A fuel cell power system efficiency optimization method based on dynamic back pressure

Abstract

The present invention provides a fuel cell power system efficiency optimization method based on dynamic back pressure, which belongs to the field of fuel cell system control technology. First, the decision vector and corresponding constraint conditions that determine the efficiency of the fuel cell power system are determined, and the sequence of stack current density points to be optimized is determined. At each stack current density point, the control system reaches the initial operating condition, the fuel cell stack power and the air compressor power are measured, and the efficiency of the fuel cell power system is calculated. The simulated annealing algorithm is used as the basis for iteratively obtaining the system iterative solution vector, and the above control, measurement and iterative optimization process are repeated to obtain a series of operating points to be selected. Finally, the decision vector and the corresponding optimal efficiency value that optimize the system efficiency can be obtained through the optimization principle. The present invention does not rely on the system model, the algorithm structure is simple, and it is easy to implement in engineering applications. The optimization dimension is expanded, and the efficiency can be further improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of fuel cell system control, and in particular relates to a method for optimizing the efficiency of a fuel cell power system based on dynamic back pressure. Background Art

[0002] Hydrogen energy has the advantages of being clean and efficient, and is the best carrier for achieving deep decarbonization in areas such as transportation. Hydrogen fuel cells have the advantages of high fuel energy conversion rate, low noise and zero emissions, and have been widely used in transportation and fixed power stations. In the commercialization process of fuel cell technology, how to increase the net output power of the system and thus improve the utilization rate of hydrogen energy has always been a key issue.

[0003] The fuel cell power system is mainly composed of a fuel cell stack module and three auxiliary subsystems: an air system, a hydrogen system, and a cooling system. The fuel cell stack is the only energy source in the system, and its output performance is affected by conditions such as the excess air coefficient and back pressure. The auxiliary subsystem provides suitable conditions for the stable and efficient operation of the stack, and suitable operating conditions can improve the power generation efficiency of the stack (greater electrical power can be generated under the same hydrogen consumption conditions). However, the construction of suitable operating conditions for the auxiliary subsystem requires the cost of power loss, and therefore greatly affects the net output power of the entire power system (the output power of the stack minus the power consumed by the auxiliary subsystem of the stack). Since the energy consumption of the air system accounts for the majority of the three auxiliary systems, in the actual operation of the power system, it is necessary to compromise between the power generation efficiency of the stack and the power consumption of the auxiliary subsystem (mainly the air compressor) to maximize the efficiency of the entire power system.

[0004] At present, the efficiency optimization of fuel cell power systems is mainly focused on the optimal control of air flow (excess oxygen ratio or excess coefficient), and the operating pressure of the stack (mainly cathode pressure, the anode usually follows the cathode pressure to maintain the pressure balance between the cathode and the cathode) is not taken into consideration separately. In addition, since the PEMFC (proton exchange membrane fuel cell) system is a strongly nonlinear, strongly coupled and time-varying system, the ability to suppress disturbances by increasing disturbances to search for the optimal operating point is limited, and it is easy to fall into the local optimum of system efficiency. The method of designing a controller to track the reference optimal operating point obtained from the experiment cannot solve the uncertainty problems caused by time-varying factors such as stack performance degradation and air compressor aging during the system operation process. The conventional method of solving the optimal operating point based on the system online identification model and intelligent optimization algorithm is too dependent on the precise system mathematical model, and the algorithm is complex and difficult to apply in engineering practice. Summary of the invention

[0005] In response to the problems existing in the above-mentioned prior art, the present invention proposes a fuel cell power system efficiency optimization method based on dynamic back pressure, which can obtain the optimal operating conditions of different operating current working point states without relying on the system object model, thereby obtaining the maximum net power output and realizing the maximization of the output efficiency of the fuel cell power system.

[0006] The specific technical solutions of the present invention are as follows:

[0007] A method for optimizing the efficiency of a fuel cell power system based on dynamic back pressure, characterized in that it comprises the following steps:

[0008] Step 1: Determine the decision vector that determines the efficiency of the fuel cell power system Among them, p rm is the cathode air pressure ratio, i.e. the ratio of cathode outlet pressure to atmospheric pressure; is the cathode air excess coefficient; I fc is the stack current density (A / cm 2 );

[0009] Step 2: According to the performance parameter manual of the fuel cell stack and the air compressor, determine the decision variable p in the decision vector X rm , I fc Constraints:

[0010]

