Fuel cell purging process optimization method, device, and storage medium

Abstract

Disclosed in the present invention are a fuel cell purging process optimization method, a device, and a storage medium. The method comprises: simulating multiple fuel cell purging processes, obtaining the consumed energy, purging time and high-frequency impedance correction value corresponding to each fuel cell purging process, excluding a fuel cell purging process corresponding to the purging time greater than a maximum purging time or corresponding to the high-frequency impedance correction value less than a target value of high-frequency impedance, selecting, as an optimal fuel cell purging process from among the remaining fuel cell purging processes, a fuel cell purging process corresponding to a minimum value of consumed energy, and using, as operation parameters of an actual fuel cell purging process, the purging time, cathode purging flow, anode purging flow and cell temperature during purging corresponding to the optimal fuel cell purging process. The present invention can improve the purging effect, reduce the energy consumption during purging, reduce the purging energy consumption and purging time as much as possible while satisfying purging performance, and provide an optimal purging method.

Description

Fuel cell purging process optimization method, device and storage medium TECHNICAL FIELD

[0001] The present application relates to a fuel cell purging process optimization method, device and storage medium, belonging to the technical field of fuel cells. BACKGROUND

[0002] Hydrogen energy and fuel cells are major strategic technologies for global energy structure upgrading and power transformation. Fuel cells, as one of the best ways to apply hydrogen energy, have received widespread attention. Fuel cells have outstanding advantages such as low operating temperature, high current density and good stability. Water management of fuel cells is an important way to improve the performance and durability of fuel cells. The gas purging process during shutdown of fuel cells can effectively remove residual water in fuel cells, thereby improving the performance of fuel cells during startup. At the same time, gas purging is also a common method to improve the success possibility of cold start of fuel cells by minimizing the residual water in fuel cells. At present, high-frequency impedance is an important means to characterize the water content in fuel cells, which is often used to evaluate the effect of fuel cell purging process and reflect the trend of internal water content in fuel cell during purging process.

[0003] The existing fuel cell purging method and system provide some fuel cell purging methods, which can improve the purging efficiency of fuel cells to a certain extent. However, the influence of water redistribution in fuel cells during shutdown process and impedance relaxation phenomenon on purging effect is not considered, and there is no accurate basis for judging the purging process. The influence of the shutdown process and the purging process on the purging method is not considered comprehensively, and the energy consumption during the purging process is not considered.

[0004] Therefore, there is an urgent need for the skilled in the art to improve the existing fuel cell purging method. SUMMARY

[0005] Objective: In order to overcome the shortcomings in the prior art, the present application provides a fuel cell purging process optimization method, device and storage medium.

[0006] Technical scheme: In order to solve the above technical problems, the technical scheme adopted by the present application is:

[0007] In the first aspect, a fuel cell purging process optimization method comprises the following steps:

[0008] Step 1: Simulate a one-dimensional purging water removal model using a simulation tool to simulate the values of membrane-bound water, catalyst layer-bound water and purging time corresponding to a fuel cell purging process.

[0009] Step 2: Calculate the average value of membrane-bound water and catalyst layer-bound water as the water content in the membrane.

[0010] Step 3: Calculate the high frequency impedance of the membrane electrode according to the water content in the membrane as the initial impedance value.

[0011] Step 4: Obtain the cathode purge flow rate, anode purge flow rate, cell temperature during purging, storage process environment temperature during shutdown, and time during placement, and take the cathode purge flow rate, anode purge flow rate, cell temperature during purging, storage process environment temperature during shutdown, time during placement, and initial impedance value as the input of the neural network, and the high frequency impedance value is output by the neural network.

[0012] Step 5: Obtain the cell temperature during purging and the high frequency impedance value to calculate the high frequency impedance value correction value.

[0013] Step 6: Calculate the consumption of purging gas and the energy consumption for controlling the cell temperature according to the purging time and the cell temperature during purging, and calculate the consumed energy of the fuel cell purging simulation process according to the consumption of purging gas and the energy consumption for controlling the cell temperature.

