A method for optimizing control of anode and cathode gases of a variable-altitude fuel cell

Abstract

The present application relates to the technical field of system optimization control, and discloses a cathode and anode gas collaborative optimization control method of a variable-altitude fuel cell, comprising: constructing a fuel cell stack model, an air supply system model, a hydrogen supply system model and an atmospheric environment parameter model; analyzing the change of the safe working area of an air compressor under a variable-altitude environment, and proposing a self-adaptive model predictive control method of the fuel cell over-oxygen ratio and the cathode pressure; setting a double-surge-discrimination mechanism of the air compressor, combining the working state of the air compressor in the prediction time domain to correct the control variable; taking the cathode pressure as a reference, designing a coordinated control method of the anode pressure and the hydrogen ratio based on the super-helix algorithm; and selecting the best exhaust strategy under the variable-altitude environment by using the entropy weight TOPSIS evaluation system. The present application proposes a variable-altitude cathode surge suppression and anode dynamic exhaust collaborative control method aiming at the problem of air compressor surge and significant anode pressure fluctuation under a variable-altitude environment, and improves the operation reliability of the fuel cell under a variable-altitude.

Description

Technical Field

[0001] This invention relates to the field of system optimization control technology, and in particular to a method for coordinated optimization control of anode and cathode gases in a variable altitude fuel cell. Background Technology

[0002] Hydrogen fuel cell technology, as an important approach to the efficient utilization of hydrogen energy, has attracted widespread attention from countries around the world and the transportation sector due to its advantages such as high efficiency, zero pollution, and low noise. Commercial vehicles, characterized by high power, long distance, and fixed routes, are highly compatible with the long driving range and high power output characteristics of fuel cell systems.

[0003] Despite the progress made in modeling and control of fuel cell gas supply systems in plain environments, reliability issues have gradually emerged under the constraints of high-altitude environments with low oxygen partial pressure. Specifically, increasing altitude narrows the safe operating range of the air compressor, and the accompanying increase in compression ratio easily induces surge risk. Model-based control methods show decreased accuracy in predicting anode and cathode flow / pressure and their surge detection capability when dealing with scenarios where the air compressor's operating characteristics dynamically shift. Furthermore, improper actuator coordination during dynamic adjustment of gas supply flow and pressure can easily cause fluctuations in the anode-cathode pressure difference, especially at high altitudes where the pressure difference between the internal and external environments of the fuel cell system is significant, ultimately leading to proton exchange membrane damage. Therefore, a coordinated control method combining adaptive surge suppression at the cathode and optimal exhaust at the anode, based on varying altitude, is proposed as a crucial guarantee for improving the reliability of fuel cell systems operating in varying altitude environments. Summary of the Invention

[0004] The purpose of this invention is to solve the problems in the prior art.

[0005] The technical solution adopted by this invention to solve its technical problem is: to provide a method for coordinated optimization control of anode and cathode gases in a variable altitude fuel cell, comprising the following steps:

[0006] A fuel cell stack model, an air supply system model, a hydrogen supply system model, and an atmospheric environmental parameter model are constructed. The air supply system model includes an air compressor MAP, intake / exhaust manifolds, cathode flow channels, and a back pressure valve. The hydrogen supply system model includes a solenoid valve, a proportional valve, an ejector, a hydrogen supply manifold, an anode flow channel, a return manifold, an exhaust valve, and a hydrogen circulation pump. The atmospheric environmental parameter model includes ambient pressure, ambient temperature, and air density. Both the air supply system model and the hydrogen supply system model are nonlinear models.

[0007] Based on the changes in the safe operating range of the air compressor under varying altitudes, adaptive predictive control of the peroxy ratio and cathode pressure of the fuel cell is performed based on the load current and altitude range.

[0008] A dual surge detection mechanism for the air compressor is set up, and the control variables of the air supply system are corrected by combining the air compressor's working status in the prediction time domain.

[0009] Using cathode pressure as a reference, a design is made to coordinate the control of anode hydrogen ratio and anode pressure in a fuel cell based on a superspiral algorithm;

[0010] Using the average efficiency, maximum efficiency, average single-cell voltage of the fuel cell stack, anode pressure fluctuation error, and nitrogen concentration as evaluation indicators, the optimal anode venting strategy under varying altitude environments was selected using the entropy weight TOPSIS evaluation system.

