# US 2023/0398498 A1 · Systems, methods, and devices for optimization of ultra filtration membrane performance for water treatment using artificial intelligence and optimization algorithms

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

- Number: US 2023/0398498 A1
- Document ID: US20230398498A1
- Status: Published application
- Kind: A1
- Application number: 17/749,814
- Filed: 2022-05-20
- Published: 2023-12-14
- Earliest claimed: 2022-05-20
- Assignees: Yokogawa Electric Corporation
- Applicants: Yokogawa Electric Corporation
- Inventors: Mahdi FARAMARZI; Gokula Krishnan SIVAPRAKASAM; Kaushik GHOSH; Yasuhiro MATSUI
- Web page: https://patentgrep.com/patent/US20230398498A1

## CPC

- B01D 65/02: Membrane cleaning or sterilisation {; Membrane regeneration}
- B01D 61/145: {Ultrafiltration}
- B01D 2321/16: Use of chemical agents
- B01D 2321/40: Automatic control of cleaning processes
- B01D: SEPARATION (separating solids from solids by wet methods B03B, B03D; by pneumatic jigs or tables B03B; by other dry methods B07; magnetic or electrostatic separation of solid materials from solid materials or fluids, separation by high-voltage electric fields B03C; centrifuges, vortex apparatus B04; presses per se for squeezing-out liquid from liquid-containing material B30B9/02; treatment of water C02F, e.g. softening by ion-exchange C02F1/42; {arrangements of air intake cleaners in gas turbine plants F02C7/05} ; arrangements or mounting of filters in air-conditioning, air-humidification or ventilation F24F13/28)

## Abstract

A method of implementing membrane cleaning is provided. The method includes accessing a membrane resistance record, where the membrane resistance record includes a plurality of membrane resistance curves, each of the plurality of membrane resistance curves includes a curve profile after a cleaning. The method further includes performing a first cleaning, determining a first membrane resistance after the first cleaning, determining a current position on the curve profile based on the first membrane resistance and a time period from a previous cleaning, and performing a second cleaning. The second cleaning may include a recovery cleaning when the current position on the curve profile exceeds a predetermined range, and a maintenance cleaning when the current position on the curve profile is within the predetermined range.

## Claims

### Claim 1

A method of implementing membrane cleaning, the method comprising:
accessing a membrane resistance record, wherein the membrane resistance record comprises a plurality of membrane resistance curves that form a curve profile that predicts membrane resistance of a membrane during a membrane operation mode, after a maintenance cleaning, and after a recovery cleaning based on computer models trained with data associated with the membrane operation mode, the maintenance cleaning, and the recovery cleaning;
performing, by equipment, a first cleaning of the membrane;
generating, by at least one sensor, sensor data indicative of a first membrane resistance of the membrane after the first cleaning;
determining, based on the sensor data from the at least one sensor, the first membrane resistance;
determining a current position on the curve profile based on the first membrane resistance; and
performing, by the equipment, a second cleaning of the membrane, wherein the second cleaning includes a recovery cleaning of the membrane when the current position on the curve profile exceeds a predetermined membrane resistance range, and a maintenance cleaning of the membrane when the current position on the curve profile is within the predetermined membrane resistance range.

### Claim 2 (depends on 1)

The method of claim 1, wherein determining the first membrane resistance after the first cleaning is based on first data associated with the first cleaning, wherein the first data includes information associated with at least one of: a membrane resistance of the membrane before the first cleaning, chemical concentrations associated with the first cleaning, flux associated with the first cleaning, and total permeate volume associated with the first cleaning.

### Claim 3 (depends on 1)

The method of claim 1, wherein the predetermined membrane resistance range is determined based on at least one of transmembrane pressure (TMP) asserted on a membrane and flux information associated with the first cleaning.

### Claim 4 (depends on 1)

The method of claim 1, further comprising determining a second membrane resistance of the membrane after the second cleaning based on second data associated with the second cleaning, wherein the second data includes membrane resistance of the membrane before the second cleaning, chemical concentrations associated with the second cleaning, flux associated with the second cleaning, and total permeate volume associated with the second cleaning.

### Claim 5 (depends on 1)

The method of claim 1, further comprising:
training the computer models with the data associated with the membrane operation mode, the maintenance cleaning, and the recovery cleaning; and
generating the membrane resistance record based on the trained computer models.

### Claim 6 (depends on 1)

The method of claim 1, wherein each of the plurality of membrane resistance curves is obtained by calculating membrane resistance of the membrane at a flux at a predetermined time interval, and by using a regression model.

### Claim 7 (depends on 1)

The method of claim 1, further comprising simulating a plurality of membrane resistance records for a predetermined future period based on the first membrane resistance, a second membrane resistance of the membrane determined after the second cleaning, and the plurality of membrane resistance curves.

### Claim 8 (depends on 7)

The method of claim 7, further comprising determining a chemical cleaning implementation for the predetermined future period based on the plurality of membrane resistance records, wherein the chemical cleaning implementation includes a chemical cleaning schedule and chemical cleaning concentrations.