[0014] Step 7: Repeat steps 1-6 to simulate multiple fuel cell purging processes, obtain the consumed energy, purging time, and high frequency impedance value correction value corresponding to each fuel cell purging process, exclude the fuel cell purging process corresponding to the purging time greater than the maximum purging time or the high frequency impedance value correction value less than the target value of the high frequency impedance, select the fuel cell purging process corresponding to the minimum consumed energy among the remaining fuel cell purging processes as the optimal fuel cell purging process, and take the purging time, cathode purge flow rate, anode purge flow rate, and cell temperature during purging of the optimal fuel cell purging process as the operation parameters of the actual fuel cell purging process.

[0015] As a preferred solution, the one-dimensional purging water removal model comprises: a model for calculating the change of the water content in the membrane with the purging process, a model for calculating the change of the water content in the catalyst layer with the purging process, a model for calculating the change of the gaseous water content in the catalyst layer with the purging process, a model for calculating the change of the gaseous water content in the gas diffusion layer with the purging process, and a model for calculating the change of the gaseous water content in the flow channel with the purging process.

[0016] As a preferred solution, the model for calculating the change of the water content in the membrane with the purging process has the following formula:

[0017] In the formula, J λ,m-cl represents the diffusion flux of the water content between the membrane and the catalyst layer during purging, λ m represents the water content in the membrane, t represents time, represents the rate of change of the water content in the membrane with the purging time during purging.

[0018] The model for calculating the change of the water content in the catalyst layer with the purging process has the following formula:

[0019] where J cl, mw-vp represents the flux of bound water in the catalyst layer to gaseous water during purging, λ cl represents the content of bound water in the catalyst layer, represents the rate of change of the content of bound water in the catalyst layer with respect to purging time.

[0020] The model for the change in gaseous water in the catalyst layer during purging is given by:

[0021] where c vp,cl is the water vapor concentration in the catalyst layer, w ion represents the concentration of ionomer in the catalyst layer, ε cl represents the porosity of the catalyst layer, p mem is the density of the membrane, EW is the equivalent weight of the membrane, J cl,lq-vp is the flux of liquid water in the catalyst layer to gaseous water during purging, J vp,cl-gdl is the flux of water vapor between the catalyst layer and the gas diffusion layer during purging. represents the rate of change of the content of gaseous water in the catalyst layer with respect to purging time during purging.

[0022] The model for the change in gaseous water in the gas diffusion layer during purging is given by:

[0023] where c vp,gdl is the water vapor concentration in the gas diffusion layer, J gdl,lq-vp is the flux of liquid water to gaseous water in the gas diffusion layer, J vp,gdl-chan is the diffusion flux of water vapor between the gas diffusion layer and the flow channel, s lq,gdl is the volume fraction of liquid water in the gas diffusion layer. represents the rate of change of the content of gaseous water in the gas diffusion layer with respect to purging time during purging.

[0024] The model for the change in gaseous water in the flow channel during purging is given by:

[0025] where A act is the effective active area of the fuel cell, J chan is the flux of water vapor removed by purging in the flow channel of the fuel cell, J represents the rate of change of the content of gaseous water in the flow channel with respect to time during purging.

[0026] As a preferred solution, the high frequency impedance of the membrane electrode is given by:

[0027] where:

[0028] In the formula, R represents the high-frequency impedance of the membrane electrode, δ mem represents the thickness of the membrane, w represents the width of the membrane, and k eff represents the effective conductivity related to the water content of the membrane, and l represents the water content in the membrane.

[0029] As a preferred solution, the high-frequency impedance value correction value is calculated according to the following formula:

[0030] wherein HFR fin represents the high-frequency impedance value correction value, HFR storage represents the high-frequency impedance value, T purge represents the battery temperature during purging.

[0031] As a preferred solution, the neural network is a BP neural network.

[0032] As a preferred solution, the consumed energy is calculated according to the following formula: E x = E cell + E gas

[0033] wherein E x represents the consumed energy, E cell represents the energy consumption for controlling the battery temperature, and E gas represents the consumption of purging gas.

[0034] In a second aspect, a computer readable storage medium has stored thereon a computer program, which, when executed by a processor, implements the method for optimizing a fuel cell purging process according to any one of the first aspect.

