# US 11,992,803 B2 · Method of determining remaining useful lifetime of membrane for separation process > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 11,992,803 B2 - Document ID: US11992803B2 - Status: Granted - Kind: B2 - Application number: 17/262,956 - Filed: 2019-07-29 - Published: 2021-10-07 - Granted: 2024-05-28 - Earliest claimed: 2018-07-27 - Assignees: Cameron International Corporation - Applicants: Schlumberger Technology Corporation - Inventors: Shu Pan; George E. Mahley, III; Atsushi Morisato; Oleg O. Medvedev; Jason M. Dietrich - Web page: https://patentgrep.com/patent/US11992803B2 ## CPC - B01D 53/22: by diffusion (manufacturing semi-permeable membranes B01D67/00; form, structure or properties of semi-permeable membranes B01D69/00; material for semi-permeable membranes B01D71/00) - B01D 53/30: Controlling by gas-analysis apparatus (regulating non electrical variables in general G05D) - G06F 18/214: Generating training patterns; Bootstrap methods, e.g. bagging or boosting - G06N 20/00: Machine learning - B01D 2311/24: Quality control - 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) - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS ## Abstract Membranes used in membrane separation technologies change over time due to changes in physical characteristics of the membrane. Predicting remaining useful lifetime of a membrane is performed by fitting an evolution model of the membrane to real-time performance characteristics recorded for the membrane and by comparing later performance characteristics of the membrane to the evolution model. Updating an evolution model during membrane operation improves estimates of remaining useful membrane lifetime and allows for accurate estimates of estimated membrane end-of-life. ## Claims ### Claim 1 A method, comprising: providing a gas mixture to a membrane separator comprising a membrane; separating the gas mixture into a first gas and a second gas using the membrane separator; measuring a performance characteristic of one or more of the gas mixture, the first gas, or the second gas, the membrane performance characteristic comprising at least one of a gas composition, a gas temperature, a gas flow rate, a thermodynamic state of a chemical in the permeate or non-permeate stream, or ratios thereof; measuring a composition of the gas mixture, the first gas, and the second gas; determining a thermodynamic state of the gas mixture; determining a current state of the membrane based on the performance characteristics and compositions, and the thermodynamic state, using a physical model; predicting a future state of the membrane using a membrane evolution model; and adjusting one of the performance characteristics, or the thermodynamic state, of the gas mixture based on the predicted future state. ### Claim 2 (depends on 1) The method of claim 1, wherein the performance characteristic is a flow rate of one or more of the gas mixture, the first gas, and the second gas. ### Claim 3 (depends on 1) The method of claim 1, further comprising recording, during an initial time interval, the measured performance characteristic of one or more of the gas mixture, the first gas, and the second gas, and selecting, as the membrane evolution model, a first membrane evolution model based on the measured performance characteristic of the initial time interval. ### Claim 4 (depends on 1) The method of claim 1, further comprising measuring the performance characteristic of one or more of the gas mixture, the first gas, and the second gas, during a second time interval and determining whether a current state of the membrane exceeds a performance characteristic threshold of the membrane. ### Claim 5 (depends on 3) The method of claim 3, further comprising recording, during a second time interval, the measured performance characteristic of one or more of the gas mixture, the first gas, and the second gas, and selecting, as the membrane evolution model, a second membrane evolution model based on the measured performance characteristic of the second time interval. ### Claim 6 (depends on 5) The method of claim 5, wherein at least one of the first and the second membrane evolution models is a machine learning model. ### Claim 7 (depends on 1) The method of claim 1, further comprising comparing the predicted future state of the membrane to a standard and adjusting one of the performance characteristics, or the thermodynamic state, of the gas mixture based on the comparison. ### Claim 8 (depends on 7) The method of claim 7, wherein the membrane evolution model is trained using historical data from multiple membranes. ### Claim 9 A method, comprising: providing a gas mixture to a membrane separator comprising a membrane; separating the gas mixture into a first gas and a second gas using the membrane separator; measuring a performance characteristic of one or more of the gas mixture, the first gas, or the second gas, the membrane performance characteristic comprising at least one of a gas composition, a gas temperature, a gas flow rate, a thermodynamic state of a chemical in the permeate or non-permeate stream, or ratios thereof; measuring a composition of the gas mixture, the first gas, and the second gas; determining a thermodynamic state of the gas mixture; determining a current state of the membrane based on the performance characteristics and compositions, and the thermodynamic state, using a physical model; predicting a future state of the membrane using a membrane evolution model selected based on the measured performance characteristic; comparing the predicted future state of the membrane to a standard; and adjusting one of the performance characteristics, or the thermodynamic state, of the gas mixture based on the comparison. ### Claim 10 (depends on 9) The method of claim 9, wherein a first membrane evolution model is selected during a first time period and a second membrane evolution model, different from the first evolution model, is selected during a second time period. ### Claim 11 (depends on 9) The method of claim 9, wherein the selected membrane evolution model is a machine learning model. ### Claim 12 (depends on 9) The method of claim 9, wherein the membrane evolution model is trained using historical data from multiple membranes. ### Claim 13 (depends on 9) The method of claim 9, further comprising comparing the current state of the membrane to a standard and adjusting one of the performance characteristics, or the thermodynamic state, of the gas mixture based on the comparison of the current state of the membrane to the standard. ### Claim 14 (depends on 13) The method of claim 13, wherein a first membrane evolution model is selected during a first time period and a second membrane evolution model, different from the first evolution model, is selected during a second time period. ### Claim 15 A method, comprising: providing a gas mixture to a membrane separator comprising a membrane; separating the gas mixture into a first gas and a second gas using the membrane separator; measuring a flowrate, temperature, pressure, and composition of one or more of the gas mixture, the first gas, or the second gas; determining a thermodynamic state of the gas mixture; determining a current membrane performance characteristic based on the flowrates, temperatures, pressures, compositions, and thermodynamic state, using a physical model, the current membrane performance characteristic comprising at least one of a gas composition, a gas temperature, a gas flow rate, a thermodynamic state of a chemical in the permeate or non-permeate stream, or ratios thereof; predicting a future membrane performance characteristic using a membrane evolution model selected based on the current membrane performance characteristic, the future membrane performance characteristic comprising at least one of a gas composition, a gas temperature, a gas flow rate, a thermodynamic state of a chemical in the permeate or non-permeate stream, or ratios thereof; comparing the predicted future membrane performance characteristic to a standard; and adjusting one of the flowrates, temperatures, pressures, compositions, or the thermodynamic state of the gas mixture based on the comparison. ### Claim 16 (depends on 15) The method of claim 15, wherein a first membrane evolution model is selected during a first time period and a second membrane evolution model, different from the first evolution model, is selected during a second time period. ### Claim 17 (depends on 16) The method of claim 16, wherein the selected membrane evolution model is a machine learning model. ### Claim 18 (depends on 17) The method of claim 17, wherein the membrane evolution model is trained using historical data from multiple membranes.