# US 2024/0005147 A1 · Computer surrogate model to predict the single-phase mixing quality in steady state mixing tanks

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

- Number: US 2024/0005147 A1
- Document ID: US20240005147A1
- Status: Published application
- Kind: A1
- Application number: 18/038,041
- Filed: 2021-09-29
- Published: 2024-01-04
- Earliest claimed: 2020-11-24
- Assignees: Amgen Inc.
- Applicants: Amgen Inc.
- Inventors: Emad Ghadirian Hoseinabadi; William Leigh Johnson
- Web page: https://patentgrep.com/patent/US20240005147A1

## CPC

- G06F 30/28: using fluid dynamics, e.g. using Navier-Stokes equations or computational fluid dynamics [CFD]
- B01F 27/808: with stirrers driven from the bottom of the receptacle
- B01F 35/2206: {Use of stored recipes for controlling the computer programs, e.g. for manipulation, handling, production or composition in mixing plants (B01F33/846 takes precedence)}
- B01F 35/717: {characterised by the means for feeding the components to the mixer}
- G06F 30/30: Circuit design
- G06N 3/0499: Feedforward networks
- G06N 3/08: Learning methods
- G06N 3/09: Supervised learning
- B01F 35/55: {Baffles; Flow breakers (B01F35/531 takes precedence)}
- B01F 2215/0404: theories or general explanations of phenomena associated with mixing or generalizations of a concept by comparison of equivalent methods
- B65G 23/08: with self-contained driving mechanisms, e.g. motors and associated gearing
- G06N 3/04: Architecture, e.g. interconnection topology
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- B01F: MIXING, e.g. DISSOLVING, EMULSIFYING OR DISPERSING (mixing paints B44D3/06)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- B65G: TRANSPORT OR STORAGE DEVICES, e.g. CONVEYORS FOR LOADING OR TIPPING, SHOP CONVEYOR SYSTEMS OR PNEUMATIC TUBE CONVEYORS (packaging B65B; handling thin or filamentary materials, e.g. paper sheets or thread, B65H; cranes B66C; portable mobile or lifting or hauling appliances, e.g. hoists, B66D; devices for lifting or lowering goods for loading or unloading purposes, e.g. fork-lift trucks, B66F9/00; emptying bottles, jars, casks, barrels or similar containers, not otherwise provided for, B67C9/00; delivering or transferring liquids B67D; filling or discharging vessels for liquefied, solidified or compressed gases F17C; pipe-line systems for fluids F17D)

## Abstract

Systems and methods of using a surrogate machine learning model, based on a CFD model, to predict the mixing quality in steady state mixing tanks are provided. An exemplary method includes generating a plurality of training CFD models for a plurality of training steady state mixing configurations based on a plurality of steady state mixing factors associated with each training steady state mixing configuration; calculating a mixing quality for each training steady state mixing configuration using each respective training CFD model; generating a training dataset that includes the steady state mixing factors associated with each training steady state mixing configuration, and the calculated mixing quality for each training steady state mixing configuration; and training a machine learning model, using the training dataset, to predict mixing qualities for steady state mixing configurations based on based on steady state mixing factors associated with the steady state mixing configurations.

## Claims

### Claim 1

A method, comprising:
generating, by one or more processors, a plurality of training computational fluid dynamic (CFD) models for a plurality of training steady state mixing configurations in which inlet streams are mixed in tanks, wherein each training CFD model is generated based on a plurality of steady state mixing factors associated with each training steady state mixing configuration;
calculating, by the one or more processors, a mixing quality for each training steady state mixing configuration using each respective training CFD model;
generating, by the one or more processors, a training dataset that includes the steady state mixing factors associated with each training steady state mixing configuration, and the calculated mixing quality for each training steady state mixing configuration;
training, by the one or more processors, a machine learning model, using the training dataset, to predict mixing qualities for steady state mixing configurations based on based on steady state mixing factors associated with the steady state mixing configurations;
recommending, by the one or more processors, one or more of a working volume or an impeller speed for a given product based on the trained machine learning model.