[0011] Among them, p max is the maximum air pressure ratio of the air compressor; The minimum excess air factor allowed for stack operation; The maximum air excess coefficient allowed for the operation of the fuel cell stack; I fc-min The minimum current density for the stack operation (A / cm 2 );I fc-max The maximum current density of the stack (A / cm 2 );

[0012] Step 3: In [I fc-min ,I fc-max ] to determine the stack current density point sequence to be optimized [I fc1 ,I fc2 ,...,I fcn ]; where n is the total number of stack current density points to be optimized; I fci , i=1,2,...,n is the i-th stack current density point;

[0013] Step 4: Set m=1;

[0014] Step 5: At the stack current density point Ifcm Under the condition of p, the simulated annealing algorithm is used to solve rm and The optimal efficiency F of the fuel cell power system is opt And the corresponding decision vector X opt ;

[0015] Step 6: Determine whether m is equal to n. If so, obtain the optimal operating condition of the fuel cell power system at all stack current density points to be optimized; otherwise, set m=m+1 and go to step 5.

[0016] Furthermore, the specific process of step 5 includes:

[0017] Step 5.1: Determine the initial temperature T0, cooling coefficient d, and termination temperature T according to the expected total number of iterations. k and the number of iterations L for each temperature; at the current temperature T = T0, the stack current density point I fcm Next, p rm and Within the constraints of 0 , let the current solution vector X cur =X 0 , the current number of iterations l = 1;

[0018] Step 5.2: At the current temperature T, by adjusting the air compressor speed and the back pressure valve opening, control the fuel cell power system to reach the current solution vector X cur Corresponding working conditions, measure the fuel cell stack power P at this time fc (X cur ) and air compressor power P air (X cur ), and then calculate the current solution vector X cur The fuel cell power system efficiency F(X cur );

[0019] Step 5.3: Based on the current solution vector X cur , superimpose the disturbance under the search speed limit, and then generate the iterative solution vector X':

[0020] X'=X cur +(-1+2×rand(1,2))×v limit

[0021] in, v pre is the maximum step value of the air pressure ratio, is the maximum step value of the air excess coefficient; rand(1,2) means generating a 1×2 dimensional random vector, and the scalar in the vector is a random number (0,1);

[0022] Step 5.4: By adjusting the air compressor speed and the back pressure valve opening, the fuel cell power system is controlled to reach the working condition corresponding to the iterative solution vector X', and the fuel cell stack power P is measured at this time. fc (X') and air compressor power P air (X'), and then calculate the fuel cell power system efficiency F(X') under the working condition corresponding to the iterative solution vector X';

[0023] Step 5.5: Calculate the increment ΔT = F(X')-F(X cur ), if ΔT>0, then accept the iterative solution vector X' as the current solution vector X cur , that is, X cur =X'; otherwise, based on the Metropolis principle, the iterative solution vector X' is accepted as the current solution vector X with probability exp(-ΔT / T) cur ;

[0024] Step 5.6: Save the current solution vector X cur And the corresponding F(X cur ), P air (X cur ), determine whether the current number of iterations l is equal to L, if so, obtain the current solution vector X at the current temperature T cur , go to step 5.7; otherwise, let l = l + 1, and go to step 5.3;

[0025] Step 5.7: Let T = dT, based on the current solution vector X cur , go to step 5.3; until the current temperature T = T k , obtain the stack current density point I fcm All current solution vectors X saved under cur And the corresponding F(X cur ), P air (X cur ), select the optimal efficiency F opt And the corresponding decision vector X opt .

[0026] Furthermore, in step 5.1, T>T k , 0 <d<1。

[0027] Furthermore, the formula for calculating the efficiency F(X) of the fuel cell power system under the working condition corresponding to the solution vector X is:

[0028]

[0029] in, is the hydrogen mass flow rate (g / s); is the lower calorific value of hydrogen, which is 1.2×105 kJ / kg; A fc is the stack reaction area (cm 2 );N fc is the number of battery stacks; is the molar mass of hydrogen, which is 2.02 g / mol; F is the Faraday constant, which is 96485 C / mol;

[0030] In step 5.2, the current solution vector X cur As the solution vector X, solve F(X) based on the formula of F(X) cur );

[0031] In step 5.4, the iterative solution vector X' is used as the solution vector X, and F(X') is solved based on the formula of F(X).

[0032] Furthermore, in step 5.7, the optimal efficiency F is selected opt And the corresponding decision vector X opt The process is: in all the saved F(X cur ), select the maximum efficiency value F best , and filter all cdd ≥αF best The candidate efficiency value F cdd , and then among all candidate efficiency values ​​F cdd In the calculation, the efficiency value corresponding to the minimum air compressor power is selected as the optimal efficiency F opt , and the corresponding decision vector X opt ; Wherein, α is the preset efficiency coefficient.

