depends on 1) The method according to claim 1, further comprising checking the solution of the modified optimization problem to determine whether violations occur. ### Claim 6 (depends on 1) The method according to claim 1, wherein solution to the modified optimization problem is provided as SC-OPF problem solution output data. ### Claim 7 (depends on 1) The method according to claim 1, wherein the power grid scenario input data comprises at least one or more information selected from: a number of nodes arranged within the power grid; a number of lines arranged within the power grid; an impedance of one or more lines and/or a flow limit of one or more lines; a number of electrical generators integrated in the power grid; a power limit of one or more electrical generators and/or a cost value of one or more electrical generators; a number of loads integrated in the power grid; and a power consumption value of one or more loads and/or a position within the power grid of one or more loads. ### Claim 8 (depends on 1) The method according to claim 1, wherein the SC-OPF problem to be solved is to satisfy an N−1 criterion, the security-constrained optimization problem to be expressible by (NL2+1)·NN, wherein NL is a number of branches of the power grid and NN is a number of nodes of the power grid. ### Claim 9 (depends on 1) The method according to claim 1, wherein solving the modified optimization problem is performed during real-time operation of the power grid. ### Claim 10 A method of operating a power grid assisted by an artificial intelligence (AI) module, the method comprising, by at least one processor executing at least one program element: during a first phase, generating a training data set that comprises a plurality of solutions of at least one security-constrained optimal power flow (SC-OPF) problem associated with at least one power grid scenario, the training data set comprising training input data and corresponding training output data, the training output data being the expected output when the corresponding input data is input; during a second phase, training the AI-module by supplying the training data set to the AI-module; and during a third phase, operating the power grid in a power grid scenario, the operation at least partly depending on SC-OPF problem solution output data obtained by solving a modified optimization problem that has a size that has been reduced by the trained AI-module in response to input data associated with the power grid scenario. ### Claim 11 (depends on 10) The method according to claim 10, wherein, during the third phase, the AI-module is retrained using SC-OPF solution data, the retraining comprising: receiving output data from the AI-module resulting from the input of the training data set; comparing the output data and the training output data corresponding to the training input data of the training data set; and adjusting parameters of the AI-module based on the comparison. ### Claim 12 (depends on 10) The method according to claim 10, wherein the modified optimization problem is solved during real-time operation of the power grid. ### Claim 13 A method of training an artificial intelligence (AI) module to determine a security-constrained optimal power flow (SC-OPF) of a power grid, the method comprising, by at least one processor executing at least one program element: providing input data associated with a power grid scenario, the input data defining a security-constrained optimization problem; generating a training data set, the training data set comprising a plurality of solutions of a SC-OPF problem associated with at least one given power grid scenario, the training data set comprising training input data and corresponding training output data, the training output data being the expected output when the corresponding input data is input; and providing the training data set to the AI-module; receiving output data from the AI-module resulting from the input of the training data set; comparing the output data and the training output data corresponding to the training input data of the training data set; and adjusting parameters of the AI-module based on the comparison. ### Claim 14 (depends on 13) The method according to claim 13, wherein the SC-OPF problem is solved a number of times and a database of these optimal solutions is created. ### Claim 15 (depends on 13) The method according to claim 13, further comprising retraining the AI-module based on additional information obtained after the power grid is operated. ### Claim 16 A power grid, comprising: a plurality of distributed electrical generators; a plurality of distributed loads; a plurality of branches, adapted to electrically connect the electrical generators and/or the loads to each other; and a control apparatus, adapted to control operation of at least a part of the power grid and to determine a security-constrained optimal power flow (SC-OPF) of the power grid, the control apparatus comprising an artificial intelligence (AI) module and a solver module; wherein the AI-module is adapted to automatically generate SC-OPF problem solution output data associated with the SC-OPF, wherein the AI-module is further adapted to, in response to received input data, from a central and/or decentralized monitoring device, measuring device, or computer model, associated with a power grid scenario, reduce the SC-OPF problem to a modified optimization problem which is smaller than the SC-OPF problem; wherein the solver module is adapted to solve, the modified optimization problem; and wherein the control apparatus is further adapted to operate the power grid based on at least the SC-OPF problem solution output data. ### Claim 17 (depends on 16) The power grid according to claim 16, wherein the AI-module further adapted to be retrained after the power grid is operated based on at least the SC-OPF problem solution output data. ### Claim 18 (depends on 16) The power grid according to claim 16, further comprising a database accessible by the control apparatus and adapted to a plurality of SC-OPF solutions. ### Claim 19 (depends on 16) The power grid according to claim 16, wherein the control apparatus is adapted to solve the modified optimization problem iteratively until a termination criterion is reached. ### Claim 20 (depends on 16) The power grid according to claim 16, wherein the control apparatus is adapted to solve the modified optimization problem iteratively until no violation is identified or violations are eliminated. ### Claim 21 (depends on 16) The power grid according to claim 16, wherein the control apparatus is adapted to solve the modified optimization problem by adding possible new flow violations iteratively.