### Claim 2 (depends on 1)

The method of claim 1, further comprising:
applying, by the one or more processors, the trained machine learning model to new steady state mixing factors associated with a new steady state mixing configuration; and
predicting, by the one or more processors, based on applying the trained machine learning model to the steady state mixing factors associated with the new steady state mixing configuration, a mixing quality for the new steady state mixing configuration.

### Claim 3 (depends on 1)

The method of claim 1, wherein the steady state mixing factors include one or more of: tank geometry, stirrer geometry, working volume, inlet configuration, outlet configuration, inlet flow rates for each inlet, outlet flow rates for each outlet, agitation speed, impeller speed, fluid Reynolds number for each substance, and other chemical and pharmaceutical properties for each substance.

### Claim 4 (depends on 1)

The method of claim 1, wherein the mixing quality is a measure of standard deviation of trace concentration in the tank.

### Claim 5 (depends on 1)

The method of claim 1, further comprising:
generating, by the one or more processors, a testing computational fluid dynamic (CFD) model for a testing steady state mixing configuration in which inlet streams are mixed in tanks, wherein the testing CFD model is generated based on a plurality of steady state mixing factors associated with the testing steady state mixing configuration;
calculating, by the one or more processors, a mixing quality for the testing steady state mixing configuration using the testing CFD model;
applying, by the one or more processors, the trained machine learning model to the steady state mixing factors associated with the testing steady state mixing configuration;
predicting, by the one or more processors, based on applying the trained machine learning model to the steady state mixing factors associated with the testing steady state mixing configuration, a quality of mixing for the testing steady state mixing configuration; and
evaluating, by the one or more processors, the trained machine learning model by comparing the mixing quality calculated for the testing steady state mixing configuration using the testing CFD model and the mixing quality predicted for the testing steady state mixing configuration using the trained machine learning model.

### Claim 6 (depends on 1)

The method of claim 1, wherein the machine learning model is a deep learning model.

### Claim 7

A computer system, comprising:
one or more processors; and
a non-transitory program memory communicatively coupled to the one or more processors and storing executable instructions that, when executed by the one or more processors, cause the processors to:
generate a plurality of training computational fluid dynamic (CFD) models for a plurality of training steady state mixing configurations in which inlet streams are mixed in tanks, wherein each training CFD model is generated based on a plurality of steady state mixing factors associated with each training steady state mixing configuration;
calculate a mixing quality for each training steady state mixing configuration using each respective training CFD model;
generate a training dataset that includes the steady state mixing factors associated with each training steady state mixing configuration, and the calculated mixing quality for each training steady state mixing configuration;
train a machine learning model, using the training dataset, to predict mixing qualities for steady state mixing configurations based on based on steady state mixing factors associated with the steady state mixing configurations; and
recommend one or more of a working volume or an impeller speed for a given product based on the trained machine learning model.

### Claim 8 (depends on 7)

The computer system of claim 7, wherein a first set of one or more processors, of the one or more processors, generate the plurality of training computational fluid dynamic (CFD) models, and wherein a second set of one or more processors, of the one or more processors, train the machine learning model.

### Claim 9 (depends on 7)

The computer system of claim 7, wherein the executable instructions, when executed by the one or more processors, further cause the processors to:
apply the trained machine learning model to new steady state mixing factors associated with a new steady state mixing configuration; and
predict, based on applying the trained machine learning model to the steady state mixing factors associated with the new steady state mixing configuration, a mixing quality for the new steady state mixing configuration.

### Claim 10 (depends on 9)

The computer system of claim 9, wherein a third set of one or more processors, of the one or more processors, apply the trained machine learning model to the new steady state mixing factors associated with the new steady state mixing configuration and predict the mixing quality for the new steady state mixing configuration.

### Claim 11 (depends on 7)

The computer system of claim 7, wherein the steady state mixing factors include one or more of: tank geometry, stirrer geometry, working volume, inlet configuration, outlet configuration, inlet flow rates for each inlet, outlet flow rates for each outlet, agitation speed, impeller speed, fluid Reynolds number for each substance, and other chemical and pharmaceutical properties for each substance.

