MEMBRANE FOR SEPARATION PROCESS

CROSS-REFEENCE TO RELATED APPLICATIONS

[0001] This application claims benefit of United States Provisional Patent Application Serial No. 62/711, 134 filed July 27, 2018, the entirety of which is incorporated herein by reference.

BACKGROUND

[0002] Membrane technology is an efficient tool in gas separation. Operational performance of gas separation membranes, including the selectivity and total throughput of the membrane, changes over time, due to factors such as physical aging, plasticization, and mechanical aging. Membranes are used in numerous fields such as oil extraction, gas separation, reverse osmosis, wastewater treatment, fuel cells, and medical applications.

[0003] Membranes are consumable components of purification and separation systems, undergoing replacement when performance and/or throughput crosses a predetermined threshold. Membrane replacement at regularly scheduled intervals contributes to excess operational costs because in some cases the membranes are replaced before replacement is needed. There is a need for methods of accurately determining the best time to replace membranes.

BRIEF DESCRIPTION OF THE DRAWINGS

[0004] Aspects of the present disclosure are best understood from the following detailed description when read with the accompanying figures. It is noted that, in accordance with the standard practice in the industry, various features are not drawn to scale. In fact, the dimensions of the various features may be arbitrarily increased or reduced for clarity of discussion. [0005] Figure l is a schematic diagram of a membrane separation system, in accordance with some embodiments.

[0006] Figure 2 is a flow diagram of a method of estimating remaining useful life of a membrane, in accordance with some embodiments.

[0007] Figure 3 is a chart of a membrane separation system performance characteristic, according to an embodiment.

DETAILED DESCRIPTION

[0008] The following disclosure provides many different embodiments, or examples, for implementing different features of the provided subject matter. Specific examples of components, values, operations, and so forth, are described below to simplify the present disclosure, but these examples are not intended to be limiting. Other values, operations, arrangements, and so forth, are contemplated. The present disclosure may repeat reference numerals and/or letters in the various examples. This repetition is for the purpose of simplicity and clarity and does not in itself dictate a relationship between the various embodiments and/or configurations discussed.

[0009] Membrane technology has proven to be an efficient process for separating gaseous materials. Membranes are used in separation of liquid materials and gaseous materials that are otherwise difficult or impossible separate. Membrane technologies are used in the oil industry, medical industry applications, water filtration, and so forth. During operation, membranes age and undergo changes in performance. Changes in membrane performance are related to factors such as physical aging, mechanical aging, and plasticization of the membranes. In some cases, the changes degrade the performance of the membrane. During operation of a membrane-based separator, in the absence of a catastrophically damaging event, the decision to replace a membrane is typically made based on a comparison between key performance indicators (KPIs) and corresponding predefined membrane replacement criteria. When membrane performance either exceeds or falls below predefined performance specifications, a membrane is replaced. However, membrane change rates are not uniform among membranes that perform a same type of separation. Individual variation between membranes results in performance lifetimes ranging from several months to many years before replacement is indicated based on predefined membrane replacement criteria.

[0010] In the present disclosure, a method is proposed to estimate remaining useful lifetime (RUL) of a gas separation membrane unit to achieve membrane replacement in a prognostic or predictive manner, rather than a responsive manner based on performance criteria exceeding a predetermined replacement criterion. The method disclosed herein is applicable to estimates of useful lifetimes of many different types of membranes, not nearly membranes used in gas separation. A characteristic of a membrane separation system amenable to using the method disclosed herein is that real-time performance data of the membrane operation can be recorded during membrane operation for both the feed stream, the permeate stream, and the non-permeate stream associated with a membrane separation system. The remaining useful lifetime determination is performed using a physical model, trained by data acquired through in-situ nondestructive monitoring of membrane operations.

[0011] FIG. 1 is a schematic diagram of a membrane separation system 100. A flow of incoming material, or a feed stream, is admitted through an inlet line 102 into membrane unit 104. In some instances, a flow of incoming material is a gaseous flow. In some instances, a flow of incoming material is a liquid flow. In some instances, a flow of material is a flow of supercritical material. Membrane unit 104 includes at least one membrane for performing a separation process on the components of the incoming material flow, and incudes at least one membrane, a membrane fastening means, and a plurality of sensors configured to detect a condition of the membrane, of the flow of incoming material, or of a flow of material after exposure to the membrane (permeate and non-permeate flows). The flow of incoming material entering the membrane unit 104, is split into a permeate flow that travels through outlet line 106, and a non-permeate flow that travels through outlet line 108. The permeate flow that travels through outlet line 106 is a flow of material that has passed through, or permeated, the membrane in membrane unit 104. The non-permeate flow is material that, having entered the membrane unit 104, is forced out of the membrane unit 104 without having passed through the membrane. The pressure of the incoming material contributes to the formation of the permeate flow in outlet line 106 by applying a physical force on the membrane, keeping a supply of mixed inlet material in contact with the membrane, and promoting e.g., a chemical binding process, physical adhesion process, capillary process, or other physiochemical process associated with separating materials into separated flows using a membrane.