[0035] In a third aspect, a computer device comprises:

[0036] a memory for storing instructions.

[0037] a processor for executing the instructions, so that the computer device performs the operations of the method for optimizing a fuel cell purging process according to any one of the first aspect.

[0038] Beneficial effects: The method, device and storage medium for optimizing a fuel cell purging process can optimize the design of the fuel cell purging method, comprehensively consider the fuel cell purging and shutdown storage processes, and consider the energy consumption of the purging process. The method can improve the purging effect, reduce the energy consumption during the purging process, and minimize the energy consumption and purging time under the condition of meeting the purging performance, and proposes an optimal purging method. BRIEF DESCRIPTION OF DRAWINGS

[0039] Fig. 1 is a flowchart of the purging method of the present application. DETAILED DESCRIPTION

[0040] The technical solutions in the embodiments of the present application will be apparently and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments of the present application, all the other embodiments obtained by a person skilled in the art without creative work fall within the protection scope of the present application.

[0041] The present application will be further described below with reference to specific embodiments.

[0042] Embodiment 1

[0043] As shown in Fig. 1, the present embodiment introduces a purging process optimization method for fuel cells, which specifically comprises: purging process simulation, shutdown process prediction, purging process optimization according to the purging simulation results and the shutdown process prediction results.

[0044] Step 1: Purging process simulation

[0045] A one-dimensional purging water removal model of the fuel cell is established, which can accurately calculate the change trend of the membrane water content with time in the purging process. The one-dimensional purging water removal model of the fuel cell includes the models of the change of the membrane bound water λ m , the catalyst layer bound water λ cl and the gaseous water c vp,cl , the gaseous water c vp,gdl in the gas diffusion layer, and the gaseous water c vp,chan in the flow channel with the purging process, which specifically comprises:

[0046] The calculation formula of the change model of the membrane bound water with the purging process is as follows:

[0047] In the formula, J λ,m-cl represents the diffusion flux of the bound water between the membrane and the catalyst layer in the purging process, λ m represents the bound water content in the membrane, t represents time, and dλ cl, / dt is used to represent the change rate of the membrane bound water with the purging time.

[0048] The calculation formula of the change model of the catalyst layer bound water with the purging process is as follows:

[0049] In the formula, J cl, mw-vp represents the flux of the catalyst layer bound water to the gaseous water in the purging process, λ cl represents the catalyst layer bound water content, and dλ cl, / dt is used to represent the change rate of the catalyst layer bound water with the purging time.

[0050] The model equation for the change of gaseous water in the catalyst layer during purging is as follows:

[0051] where c is the water vapor concentration in the catalyst layer, w is the concentration of the ionomer in the catalyst layer, ε is the porosity of the catalyst layer, ρ is the density of the membrane, EW is the equivalent weight of the membrane, J is the flux of liquid water in the catalyst layer converted to gaseous water during purging, J is the flux of water vapor between the catalyst layer and the gas diffusion layer during purging. vp,cl ion cl mem cl,lq-vp vp,cl-gdl is used to represent the rate of change of the gaseous water content in the catalyst layer during purging with respect to the purging time.

[0052] The model equation for the change of gaseous water in the gas diffusion layer during purging is as follows:

[0053] where c is the water vapor concentration in the gas diffusion layer, J is the flux of liquid water in the gas diffusion layer converted to gaseous water during purging, J is the diffusion flux of water vapor between the gas diffusion layer and the flow channel, s is the volume fraction of liquid water in the gas diffusion layer. vp,gdl gdl,lq-vp vp,gdl-chan lq,gdl is used to represent the rate of change of the gaseous water content in the gas diffusion layer during purging with respect to the purging time.

[0054] The model equation for the change of gaseous water in the flow channel during purging is as follows:

[0055] where A is the effective active area of the fuel cell, J is the flux of water vapor purged away in the flow channel of the fuel cell, s is the volume fraction of liquid water in the flow channel. act chan is used to represent the rate of change of the gaseous water content in the flow channel during purging with respect to the time.

[0056] A one-dimensional purging water removal model is simulated using a simulation tool to simulate a purging process of a fuel cell to obtain the values of the bound water in the membrane λ m , the bound water in the catalyst layer λ cl , and the purging time t purge .