[0033] The beneficial effects of the present invention are:

[0034] The present invention proposes a fuel cell power system efficiency optimization method based on dynamic back pressure. The method is independent of the system model, has a simple algorithm structure, and is easy to implement in engineering applications. In the process of searching for the optimal efficiency, the method effectively avoids falling into the local minimum and eventually tends to the global optimum. Compared with the strategy of only optimizing the peroxygen ratio under normal pressure in the prior art, the fuel cell system efficiency optimization strategy based on dynamic back pressure in the present invention expands the optimization dimension and can further improve the efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 This is a flow chart of a fuel cell power system efficiency optimization method based on dynamic back pressure proposed in Example 1 of the present invention;

[0036] Figure 2 This is a system block diagram used in Example 1 of the present invention;

[0037] Figure 3 0.1-1.2A / cm in Example 1 of the present invention2 The optimal operating point under all the current density points of the battery stack to be optimized;

[0038] Figure 4 0.9A / cm in Example 1 of the present invention 2 The optimal efficiency solution at the stack current density point. DETAILED DESCRIPTION

[0039] In order to make the objectives, technical solutions and advantages of the present invention more clear, the present invention is further described in conjunction with the following specific embodiments and with reference to the accompanying drawings.

[0040] The following non-limiting embodiments may enable a person skilled in the art to more fully understand the present invention, but are not intended to limit the present invention in any way.

[0041] Example 1

[0042] This embodiment proposes a fuel cell power system efficiency optimization method based on dynamic back pressure. Figure 2 The system block diagram shown is implemented. The method first selects the decision variable X that affects the output efficiency of the power generation system; then determines the constraint conditions of the decision variable and the sequence of stack current density to be optimized; then selects each stack current density as the optimization background in sequence; under different stack current densities, the air compressor speed and the back pressure valve opening are adjusted by the controller to make the system reach the working condition X, and the hydrogen mass flow, fuel cell stack power and air compressor power of the fuel cell system power system under this working condition are measured to obtain the fuel cell power system efficiency F(X), and the efficiency value is used as the basis for the simulated annealing algorithm to iteratively obtain the next working condition solution vector X' of the system, repeat the above control, measurement and iterative optimization process, and a series of working condition points to be selected can be obtained. Finally, the decision variables and corresponding optimal values ​​that optimize the system efficiency can be obtained through the optimization principle.

[0043] The process of the fuel cell power system efficiency optimization method based on dynamic back pressure proposed in this embodiment is as follows: Figure 1 As shown, the following steps are included:

[0044] Step 1: Determine the decision vector that determines the efficiency of the fuel cell power system based on the factors that affect the output performance of the fuel cell stack and the power consumption of auxiliary components. Among them, p rm is the cathode air pressure ratio, i.e. the ratio of cathode outlet pressure to atmospheric pressure; is the cathode air excess coefficient; I fc is the stack current density (A / cm 2 );

[0045] Step 2: According to the performance parameter manual of the fuel cell stack and the air compressor, determine the decision variable p in the decision vector X rm , I fc Constraints:

[0046]

[0047] Among them, p max =1.8 is the maximum air pressure ratio of the air compressor; The minimum excess air factor allowed for stack operation; The maximum air excess coefficient allowed for the operation of the fuel cell stack; I fc-min =0.1A / cm 2 The minimum current density for the stack operation; I fc-max =1.2A / cm 2 The maximum current density for the stack operation;

[0048] Step 3: In [I fc-min ,I fc-max ] to determine the stack current density point sequence to be optimized [I fc1 ,I fc2 ,...,I fcn ]=[0.1,0.2,...,1.2]; where n=12 is the total number of stack current density points to be optimized; I fci , i=1,2,...,12 is the i-th stack current density point;

[0049] Step 4: Set m=1;

[0050] Step 5: At the stack current density point I fcm Under the condition of p, the simulated annealing algorithm is used to solve rm and λ O2 The optimal efficiency F of the fuel cell power system is opt And the corresponding decision vector X opt The specific process includes:

[0051] Step 5.1: According to the expected total number of iterations, determine the initial temperature T0 = 10, the cooling coefficient d = 0.5, and the termination temperature T k = 1 and the number of iterations for each temperature is L = 4; at the current temperature T = T0 and the stack current density point I fcm Next, p rm and Within the constraints of 0 , let the current solution vector X cur =X 0 , the current number of iterations l = 1;