[0012] In some embodiments, the feed stream or the flow of incoming material is a mixture of gaseous components. In some embodiments, the feed stream is at an elevated pressure and or elevated temperature in order to promote separation of components of the gaseous mixture by a membrane in a membrane separation system. In some embodiments, the gaseous components include one or more of carbon dioxide, methane, ethane, propane, butane, isobutane, or other saturated hydrocarbons— for example a wellhead natural gas stream. In some embodiments, the gaseous components include unsaturated hydrocarbons such as ethene, propene, one or more butene isomers, one or more pentene isomers, and so forth. In some embodiments, one or more of the gaseous components includes mixtures of inorganic molecules such as ammonia, hydrogen cyanide, carbon monoxide, carbon dioxide, hydrogen sulfide, and so forth. In some embodiments, the feed stream or flow of incoming material includes liquid components, solutions, azeotropic mixtures, or other fluids amenable to separation using membrane-based separation systems to extract one fraction of the feed stream into a permeate stream and leaving a remainder, or non-permeate stream having a depleted concentration of a component separated into the permeate stream. In some embodiments, fluid feed streams used in membrane separation systems include biological fluids such as blood and/or urine.

[0013] Individual membrane modeling involves the measurement of variables flowrate Q , temperature T, pressure P and compositions (the index number i identifies different components in the stream exiting a membrane separator) in each stream allow the non destructive monitoring of membrane conditions. These variables can be classified into two categories. The first category is feedstock and control variables X. These parameters are determined before membrane gas separation and typically include flowrate, temperature, pressure in a line (e.g., inlet pressure), partial pressure of gaseous components in a line, chemical composition of the inlet flow or feed stream, and so forth. These parameters also include other non-process-defined operational parameters, such as the membrane maintenance or replacement schedule, profile of operators or technicians, supervising the operations, as well as the history of past conditions, and the history of specific membrane units. The second category is outlet variables Ϋ. These parameters represent the results of membrane gas separation, such as pressure of outlet lines (e.g., permeate line pressure, and non-permeate line pressure), flow rates of permeate and non- permeate fractions through outlet lines, chemical composition of the permeate and non- permeate fractions in the outlet lines, and so forth.

[0014] Additionally, the membrane units are characterized by membrane characteristic parameters M including membrane areas S and pressure normalized fluxes of each component P and so forth.

[0015] For individual membrane degradation modeling, given X and M, one obtains Y through a membrane separation model F, given below in Equation 1 :

Ϋ = F(X, M ) (Equation 1)

The function F can be a physical model of the membrane system, a statistical model, a heuristic model, or any type of model that produces useful results. And reversely, one may obtain membrane characteristics M from X and f by the relationship described in Equation 2:

M = F ^~1 (X, Y ) (Equation 2)

Equation 2 provides a mathematical foundation to monitor membrane condition through the measurement of feed, perm and non-perm streams.

[0016] The change in performance of a membrane unit is represented by the variation of M, particularly fluxes P _t of each component, with time. The modelling and forecast of these variations will lead to an estimation of RUL. The workflow to estimate the RUL of a membrane unit is summarized by the method 200 provided in FIG. 2. At operation 202, a membrane separation control system or user obtains the real time acquisition of feedstock and control variables X and outlet variables f as described above. In some embodiments, the feedstock and control variable, and outlet variable, information may also include the history of the current unit, events related to the maintenance, replacement, change of operational conditions, and so forth.

[0017] In some embodiments, a user obtains real time membrane separation characteristic parameters M using the relationship described above in Equation 2 based on the sensor readouts of sensors connected to outlet lines 106 and 108 of the membrane separation system 100 presented in FIG. 1. In some embodiments, membrane performance characteristics identified in outlet lines 106 and 108 include gas composition, gas temperature, gas flow rate, a thermodynamic state of a chemical in the permeate or non-permeate stream, or ratios of one or more membrane performance characteristics.

[0018] The separate event detection model may be deployed to recognize/record different events and assess their impact on membrane lifetime, e.g. reduction in temperature may cause temporary reduction in performance due to condensation. This model may be rule-based or machine learning based. A membrane evolution model is trained by history matching the variation of M, It may also include the history of events and operations, including maintenance and replacement. The membrane evolution model can be established by different machine learning approaches such as linear/non-linear regression, artificial neural network (ANN), etc.

[0019] Method 200 includes operation 204, in which membrane characteristic parameters are estimated using the technique summarized by Equation 2.