### Claim 12 (depends on 7)

The computer system of claim 7, wherein the mixing quality is a measure of standard deviation of trace concentration in the tank.

### Claim 13 (depends on 7)

The computer system of claim 7, wherein the executable instructions, when executed by the one or more processors, further cause the processors to:
generate a testing computational fluid dynamic (CFD) model for a testing steady state mixing configuration in which inlet streams are mixed in tanks, wherein the testing CFD model is generated based on a plurality of steady state mixing factors associated with the testing steady state mixing configuration;
calculate a mixing quality for the testing steady state mixing configuration using the testing CFD model;
apply the trained machine learning model to the steady state mixing factors associated with the testing steady state mixing configuration;
predict, based on applying the trained machine learning model to the steady state mixing factors associated with the testing steady state mixing configuration, a quality of mixing for the testing steady state mixing configuration; and
evaluate the trained machine learning model by comparing the mixing quality calculated for the testing steady state mixing configuration using the testing CFD model and the mixing quality predicted for the testing steady state mixing configuration using the trained machine learning model.

### Claim 14 (depends on 7)

The computer system of claim 7, wherein the machine learning model is a deep learning model.

### Claim 15

A non-transitory computer readable storage medium storing computer-readable instructions that, when executed by one or more processors, cause the one or more processors to:
generate a plurality of training computational fluid dynamic (CFD) models for a plurality of training steady state mixing configurations in which inlet streams are mixed in tanks, wherein each training CFD model is generated based on a plurality of steady state mixing factors associated with each training steady state mixing configuration;
calculate a mixing quality for each training steady state mixing configuration using each respective training CFD model;
generate a training dataset that includes the steady state mixing factors associated with each training steady state mixing configuration, and the calculated mixing quality for each training steady state mixing configuration;
train a machine learning model, using the training dataset, to predict mixing qualities for steady state mixing configurations based on based on steady state mixing factors associated with the steady state mixing configurations; and
recommend one or more of a working volume or an impeller speed for a given product based on the trained machine learning model.

### Claim 16 (depends on 15)

The non-transitory computer readable storage medium of claim 15, wherein the computer-readable instructions, when executed by the one or more processors, further cause the processors to:
apply the trained machine learning model to new steady state mixing factors associated with a new steady state mixing configuration; and
predict, based on applying the trained machine learning model to the steady state mixing factors associated with the new steady state mixing configuration, a mixing quality for the new steady state mixing configuration.

### Claim 17 (depends on 15)

The non-transitory computer readable storage medium of claim 15, wherein the steady state mixing factors include one or more of: tank geometry, stirrer geometry, working volume, inlet configuration, outlet configuration, inlet flow rates for each inlet, outlet flow rates for each outlet, agitation speed, impeller speed, fluid Reynolds number for each substance, and other chemical and pharmaceutical properties for each substance.

### Claim 18 (depends on 15)

The non-transitory computer readable storage medium of claim 15, wherein the mixing quality is a measure of standard deviation of trace concentration in the tank.

### Claim 19 (depends on 15)

The non-transitory computer readable storage medium of claim 15, wherein the computer-readable instructions, when executed by the one or more processors, further cause the processors to:
generate a testing computational fluid dynamic (CFD) model for a testing steady state mixing configuration in which inlet streams are mixed in tanks, wherein the testing CFD model is generated based on a plurality of steady state mixing factors associated with the testing steady state mixing configuration;
calculate a mixing quality for the testing steady state mixing configuration using the testing CFD model;
apply the trained machine learning model to the steady state mixing factors associated with the testing steady state mixing configuration;
predict, based on applying the trained machine learning model to the steady state mixing factors associated with the testing steady state mixing configuration, a quality of mixing for the testing steady state mixing configuration; and
evaluate the trained machine learning model by comparing the mixing quality calculated for the testing steady state mixing configuration using the testing CFD model and the mixing quality predicted for the testing steady state mixing configuration using the trained machine learning model.

### Claim 20 (depends on 15)

The non-transitory computer readable storage medium of claim 15, wherein the machine learning model is a deep learning model.