[0052] Step 5.2: At the current temperature T, by adjusting the air compressor speed and the back pressure valve opening (i.e., dynamic back pressure regulation), control the fuel cell power system to reach the current solution vector X cur Corresponding working conditions, measure the fuel cell stack power P at this time fc (X cur ) and air compressor power P air (X cur ), and then calculate the current solution vector X cur The fuel cell power system efficiency F(X cur ):

[0053]

[0054] in, is the hydrogen mass flow rate (g / s); is the lower calorific value of hydrogen, which is 1.2×10 5 kJ / kg; A fc =350cm 2 is the stack reaction area; N fc =340 is the number of battery stack nodes; is the molar mass of hydrogen, which is 2.02 g / mol; F is the Faraday constant, which is 96485 C / mol;

[0055] Step 5.3: Based on the current solution vector X cur , under the search speed limit (to avoid too large a step in the pressure ratio / air excess coefficient, which would affect the stable operation of the system), the disturbance is superimposed to generate the iterative solution vector X':

[0056] X'=X cur +(-1+2×rand(1,2))×v limit

[0057] in, v pre is the maximum step value of the air pressure ratio, is the maximum step value of the air excess coefficient; rand(1,2) means generating a 1×2 dimensional random vector, and the scalar in the vector is a random number (0,1);

[0058] Step 5.4: By adjusting the air compressor speed and the back pressure valve opening, the fuel cell power system is controlled to reach the working condition corresponding to the iterative solution vector X', and the fuel cell stack power P is measured at this time. fc (X') and air compressor power P air (X'), and then calculate the fuel cell power system efficiency F(X') under the working condition corresponding to the iterative solution vector X':

[0059]

[0060] Step 5.5: Calculate the increment ΔT = F(X')-F(X cur ), if ΔT>0, then accept the iterative solution vector X' as the current solution vector X cur , that is, X cur =X'; otherwise, based on the Metropolis principle, the iterative solution vector X' is accepted as the current solution vector X with probability exp(-ΔT / T) cur ;

[0061] Step 5.6: Save the current solution vector X cur And the corresponding F(X cur ), P air (X cur ), determine whether the current number of iterations l is equal to L, if so, obtain the current solution vector X at the current temperature T cur , go to step 5.7; otherwise, let l = l + 1, and go to step 5.3;

[0062] Step 5.7: Let T = dT, based on the current solution vector X cur , go to step 5.3; until the current temperature T = T k , obtain the stack current density point I fcm All current solution vectors X saved under cur And the corresponding F(X cur ), P air (X cur ), select the optimal efficiency F opt And the corresponding decision vector X opt ;

[0063] The selection process is: among all the saved F(X cur ), select the maximum efficiency value F best = 66.72%, and screen all those that meet F cdd ≥αF best The candidate efficiency value F cdd (66.34%, 66.38%, 66.47%, 66.65%, 66.72%), and then among all the candidate efficiency values ​​F cdd In the calculation, the efficiency value corresponding to the minimum air compressor power (slowing down the aging of the air compressor) is selected as the optimal efficiency F opt = 66.47%, and the corresponding decision vector X opt =(1,2.4181,0.1), such as Figure 4 As shown; wherein, α = 99%;

[0064] Step 6: Determine whether m is equal to n. If so, obtain the optimal operating condition of the fuel cell power system at all stack current density points to be optimized; otherwise, set m=m+1 and go to step 5.

[0065] After the above-mentioned efficiency optimization of each current density point is completed, each optimal working point is calibrated to the MAP diagram of the air compressor (the change diagram of the relationship between the pressure flow and the speed of the air compressor) to obtain the operating condition spectrum with the best system efficiency, such as Figure 3 As indicated by the Chinese point.

[0066] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the embodiments, a person skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the purpose and scope of the technical solution, which should be included in the scope of the claims of the present invention.