### Claim 9 (depends on 8)

The method of claim 8, wherein determining the chemical cleaning implementation for the predetermined future period includes:
performing use optimization associated with each of the plurality of membrane resistance records; and
determining the chemical cleaning implementation based on comparison of a cost associated with each of the plurality of membrane resistance records.

### Claim 10 (depends on 9)

The method of claim 9, wherein performing the use optimization associated with each of the plurality of membrane resistance records is based on a set of considerations associated with the first cleaning, the second cleaning, and the chemical cleaning implementation for the predetermined future period, wherein the set of considerations comprises, for each of the cleanings:
a first consideration associated with energy consumption during the membrane operation mode; and
a second consideration including at least one of energy consumption and chemical costs associated with a chemical cleaning concentration.

### Claim 11 (depends on 1)

The method of claim 1, wherein the first cleaning includes a maintenance cleaning or a recovery cleaning.

### Claim 12 (depends on 1)

The method of claim 1, wherein each of the first membrane resistance and a second membrane resistance of the membrane determined after the second cleaning includes at least one of a membrane fouling resistance and a membrane initial resistance.

### Claim 13

A device comprising:
a processor; and
a memory coupled with the processor, wherein the memory stores instructions that, when executed by the processor, enables the processor to:
access a membrane resistance record, wherein the membrane resistance record comprises a plurality of membrane resistance curves that form a curve profile that predicts membrane resistance of a membrane during a membrane operation mode, after a maintenance cleaning, and after a recovery cleaning based on computer models trained with data associated with the membrane operation mode, the maintenance cleaning, and the recovery cleaning;
send a control signal to equipment that instructs the equipment to perform a first cleaning, wherein the first cleaning includes a maintenance cleaning or a recovery cleaning;
receive, from at least one sensor over a communication network, sensor data indicative of a first membrane resistance of the first membrane after the first cleaning; determine, based on the received sensor data, the first membrane resistance;
determine a current position on the curve profile based on the first membrane resistance; and
send a control signal to the equipment that instructs the equipment to perform a second cleaning, wherein the second cleaning includes:
a recovery cleaning when the current position on the curve profile exceeds a predetermined membrane resistance range; and
a maintenance cleaning when the current position on the curve profile is within the predetermined membrane resistance range.

### Claim 14 (depends on 13)

The device of claim 13, wherein determining the first membrane resistance after the first cleaning is based on first data associated with the first cleaning, wherein the first data includes information associated with at least one oft a membrane resistance of the membrane before the first cleaning, chemical concentrations, flux associated with the first cleaning, and total permeate volume associated with the first cleaning.

### Claim 15 (depends on 13)

The device of claim 13, wherein the predetermined membrane resistance range is determined based on at least one of transmembrane pressure (TMP) asserted on a membrane and flux information associated with the first cleaning.

### Claim 16 (depends on 13)

The device of claim 13, wherein the instructions are further executable by the processor to determine a second membrane resistance of the membrane after the second cleaning based on second data associated with the second cleaning, wherein the second data includes membrane resistance of the membrane before the second cleaning, chemical concentrations associated with the second cleaning, flux associated with the second cleaning, and total permeate volume associated with the second cleaning.

### Claim 17 (depends on 13)

The device of claim 13, wherein the instructions are further executable by the processor to:
train the computer models with the data associated with the membrane operation mode, the maintenance cleaning, and the recovery cleaning; and
generate the membrane resistance record based on the trained computer models.

### Claim 18 (depends on 13)

The device of claim 13, wherein each of the plurality of membrane resistance curves is obtained by calculating the membrane resistance at a flux at a predetermined time interval, and using a regression model to determine each of the plurality of membrane resistance curves.

### Claim 19 (depends on 13)

The device of claim 13, wherein the instructions are further executable by the processor to simulate a plurality of membrane resistance records for a predetermined future period based on the first membrane resistance, a second membrane resistance of the membrane determined after the second cleaning, and the plurality of membrane resistance curves.

### Claim 20

A system comprising:
a processor; and
a memory coupled with the processor, wherein the memory stores instructions that, when executed by the processor, enables the processor to:
access a membrane resistance record, wherein the membrane resistance record comprises a plurality of membrane resistance curves that form a curve profile that predicts membrane resistance of a membrane during a membrane operation mode, after a maintenance cleaning, and after a recovery cleaning based on computer models trained with data associated with the membrane operation mode, the maintenance cleaning, and the recovery cleaning;
send a control signal to equipment that instructs the equipment to perform a first cleaning;
receive, from at least one sensor over a communication network, sensor data indicative of a first membrane resistance of the first membrane after the first cleaning;
determine, based on the sensor data, the first membrane resistance of the membrane;
determine a current position on the curve profile based on the first membrane resistance; and
send a control signal to the equipment that instructs the equipment to perform a second cleaning, wherein the second cleaning includes:
a recovery cleaning when the current position on the curve profile exceeds a predetermined membrane resistance range; and
a maintenance cleaning when the current position on the curve profile is within the predetermined membrane resistance range,
wherein each of the first membrane resistance and a second membrane resistance of the membrane determined after the second cleaning includes at least one of a membrane fouling resistance and a membrane initial resistance.