[0011] Preferably, the adaptive predictive control of the cathode oxygen ratio and cathode pressure of the fuel cell based on the load current and altitude range, according to the changes in the safe operating range of the air compressor under varying altitude conditions, includes the following steps:

[0012] Obtain the operating safety boundary of the air compressor under varying altitudes, represented as:

[0013] ;

[0014] Among them, W surge,h and W choke,h The corresponding air compressor outlet pressure p at altitude h comp,h Surge flow and blockage flow, c 00,h c 01,h c 02,h c 10,h c 11,h c 12,h All are fitted parameters;

[0015] The load current and altitude are divided into intervals. A Taylor series expansion of the air supply system model is performed in each interval. A linearized approximate model is established by truncating higher-order components, as follows:

[0016] ;

[0017] Among them, A and B u B d C, D u D d B w D w All are Jacobian matrices; X(t) represents the system state variables. Differentiate the system state variables. For system output variables, For system control variables, For load disturbance, The linearization error disturbance is represented by t, which is a time variable.

[0018] Consider using integration to reduce or eliminate static error, transforming the linearized approximation model into a linear discrete-time state-space model, and writing it in incremental form as follows:

[0019] ;

[0020] Where Δ represents the increment, k+1 and k represent time k+1 and time k respectively, and dis represents discretization processing;

[0021] Based on the incremental linear discrete-time state-space model, a prediction equation is constructed based on the output variables in the prediction time domain, thereby establishing the objective function for regulating the oxygen ratio and cathode pressure, expressed as:

[0022] ;

[0023] Among them, Γ y To predict the weighting factor, Γ u To control the weighting factor, S u Let E be the system state coefficient matrix. p To predict the step size error, This represents the square of the norm.

[0024] Preferably, the division of load current and altitude into intervals specifically involves:

[0025] The load current is divided into five ranges, including (0, 100] A, (100, 200] A, (200, 300] A, (300, 400] A, and (400, 450] A;

[0026] The altitude is divided into five ranges, including (0, 800] m, (800, 1600] m, (1600, 2400] m, (2400, 3200] m, and (3200, 4000] m.

[0027] Preferably, the setting of the dual surge discrimination mechanism for the air compressor, combined with the correction of control variables based on the air compressor's operating state in the prediction time domain, includes the following steps:

[0028] The first surge warning detector is set up as follows:

[0029] ;

[0030] Among them, J 1,surge,k,i W is the judgment signal of the first surge detector in the i-th prediction step of the k-th control step. comp,k,i and p comp,k,i W represents the predicted flow rate and pressure of the air compressor output in the i-th prediction step within the k-th control step.surge,k,i The surge judgment signal for the air compressor output pressure corresponding to the i-th prediction step in the k-th control step;

[0031] Setting a second surge warning detector is represented as follows:

[0032] ;

[0033] ;

[0034] Among them, J 2,surge,k For the judgment signal of the second surge detector at the k-th control step, NC Wcomp,low The low threshold for the normalized rate of change of air compressor flow rate; NC Wcomp,k The normalized rate of change of air compressor flow rate is represented by C. Wcomp,k The normalized result, T s For sampling time, W comp,k, Predict the output flow rate of the air compressor in the k-th control step;

[0035] In the subsequent k-th control step, when J 1,surge,k,i =1 and J 2,surge,k When =1, the controller adjusts two control variables—the current air compressor voltage and the back pressure valve opening—based on the predicted sequence:

[0036] ;

[0037] Among them, u comp,af,k and u comp,bef,k and u comp,cor,k These are the corrected operating voltage, the original operating voltage, and the amount of voltage correction for the air compressor. bpv,af,k and u bpv,bef,k and u bpv,cor,k These are the corrected back pressure valve opening, the original back pressure valve opening, and the correction amount for the back pressure valve opening, respectively.

[0038] Preferably, the step of adjusting the current air compressor voltage and back pressure valve opening based on the predicted sequence is specifically as follows: using the 110% surge line as the surge warning line, that is, the value of the control variable = 110% × |surge value - predicted control output value|.