[0057] The calculation model of the high-frequency impedance of the membrane electrode after the change in water content outputs the high-frequency impedance value R of the cell after purging as the input HFR(init) of the shutdown prediction process. ​​​​​​​​​​​​

[0058] The calculation model of the high frequency impedance of the membrane electrode is calculated by the following formula:

[0059] In the formula, R represents the high frequency impedance of the membrane electrode, δ mem represents the thickness of the membrane, w represents the width of the membrane, κ eff (λ) represents the effective conductivity related to the water content of the membrane, λ represents the water content in the membrane electrode, and 1 is a dimensionless quantity.

[0060] The average value of λ cl and λ m is calculated as the water content in the membrane λ, and the high frequency impedance value R under different purging times is finally calculated as the input value HFR init in the shutdown prediction process.

[0061] Step 2: Prediction of impedance value in the shutdown storage process:

[0062] Specifically, a neural network model is established to predict the change trend of the high frequency impedance of the fuel cell in the shutdown storage process. In terms of the purging process, as the purging time increases, the water removed from the inside of the fuel cell is also more, and the measured impedance value of the fuel cell is also rising, which generally presents a slow-then-fast trend. The cell temperature, purging gas flow and purging gas temperature during purging have a relatively obvious influence on the purging effect. The shutdown storage environmental parameters also have a very obvious influence on the change of the impedance, including the purging flow, the cell temperature, the storage environment temperature, the high frequency impedance value at the end of purging, etc. The purging conditions and the environmental temperature affect the time required for the fuel cell to reach the final stable state.

[0063] Therefore, the initial impedance value HFR init is obtained, the shutdown storage process environmental temperature T amb is obtained through a temperature acquisition device, the placement process time t in the shutdown storage process is set, the cathode purging flow Q a , the anode purging flow Q c , the BP neural network model is adopted, the cathode purging flow Q a , the anode purging flow Q c , the cell temperature T purge during purging, the shutdown storage process environmental temperature T amb , the initial impedance value HFR init and the placement process time t are selected as the input parameters of the neural network, and the high frequency impedance value HFR storage of the fuel cell is selected as the output parameter of the neural network.

[0064] The high frequency impedance value of the fuel cell is affected by the cell temperature, and the high frequency impedance value HFR of the fuel cell needs to be corrected to eliminate the influence of the set cell temperature T purge at purging on the high frequency impedance value HFR fin The calculation formula is as follows:

[0065] The structure of the BP neural network includes an input layer, a hidden layer, and an output layer. Through the neural network prediction model, the high frequency impedance value HFR fin corresponding to the end of the shutdown storage stage and related to the cell temperature can be calculated as one of the constraint conditions for optimizing the third-stage purging process, according to the high frequency impedance value R at the end of purging as the input of the neural network prediction model.

[0066] Step 3: Optimization of the purging process:

[0067] The consumption E gas of purging gas and the energy consumption E cell for controlling the cell temperature are calculated according to the purging time t purge and the cell temperature T purge at purging, and the consumed energy E x of the first fuel cell purging simulation process is calculated.

[0068] The purging process is optimized with the purging effect HFR fin and the energy consumption E x in the purging process as the optimization objectives. When the purging requirements are met, the purging time is as short as possible, and the additional consumption in the purging process is as little as possible. Therefore, the optimization objectives of the purging parameters are as follows: E x = min(E cell +E gas ) HFR fin ≥ HFR min s.t.t purge ≤ t max

[0069] In the formula, E x is the consumed energy, E cell is the energy consumption for controlling the cell temperature, E gas is the consumption of purging gas, t purge is the purging time, t max is the maximum purging time, and HFR min is the target value of the high frequency impedance.

[0070] According to the purging model, the purging simulation is performed to find the purging operation parameters under which the purging time t maxa plurality of groups of operation parameters meeting the purging requirement. The above operation parameters are taken as inputs, and the neural network prediction model is used to calculate the fuel cell impedance value HFR after the storage process fin whether the requirement is met. Among them, the purging method with the lowest energy consumption under the premise of ensuring the purging effect is recorded as the optimized selection of operation parameters for finally realizing the purging process, and the optimized operation parameters are executed to realize the purging, thereby improving the efficiency of the purging and reducing the energy consumption of the purging. The operation parameters include: purging time t purge , cathode purging flow Q a , anode purging flow Q c , and cell temperature T purge during purging.

