# US 2022/0285938 A1 · Method and apparatus for managing predicted power resources for an industrial gas plant complex

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2022/0285938 A1
- Document ID: US20220285938A1
- Status: Published application
- Kind: A1
- Application number: 17/193,029
- Filed: 2021-03-05
- Published: 2022-09-08
- Earliest claimed: 2021-03-05
- Assignees: Air Products and Chemicals, Inc.
- Applicants: Air Products and Chemicals, Inc.
- Inventors: Sanjay Mehta; Pratik Misra
- Web page: https://patentgrep.com/patent/US20220285938A1

## CPC

- H02J 3/004: {Generation forecast, e.g. methods or systems for forecasting future energy generation}
- G05B 13/0265: {the criterion being a learning criterion}
- G05B 13/048: {using a predictor}
- G06N 3/0442: characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU]
- G06N 3/09: Supervised learning
- G06Q 10/06375: {Prediction of business process outcome or impact based on a proposed change}
- G06Q 50/04: Manufacturing
- G06Q 50/06: Energy or water supply
- H02J 3/28: Arrangements for balancing of the load in networks by storage of energy
- H02J 3/32: using batteries or super capacitors with converting means
- H02J 3/381: {Dispersed generators}
- G06N 3/044: Recurrent networks, e.g. Hopfield networks
- G06N 5/01: Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound
- G06N 20/10: using kernel methods, e.g. support vector machines [SVM]
- G06N 20/20: Ensemble learning
- H02J 2101/20: Dispersed power generation using renewable energy sources
- H02J 2101/40: Hybrid power plants, i.e. a plurality of different generation technologies being operated at one power plant
- H02J 2103/30: Simulating, planning, modelling, reliability check or computer assisted design [CAD] of electric power networks
- H02J 2105/10: Local stationary networks having a local or delimited stationary reach
- H02J: ELECTRIC POWER NETWORKS; CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- G05B: CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS (systems for controlling or regulating non-electric variables G05D; systems for regulating electric or magnetic variables G05F; control devices or systems insofar as characterised by mechanical features only G05G)
- 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

## Abstract

There is provided a method of determining and utilizing predicted available power resources from one or more renewable power sources for one or more industrial gas plants comprising one or more storage resources. The method is executed by at least one hardware processor and comprises: obtaining historical time-dependent environmental data associated with the one or more renewable power sources; obtaining historical time-dependent operational characteristic data associated with the one or more renewable power sources; training a machine learning model based on the historical time-dependent environmental data and the historical time-dependent operational characteristic data; executing the trained machine learning model to predict available power resources for the one or more industrial gas plants for a pre-determined future time period; and controlling the one or more industrial gas plants in response to the predicted available power resources for the pre-determined future time period.

## Claims

### Claim 1

A method of determining and utilizing predicted available power resources from one or more renewable power sources for one or more industrial gas plants comprising one or more storage resources, the method executed by at least one hardware processor, the method comprising:
obtaining historical time-dependent environmental data associated with the one or more renewable power sources;
obtaining historical time-dependent operational characteristic data associated with the one or more renewable power sources;
training a machine learning model based on the historical time-dependent environmental data and the historical time-dependent operational characteristic data;
executing the trained machine learning model to predict available power resources for the one or more industrial gas plants for a pre-determined future time period; and
controlling the one or more industrial gas plants in response to the predicted available power resources for the pre-determined future time period.

### Claim 2 (depends on 1)

The method of claim 1, wherein controlling the one or more industrial gas plants comprises maximizing the usage of the predicted available power resources for the pre-determined future time period.

### Claim 3 (depends on 1)

The method of claim 1, wherein the storage resources comprise one or more industrial gas storage vessels and/or one or more energy storage resources.

### Claim 4 (depends on 3)

The method of claim 3, wherein the one or more energy storage resources comprises one or more of: battery energy storage systems; compressed air energy storage; liquid air energy storage; or pumped hydroelectric energy storage.