[0039] Preferably, the step of designing a coordinated control of the hydrogen permeation ratio and anode pressure of the fuel cell based on a superspiral algorithm, with the cathode pressure as a reference, includes the following steps:

[0040] The sliding variables, including sliding mode variables s1(x) and s2(x), are directly calculated using the anode-cathode pressure difference and the hydrogen permeation ratio deviation. Sliding mode variable s1(x) represents the difference between the target cathode pressure and the anode pressure, and sliding mode variable s2(x) represents the difference between the target hydrogen permeation ratio and the actual feedback hydrogen permeation ratio of the system.

[0041] The control variable u of the anode system is designed based on the sliding variable, and is expressed as:

[0042] ;

[0043] Among them, u=[u fcv , u pump ] T , s=[s1, s2] T , γ=[γ1, γ2] T α = [α1, α2] T u fcv For the proportional valve opening, u pump denoted as the rotational speed of the hydrogen circulation pump, γ and α are both convergence coefficients, γ1 and α1 are the convergence coefficients of the sliding mode variable s1(x), γ2 and α2 are the convergence coefficients of the sliding mode variable s2(x); sign represents the sign function.

[0044] Preferably, the step of using the average efficiency, maximum efficiency, average single-cell voltage of the fuel cell stack, anode pressure fluctuation error, and nitrogen concentration as evaluation indicators, and selecting the optimal anode venting strategy under varying altitude environments using the entropy-weighted TOPSIS evaluation system, includes the following steps:

[0045] Evaluation metrics for designing an anode system include average efficiency, maximum efficiency, average single-cell voltage of the fuel cell stack, anode pressure fluctuation error, and nitrogen concentration.

[0046] Data is standardized according to the direction of indicator attributes, and information entropy and weight values ​​are calculated. The weight values ​​are multiplied by the data to obtain new data, which is then calculated using the TOPSIS method.

[0047] A segmented analysis of the current density was performed, and several exhaust strategies were set for each current range. After comprehensive evaluation, the optimal exhaust strategy for different current densities and altitude environments was obtained.

[0048] Preferably, the formula for calculating the efficiency of the anode system is as follows:

[0049] ;

[0050] Where, η an For the efficiency of the anode system, P st For the fuel cell stack power, P pump For the power consumption of the hydrogen circulation pump, LHV H2Because of the low calorific value of hydrogen, This refers to the outlet hydrogen flow rate of the anode exhaust valve. The flow rate of hydrogen gas consumed in the reaction is denoted by 'cycle', which represents one cycle.

[0051] Preferably, the anode pressure fluctuation error is expressed as:

[0052] ;

[0053] in, For the anode pressure fluctuation error, p an p is the anode pressure. ref This is a pressure reference value.

[0054] Preferably, the step of performing segmented analysis of current density, setting several exhaust strategies for each current range, and obtaining the optimal exhaust strategy for different current densities and altitude environments after comprehensive evaluation includes the following steps:

[0055] The current density in the piecewise analysis is divided into four intervals, including (0, 0.7] A / cm. 2 (0.7, 1.0] A / cm 2 (1.0, 1.3] A / cm 2 (1.3, 1.6] A / cm 2 ;

[0056] Fifteen exhaust strategies are set for each current range, where (0, 0.7] A / cm 2 The exhaust intervals at different current densities were 10 s, 20 s, and 30 s, and the exhaust durations were 0.4 s, 0.6 s, 0.8 s, 1.0 s, and 1.2 s; (0.7, 1.0] A / cm 2 The exhaust intervals at different current densities included 7 s, 14 s, and 18 s, and the exhaust durations included 0.4 s, 0.6 s, 0.8 s, 1.0 s, and 1.2 s; >1.0 A / cm 2 The exhaust intervals at different current densities include 4 s, 8 s, and 12 s, and the exhaust durations include 0.4 s, 0.6 s, 0.8 s, 1.0 s, and 1.2 s.