Claims

1. A method for optimizing the efficiency of a fuel cell power system based on dynamic back pressure, characterized in that: The following steps are involved: Step 1: Determine the decision vector that determines the efficiency of the fuel cell power system Among them, p rm is the cathode air pressure ratio; is the cathode air excess coefficient; I fc is the stack current density; Step 2: Determine the decision variable p in the decision vector X rm , I fc Constraints: Among them, p max is the maximum air pressure ratio of the air compressor; The minimum excess air factor allowed for stack operation; The maximum air excess coefficient allowed for the operation of the fuel cell stack; I fc-min The minimum current density for the stack operation; I fc-max The maximum current density for the stack operation; Step 3: In [I fc-min ,I fc-max ] to determine the stack current density point sequence to be optimized [I fc1 ,I fc2 ,...,I fcn ]; where n is the total number of stack current density points to be optimized; I fci , i=1,2,...,n is the i-th stack current density point; Step 4: Set m=1; Step 5: At the stack current density point I fcm Under the condition of p, the simulated annealing algorithm is used to solve rm and The optimal efficiency F of the fuel cell power system is opt And the corresponding decision vector X opt ; Step 6: Determine whether m is equal to n. If so, obtain the optimal operating condition of the fuel cell power system at all stack current density points to be optimized; otherwise, set m=m+1 and go to step 5.

2. The method for optimizing the efficiency of a fuel cell power system based on dynamic back pressure according to claim 1, characterized in that: The specific process of step 5 includes: Step 5.1: Determine the initial temperature T0, cooling coefficient d, and termination temperature T according to the expected total number of iterations. k and the number of iterations L for each temperature; at the current temperature T = T0, the stack current density point I fcm Next, p rm and Within the constraints of 0 , let the current solution vector X cur =X 0 , the current number of iterations l = 1; Step 5.2: At the current temperature T, by adjusting the air compressor speed and the back pressure valve opening, control the fuel cell power system to reach the current solution vector X cur Corresponding working conditions, measure the fuel cell stack power P at this time fc (X cur ) and air compressor power P air (X cur ), and then calculate the current solution vector X cur The fuel cell power system efficiency F(X cur ); Step 5.3: Based on the current solution vector X cur , superimpose the disturbance under the search speed limit, and then generate the iterative solution vector X': X'=X cur +(-1+2×rand(1,2))×v limit in, v pre is the maximum step value of the air pressure ratio, is the maximum step value of the air excess coefficient; rand(1,2) means generating a 1×2 dimensional random vector, and the scalar in the vector is a random number (0,1); Step 5.4: By adjusting the air compressor speed and the back pressure valve opening, the fuel cell power system is controlled to reach the working condition corresponding to the iterative solution vector X', and the fuel cell stack power P is measured at this time. fc (X') and air compressor power P air (X'), and then calculate the fuel cell power system efficiency F(X') under the working condition corresponding to the iterative solution vector X'; Step 5.5: Calculate the increment ΔT = F(X')-F(X cur ), if ΔT>0, then accept the iterative solution vector X' as the current solution vector X cur , that is, X cur =X'; otherwise, based on the Metropolis principle, the iterative solution vector X' is accepted as the current solution vector X with probability exp(-ΔT / T) cur ; Step 5.6: Save the current solution vector X cur And the corresponding F(X cur ), P air (X cur ), determine whether the current number of iterations l is equal to L, if so, obtain the current solution vector X at the current temperature T cur , go to step 5.7; otherwise, let l = l + 1, and go to step 5.3; Step 5.7: Let T = dT, based on the current solution vector X cur , go to step 5.3; until the current temperature T = T k , obtain the stack current density point I fcm All current solution vectors X saved under cur And the corresponding F(X cur ), P air (X cur ), select the optimal efficiency F opt And the corresponding decision vector X opt .

3. The method for optimizing the efficiency of a fuel cell power system based on dynamic back pressure according to claim 2, characterized in that: In step 5.1, T>T k , 0 <d<1。 4. The method for optimizing the efficiency of a fuel cell power system based on dynamic back pressure according to claim 2, characterized in that: The formula for calculating the fuel cell power system efficiency F(X) under the working condition corresponding to the solution vector X is: in, is the mass flow rate of hydrogen; is the lower calorific value of hydrogen; A fc is the stack reaction area; N fc is the number of battery stacks; is the molar mass of hydrogen; F is the Faraday constant; In step 5.2, the current solution vector X cur As the solution vector X, solve F(X) based on the formula of F(X) cur ); In step 5.4, the iterative solution vector X' is used as the solution vector X, and F(X') is solved based on the formula of F(X).

5. The method for optimizing the efficiency of a fuel cell power system based on dynamic back pressure according to claim 2, characterized in that: Select the optimal efficiency F in step 5.7 opt And the corresponding decision vector X opt The process is: in all the saved F(X cur ), select the maximum efficiency value F best , and filter all cdd ≥αF best The candidate efficiency value F cdd , and then among all candidate efficiency values ​​F cdd The efficiency value corresponding to the minimum air compressor power is the optimal efficiency F opt , and the corresponding decision vector X opt ; Wherein, α is the preset efficiency coefficient.

Images