### Claim 5 (depends on 3)

The method of claim 3, wherein maximizing the usage of the predicted power resources further comprises controlling the utilization of the industrial gas storage vessels and/or one or more energy storage resources in response to the predicted available power resources.

### Claim 6 (depends on 5)

The method of claim 5, wherein controlling the utilization comprises utilizing an algorithm to select one or more storage resources from a group of storage resources for a given pattern of predicted power availability as a function of time.

### Claim 7 (depends on 6)

The method of claim 6, wherein selection of storage resources is based on physical characteristics of the storage resources.

### Claim 8 (depends on 1)

The method of claim 1, wherein the one or more renewable power sources comprise one or more of: solar power sources; wind power sources; tidal; hydro power; or geothermal power sources

### Claim 9 (depends on 1)

The method of claim 1, wherein the environmental data is selected from one or more of: wind speed; cloud cover; precipitation; humidity; air temperature; atmospheric pressure; solar intensity; and tide times.

### Claim 10 (depends on 1)

The method of claim 1, wherein the operational characteristic data comprises power output from the one or more renewable power sources.

### Claim 11 (depends on 1)

The method of claim 1, wherein the step of training the machine learning model is carried out periodically at a pre-determined training time.

### Claim 12 (depends on 11)

The method of claim 11, wherein at the training time the machine learning model is trained based on historical time-dependent environmental data and the historical time-dependent operational characteristic data obtained within one or more pre-determined historical time windows.

### Claim 13 (depends on 11)

The method of claim 11, further comprising comparing the value of the predicted power resources for a pre-determined future time period with the actual power resources at the end of the predicted period to generate a prediction error value.

### Claim 14 (depends on 13)

The method of claim 13, wherein the pre-determined training time is selected when the prediction error value exceeds a pre-determined threshold.

### Claim 15 (depends on 14)

The method of claim 14, wherein the pre-determined training time is selected based on a pre-determined empirical interval unless the prediction error value exceeds the pre-determined threshold within the pre-determined empirical interval.

### Claim 16 (depends on 1)

The method of claim 1, wherein the one or more industrial gas plants comprise a Hydrogen production plant comprising at least one electrolyzer.

### Claim 17 (depends on 16)

The method of claim 16, wherein the one or more industrial gas plants comprise an Ammonia production plant complex including the Hydrogen production plant.

### Claim 18 (depends on 1)

The method of claim 1, wherein the machine learning model comprises one or more of: Gradient boosting algorithm; Long short-term memory (LSTM) algorithm; support vector machine (SVM) algorithm; or random decision forest algorithm.

### Claim 19

A system for determining and utilizing predicted available power resources from one or more renewable power sources for one or more industrial gas plants comprising one or more storage resources, the system comprising:
at least one hardware processor operable to perform:
obtaining historical time-dependent environmental data associated with the one or more renewable power sources;
obtaining historical time-dependent operational characteristic data associated with the one or more renewable power sources;
training a machine learning model based on the historical time-dependent environmental data and the historical time-dependent operational characteristic data;
executing the trained machine learning model to predict available power resources for the one or more industrial gas plants for a pre-determined future time period; and
controlling the one or more industrial gas plants in response to the predicted available power resources for the pre-determined future time period.

### Claim 20

A computer readable storage medium storing a program of instructions executable by a machine to perform a method of determining and utilizing predicted available power resources from one or more renewable power sources for one or more industrial gas plants comprising one or more storage resources, the method comprising:
obtaining historical time-dependent environmental data associated with the one or more renewable power sources;
obtaining historical time-dependent operational characteristic data associated with the one or more renewable power sources;
training a machine learning model based on the historical time-dependent environmental data and the historical time-dependent operational characteristic data;
executing the trained machine learning model to predict available power resources for the industrial gas plant for a pre-determined future time period; and
controlling the one or more industrial gas plants in response to the predicted available power resources for the pre-determined future time period.