[0071] Embodiment 2

[0072] This embodiment introduces a computer readable storage medium, which stores a computer program. When the computer program is executed by a processor, a fuel cell purging process optimization method as described in any one of embodiments 1 is realized.

[0073] Embodiment 3

[0074] A computer device comprises:

[0075] a memory for storing instructions.

[0076] a processor for executing the instructions, so that the computer device performs the operations of a fuel cell purging process optimization method as described in any one of embodiments 1.

[0077] Embodiment 4

[0078] Taking a single cell as an example, it is assumed that the ambient temperature is-5℃, and the fuel cell single cell impedance value needs to be greater than 16.82mΩ. Given that the flow range of the anode is 1-2SLPM, the flow range of the cathode is 3-5SLPM, the temperature range of the cell is 50℃-80℃, and the shutdown storage time is 10h. The optimized method of the present application is used to simulate multiple purging, and the optimized operation parameters are obtained, as shown in Table 1. Compared with common purging operation parameters, 2437.06J of energy is saved, and the purging time is reduced by 32s.

[0079] Table 1 Comparison of purging simulation parameters

[0080] Those skilled in the art will appreciate that embodiments of the present application can be readily used as software, hardware, or a combination of software and hardware. In a software embodiment, the present application can be implemented with computer-executable instructions, such as programs stored in memory of a computer and executed by a processor of the computer. Of course, the present application can be implemented with a combination of programs and computer-executable instructions.

[0081] The present application is described in relation to flow diagrams and / or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the present application. It is understood that each flow and / or block in the flow diagrams and / or block diagrams, and combinations of flows and / or blocks in the flow diagrams and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flow diagram flow or flows and / or block diagram block or blocks.

[0082] These computer program instructions can also be stored in a computer- readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instructions which implement the function specified in the flow diagram flow or flows and / or block diagram block or blocks.

[0083] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flow diagram flow or flows and / or block diagram block or blocks.

[0084] The above description is only preferred embodiments of the present application, and it should be pointed out that for those skilled in the art, some improvements and refinements can be made without departing from the principles of the present application, and these improvements and refinements should be considered as falling within the scope of the present application.

Claims

1. A method of optimizing a fuel cell purge process, characterized by: The method comprises the following steps: Step 1: Simulate a one-dimensional water removal model by using a simulation tool to simulate the values of the membrane-bound water, the catalyst layer-bound water and the purging time corresponding to a fuel cell purging process; Step 2: Calculate the average value of the membrane-bound water and the catalyst layer-bound water as the water content in the membrane; Step 3: Calculate the high-frequency impedance of the membrane electrode as an initial impedance value according to the water content in the membrane; Step 4: Obtain the cathode purging flow, the anode purging flow, the cell temperature during purging, the storage process environment temperature and the storage time, and take the cathode purging flow, the anode purging flow, the cell temperature during purging, the storage process environment temperature, the storage time and the initial impedance value as the inputs of the neural network, and take the high-frequency impedance value output by the neural network as the output of the neural network; Step 5: Obtain the cell temperature during purging and the high-frequency impedance value to calculate the high-frequency impedance value correction value; Step 6: Calculate the consumption of the purging gas and the energy consumption for controlling the cell temperature according to the purging time and the cell temperature during purging, and calculate the consumed energy of the fuel cell purging simulation process according to the consumption of the purging gas and the energy consumption for controlling the cell temperature; Step 7: Repeat steps 1-6 to simulate multiple fuel cell purging processes, obtain the consumed energy, the purging time and the high-frequency impedance value correction value corresponding to each fuel cell purging process, exclude the fuel cell purging processes corresponding to the purging time greater than the maximum purging time or the high-frequency impedance value correction value less than the target value of the high-frequency impedance, and select the fuel cell purging process corresponding to the minimum consumed energy among the remaining fuel cell purging processes as the optimal fuel cell purging process, and take the purging time, the cathode purging flow, the anode purging flow and the cell temperature during purging of the optimal fuel cell purging process as the operation parameters of the actual fuel cell purging process.

