# US 2026/0094677 A1 · Methods of predicting properties of a chemical system using surrogate models

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

- Number: US 2026/0094677 A1
- Document ID: US20260094677A1
- Status: Published application
- Kind: A1
- Application number: 19/112,646
- Filed: 2023-09-19
- Published: 2026-04-02
- Earliest claimed: 2022-09-19
- Assignees: Cameron International Corporation
- Applicants: Cameron International Corporation
- Inventors: Sergey MAKARYCHEV-MIKHAILOV; Jesse FARRELL
- Web page: https://patentgrep.com/patent/US20260094677A1

## CPC

- E21B 43/25: Methods for stimulating production {(by vibrating earth formations E21B43/003)}
- G16C 20/30: Prediction of properties of chemical compounds, compositions or mixtures
- E21B 41/00: Equipment or details not covered by groups E21B15/00 - E21B40/00
- E21B 43/00: Methods or apparatus for obtaining oil, gas, water, soluble or meltable materials or a slurry of minerals from wells (applicable only to water E03B)
- E21B 49/00: Testing the nature of borehole walls; Formation testing; Methods or apparatus for obtaining samples of soil or well fluids, specially adapted to earth drilling or wells
- G06N 3/08: Learning methods
- G06N 5/01: Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound
- G06N 20/00: Machine learning
- G16C 20/70: Machine learning, data mining or chemometrics
- E21B 2200/20: Computer models or simulations, e.g. for reservoirs under production, drill bits
- E21B 2200/22: Fuzzy logic, artificial intelligence, neural networks or the like
- E21B: EARTH OR ROCK DRILLING; OBTAINING OIL, GAS, WATER, SOLUBLE OR MELTABLE MATERIALS OR A SLURRY OF MINERALS FROM WELLS
- G16C: COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

Methods of predicting physicochemical properties of a chemical system using a family of surrogate or reduced order models, trained on first principle simulation results. The models are created using machine learning techniques. The chemical system can be a complex multicomponent and multiphase system such as produced water.

## Claims

### Claim 1

A method of predicting properties of a chemical system, the method comprising:
generating a family of surrogate or reduced order models; and
training the family of surrogate or reduced order models on first principle simulation results.

### Claim 2 (depends on 1)

The method of claim 1, wherein the predicted properties comprise one or more of physicochemical properties, thermodynamic properties, and phase speciation properties.

### Claim 3 (depends on 1)

The method of claim 1, wherein the chemical system comprises water with dissolved species in contact with solid precipitates, gases, and other liquids.

### Claim 4 (depends on 3)

The method of claim 3, wherein the other liquids comprise oil.

### Claim 5 (depends on 1)

The method of claim 1, wherein the family of surrogate models is based on one or more machine learning algorithms.

### Claim 6 (depends on 1)

The method of claim 1, further comprising constructing a training dataset from data subjected to data engineering steps.

### Claim 7 (depends on 6)

The method of claim 6, wherein the data engineering steps comprise one or more of sampling, feature engineering and pre-processing, outlier removal, and target transformations.

### Claim 8 (depends on 1)

The method of claim 1, further comprising using the surrogate models as stand-alone simulators, incorporating the surrogate models into user applications, or deploying the surrogate models on edge devices on or near sensors and/or analytical instruments.

### Claim 9 (depends on 8)

The method of claim 8 further comprising using multiple simulators to form a range of family of results.

### Claim 10 (depends on 9)

The method of claim 9 wherein the range of family of results comprises statistical evaluation of quality of the surrogate models result such as confidence interval or min/max/avg.

### Claim 11 (depends on 10)

The method of claim 10 further comprising using the surrogate models to predict deviations or potential errors in one or more simulators.

### Claim 12 (depends on 1)

The method of claim 1 further comprising using the surrogate models to speed up reaction time of a response to control or to optimize a process.

### Claim 13 (depends on 12)

The method of claim 12 wherein the process comprises dosing of a chemical, adjustment of a pump flow rate, regulation of a pressure, adjustment of fluid temperature within equipment, or actuating a valve.

### Claim 14

A method of predicting physicochemical properties of a chemical system, the method comprising:
training and optimizing reduced order models (ROMs) for one or more target properties using one or more machine learning models; and
predicting the physicochemical properties of the chemical system using the trained and optimized ROMs.

### Claim 15 (depends on 14)

The method of claim 14, further comprising:
screening multiple machine learning models for the one or more target properties using model metrics to compare models; and
selecting one or more of the machine learning models based on the model metrics,
wherein training and optimizing the ROMs comprising training and optimizing the ROMs using the one or more machine learning models selected based on the model metrics.

### Claim 16 (depends on 14)

The method of claim 14, wherein the chemical system comprises produced water.

### Claim 17 (depends on 14)

The method of claim 14, further comprising performing principal component analysis on a representative dataset to produce an engineered dataset.

### Claim 18 (depends on 17)

The method of claim 17, further comprising using the engineered dataset to train the ROMs.

### Claim 19 (depends on 14)

The method of claim 14, further comprising deploying the trained and optimized ROMs in one or more digital workflows.

### Claim 20 (depends on 19)

The method of claim 19, wherein the one or more digital workflows comprise optimizing one or more processes, the one or more processes comprising one or more of dosing of a chemical, adjustment of a pump flow rate, regulation of a pressure, adjustment of fluid temperature within equipment, and actuating a valve.