[0057] The present invention has the following beneficial effects:

[0058] (1) This invention mainly proposes a method for coordinated optimization control of anode and cathode gases in a variable altitude fuel cell, which takes into account reliability improvement. This method overcomes the problems of air compressor surge boundary drift and anode exhaust dynamic instability in variable altitude environments, thereby improving the operational reliability of the fuel cell system.

[0059] (2) The model adaptive update strategy based on the present invention designs an adaptive model predictive control method for cathode gas flow and pressure, integrates a dual-layer surge discrimination mechanism of flow / pressure and flow change rate, realizes active anti-surge of altitude gradient air compressor under extreme working conditions, and improves the prediction accuracy and suppression capability of surge.

[0060] (3) This invention utilizes the entropy weight TOPSIS evaluation system to optimize the anode exhaust strategy, thereby optimizing anode efficiency and gas supply stability, and better solving the problems of anode pressure instability and gas permeation.

[0061] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments, but the present invention is not limited to the embodiments. Attached Figure Description

[0062] Figure 1 This is a flowchart illustrating the method steps of a variable altitude fuel cell anode and cathode gas co-optimization control method according to an embodiment of the present invention.

[0063] Figure 2 This is a cathode gas predictive control block diagram of a method for coordinated optimization control of cathode and anode gases in a variable altitude fuel cell according to an embodiment of the present invention.

[0064] Figure 3 This invention relates to a method for coordinated optimization control of anode and cathode gases in a variable altitude fuel cell, and describes the variation trends of surge and blockage lines of an air compressor in a variable altitude environment.

[0065] Figure 4 This is the air compressor anti-surge result under altitude gradient of a variable altitude fuel cell anode and cathode gas co-optimization control method according to an embodiment of the present invention;

[0066] Figure 5 This is a research block diagram of the anode pressure tracking method for a variable altitude fuel cell based on the optimal exhaust strategy in a method for coordinated optimization control of anode and cathode gases in a variable altitude fuel cell according to an embodiment of the present invention.

[0067] Figure 6 This invention relates to a method for coordinated optimization control of anode and cathode gases in a variable altitude fuel cell, applicable at different altitudes of 1.0 A / cm. 2 The evolution of anode gas pressure, concentration, and efficiency over time during purging at current density;

[0068] Figure 7 This is the anode system performance control result under the test condition (CHTC) of a heavy-duty commercial vehicle at an altitude of 4000m, according to an embodiment of the present invention, which is a method for coordinated optimization control of anode and cathode gases in a variable altitude fuel cell. Detailed Implementation

[0069] The present invention will now be described in more detail with reference to the accompanying drawings and embodiments.

[0070] Please refer to Figure 1 This invention provides a method for coordinated optimization control of anode and cathode gases in a variable altitude fuel cell, taking into account reliability improvement. The method is characterized by the following steps:

[0071] S101, construct fuel cell stack model, air supply system model, hydrogen supply system model and atmospheric environmental parameter model;

[0072] S102, based on the change of the safe operating range of the air compressor in the variable altitude environment, adaptively controls the cathode oxygen ratio and cathode pressure of the fuel cell based on the load current and altitude range.

[0073] S103, set up a dual surge discrimination mechanism for the air compressor, and correct the control variables by combining the air compressor working status in the prediction time domain;

[0074] S104, using cathode pressure as a reference, designs a method for coordinated control of the anode hydrogen ratio and anode pressure of a fuel cell based on a superspiral algorithm;

[0075] S105 uses the average efficiency, maximum efficiency, average single-cell voltage of the fuel cell stack, anode pressure fluctuation error, and nitrogen concentration as evaluation indicators, and employs the entropy weight TOPSIS evaluation system to select the optimal anode venting strategy under varying altitude environments.

[0076] The air supply system model includes an air compressor (MAP), intake / exhaust manifolds, cathode channels, and back pressure valves; the hydrogen supply system model includes a solenoid valve, proportional valve, ejector, hydrogen supply manifold, anode channels, return manifold, exhaust valve, and hydrogen circulation pump; and the atmospheric environmental parameter model includes ambient pressure, ambient temperature, and air density.