[0020] Method 200 includes operation 206, in which a membrane evolution model is developed based on the recorded real-time membrane characteristic parameters collected during operation 202. In some embodiments, a membrane evolution model is determined after recording membrane characteristic parameters in an initial time interval and one or more predetermined evolution models are fitted to the recorded membrane characteristic parameters for evaluation and selection. In some embodiments, a membrane evolution model is developed ab initio based on a lack of predetermined models that fit data collected during the initial time interval. In some embodiments, a model selection for a membrane is modified, subsequent to operation 218, based on a deviation of monitored membrane performance characteristic parameters from an initial model selected in operation 206.

[0021] Method 200 includes an operation 208, wherein a set of future membrane performance characteristic parameters M is generated based on the membrane evolution model developed during operation 206. With future M and X , future membrane outlets F can be estimated using Equation (1), given above.

[0022] Method 200 includes an operation 210, in which feedstock parameters and membrane control system parameters X are estimated for future intervals of membrane operation in order to compensate for anticipated membrane evolution according to the model developed during operation 206.

[0023] Method 200 includes an operation 212, in which a set of threshold values or membrane performance characteristic thresholds are determined based on the real-time membrane performance characteristics of operation 202 and the membrane evolution model of operation 206 (or, the future membrane performance characteristic parameters of operation 208).

[0024] Method 200 includes an operation 214, in which the threshold values, or membrane performance characteristic thresholds, are used to determine an anticipated membrane replacement time according to a time difference between a measurement interval of the real-time membrane characteristic collection performed an operation 202 and an estimated time at which membrane performance characteristics will correspond to the threshold values based on the membrane evolution model. With future F and X , future key performance indicators are estimated and compare with the membrane replacement criteria (e.g., the threshold values of operation 212) to determine the time when the membrane end of life (EOL) is achieved. And therefore RETL is obtained by subtracting the current time interval for measuring real time characteristics from the estimated end of life (EOL) of the membrane. [0025] Method 200 includes an operation 216, in which, at a second time interval after the first time interval wherein real-time membrane characteristic parameters were recorded (e.g., the time interval corresponding to operation 202), additional real-time membrane characteristic parameters are recorded and compared to the threshold values determined in operation 212. When real-time membrane characteristic parameters recorded during operation 206 fall within an anticipated range of membrane characteristic parameters corresponding to normal operation of the membrane, or the membrane separation system, the method continues to operation 218. When real-time membrane characteristic parameters recorded during operation 206 exceed a threshold value, or fall outside of an anticipated range of membrane characteristic parameters corresponding to normal operation of the membrane, the membrane separation system, the method continues to operation 220.

[0026] In operation 220, the membrane separation system is turned off, feed inlet and permeate and non-permeate flows are removed, and the membrane is replaced. In operation 218, the time interval between recording of current membrane performance characteristic parameters and a previous time interval anticipated membrane replacement time is determined. When the time interval of current membrane performance characteristic parameters exceeds the previously anticipated membrane replacement time, membrane performance characteristic parameters are recorded and a user is notified that membrane replacement may be warranted. Upon receiving an input from a user, operation 220 is performed, wherein membrane replacement occurs. In the absence of an input from a user, operation 206 is performed, wherein membrane evolution models are updated based on current membrane performance characteristic parameters for continued remaining useful lifetime determinations as described above. Thus, deviations of membrane characteristic parameters are used to update a membrane evolution model, or coefficients related to the equation (2), to improve accuracy of end-of-life, or remaining useful lifetime, determinations according to the method 200.

[0027] FIG. 3 is a chart 300 representing a performance characteristic value on a first axis 302, measurement time on a 2 ^nd axis 304, and a component of a membrane evolution model 306 corresponding to values of the performance characteristic as a function of time. A first measurement of the performance characteristic is performed and a first observation 308 is recorded and compared to the degradation model 306. Observation 308 is recorded at time 318 on axis 304, and has a value 308V. Observation 310 is recorded at time 310T and has a value 310V. In FIG. 3, observation 310 falls on the model 306 and within threshold values 312 for observations corresponding to normal anticipated operation of the membrane according to the model 306. In FIG. 3, the evolution model 306 is used to estimate the remaining useful lifetime 314 corresponding to an end of life measurement at time of 316T and/or at performance value 316V of the performance characteristic monitored on axis 302.

[0028] The foregoing outlines features of several embodiments so that those skilled in the art may better understand the aspects of the present disclosure. Those skilled in the art should appreciate that they may readily use the present disclosure as a basis for designing or modifying other processes and structures for carrying out the same purposes and/or achieving the same advantages of the embodiments introduced herein. Those skilled in the art should also realize that such equivalent constructions do not depart from the spirit and scope of the present disclosure, and that they may make various changes, substitutions, and alterations herein without departing from the spirit and scope of the present disclosure.