2. A method of purging process optimization for a fuel cell as recited in claim 1, characterized by: The one-dimensional water removal model comprises: a model of the change of the membrane-bound water with the purging process, a model of the change of the catalyst layer-bound water with the purging process, a model of the change of the gaseous water in the catalyst layer with the purging process, a model of the change of the gaseous water in the gas diffusion layer with the purging process, and a model of the change of the gaseous water in the flow channel with the purging process.

3. The fuel cell purging process optimization method according to claim 1, wherein: The model formula for calculating the change of the water in the membrane with the purging process is as follows: where J λ,m-cl represents the bound water diffusion flux between the membrane and the catalytic layer during the purging process, λ m represents the bound water content in the membrane, t represents time, is used to represent the change rate of the membrane-bound water with the purging time during the purging process; The model calculation formula of the change of the water combined in the catalytic layer with the blowing process is as follows: wherein J cl,mw-vp represents the flux of the conversion of the bound water in the catalytic layer into gaseous water during the purging process, λ cl represents the content of the bound water in the catalytic layer, is used to represent the change rate of the catalyst layer-bound water with the purging time; The model calculation formula of gaseous water in the catalytic layer with the change of the blowing process is as follows: where c vp,cl is the water vapor concentration in the catalyst layer, w ion denotes the concentration of the ionomer in the catalyst layer, ε cl denotes the porosity of the catalyst layer, p mem is the density of the membrane, EW is the equivalent weight of the membrane, J cl,lq-vp is the flux of liquid water converted to gaseous water in the catalyst layer during purging, J vp,cl-gdl is the flux of water vapor between the catalyst layer and the gas diffusion layer during purging; is used to represent the change rate of the gaseous water content in the catalyst layer with the purging time during the purging process; The model calculation formula of gaseous water in the gas diffusion layer with the change of the blowing process is as follows: Where, c vp,gdl is the concentration of water vapor in the gas diffusion layer, J gdl,lq-vp is the flux of liquid water to gaseous water in the gas diffusion layer, J vp,gdl-chan is the diffusion flux of water vapor between the gas diffusion layer and the flow channel, s lq,gdl is the volume fraction of liquid water in the gas diffusion layer; is used to represent the change rate of the gaseous water content in the gas diffusion layer with the purging time during the purging process; The model calculation formula of the change of gaseous water in the flow channel with the blowing process is as follows: wherein A act is the effective active area of the fuel cell, J chan is the water vapor flux swept away by the purge in the fuel cell flow channel, is used to represent the change rate of the gaseous water content in the flow channel with the purging time during the purging process.

4. The fuel cell purging process optimization method according to claim 1, wherein: The high frequency impedance of the membrane electrode is calculated by the following formula: In the formulae: where R represents the high frequency impedance of the membrane electrode, δ mem represents the thickness of the membrane, w represents the width of the membrane, κ eff (λ) represents the effective conductivity related to the water content of the membrane, λ represents the water content in the membrane.

5. The fuel cell purging process optimization method according to claim 1, wherein: The high-frequency impedance value correction value is calculated by the following formula: wherein HFR fin represents a high frequency impedance value correction value, HFR storage represents a high frequency impedance value, T purge represents a battery temperature at the time of purging.

6. A method of purging process optimization for a fuel cell as recited in claim 1, further comprising: The neural network adopts a BP neural network.

7. The fuel cell purging process optimization method according to claim 1, wherein: The consumed energy calculation formula is as follows: E x = E cell + E gas where E x represents the energy consumed, E cell represents the energy consumed for controlling the battery temperature, E gas represents the consumption of purge gas.

8. A computer-readable storage medium, characterized in that: A computer program product comprising a computer program stored thereon, which, when executed by a processor, implements a method of optimizing a fuel cell purge process according to any one of claims 1-7.

9. A computer device, characterized by: The computer program product comprises: a memory for storing instructions; a processor for executing the instructions to cause the computer device to perform operations of a method of optimizing a fuel cell purge process according to any one of claims 1-7.

Images