[0077] Specifically, the fuel cell stack model is represented as follows:

[0078] (1)

[0079] Among them, E nernst V is the open-circuit voltage of the fuel cell. act To activate the polarization voltage, V ohm V is the ohmic polarization voltage. con n is the concentration polarization voltage. cell Where is the number of individual battery cells, and F is the Faraday constant. Let T be the Gibbs free energy, R be the gas constant, and T be the gas free energy. fc p is the temperature of the fuel cell stack. H2 For the partial pressure of hydrogen, p O2 For the partial pressure of oxygen, i fcLet i be the current density. fc,max R is the limiting current density. ohm For ohmic resistors, C1, C2, C3, V0, V a These are the fitting parameters.

[0080] See Figure 2 As shown, step S102 specifically includes:

[0081] Step S21: Obtaining the safe operating boundary of the air compressor under varying altitudes; this is based on the changes in the safe operating area of ​​the air compressor under varying altitudes, such as... Figure 3 As shown; mathematical representations of surge lines and blocking lines under different altitude conditions are established using a polynomial fitting method:

[0082] (2)

[0083] Among them, W surge,h and W choke,h The corresponding air compressor outlet pressure p at altitude h comp,h Surge flow and blockage flow, c ij,h These are the fitting parameters.

[0084] Step S22: Cathode gas predictive control based on model adaptive update

[0085] Based on the load current and altitude range, an adaptive update strategy for the model is set. To address the strong nonlinear characteristics of the air supply system, a Taylor series expansion is performed on the original nonlinear model at the stable operating point, and a linearized approximate model is established by truncating higher-order components.

[0086] (3)

[0087] Among them, A and B u B d C, D u D d All are Jacobian matrices.

[0088] Consider using integration to reduce or eliminate static error, transforming the above equation into a linear discrete-time state-space model, and further rewriting it in incremental form:

[0089] (4)

[0090] Based on the incremental linear discrete-time state-space model, a prediction equation is constructed based on the output variables in the prediction time domain, thereby establishing an objective function for regulating the oxygen ratio and cathode pressure. This objective function is then transformed into a quadratic programming problem for solution, and rewritten as follows:

[0091] (5)

[0092] Among them, Γ y To predict the weighting factor, Γ u To control the weighting factor, S u This is the system state coefficient matrix.

[0093] Specifically, the prediction equation based on the output variable in the prediction time domain is constructed according to the incremental linear discrete-time state-space model, using the following formula:

[0094] (6)

[0095] in, These are the reference values ​​for the model predictive controller, namely the target values ​​for the oxygen ratio and cathode pressure.

[0096] Specifically, step S103 is as follows:

[0097] Step S31: Setting up the first surge warning detector

[0098] The optimal control sequence is generated based on the prediction model, and the surge risk is pre-determined through comparative calculation. During the control process of the k-th control step, the surge behavior at the i-th time in the prediction time domain is marked, and the first surge early warning detector is designed by following the judgment equation:

[0099] (7)

[0100] Among them, J 1,surge,k,i W is the judgment signal of the first surge detector in the i-th prediction step of the k-th control step. comp,k,i and p comp,k,i W represents the predicted flow rate and pressure of the air compressor output in the i-th prediction step within the k-th control step. surge,k,i This is the surge judgment signal for the air compressor output pressure corresponding to the i-th prediction step in the k-th control step.

[0101] Step S32: Setting up the second surge warning detector

[0102] Considering the limited predictive capability and accuracy of the controller, the air compressor flow rate change characteristic is introduced in the k-th control step to further identify surge characteristics.

[0103] (8)

[0104] Among them, C Wcomp,k This represents the rate of change of air compressor flow.

[0105] To standardize the evaluation criteria for different air compressor operating conditions, the tanh function is used to normalize the air compressor flow rate change to the range of [-1, 1], and it is set to NC. Wcomp,k Here, the flow rate change rate is set as the second detector. Surge is identified by comparing it to a threshold set for the air compressor flow rate change rate. The judgment equation is:

[0106] (9)

[0107] Among them, J 2,surge,k For the judgment signal of the second surge detector at the k-th control step, NC Wcomp,low The low threshold for the normalized rate of change of air compressor flow.

[0108] In the subsequent k-th control step, the controller adjusts two control variables—the current air compressor voltage and the back pressure valve opening—based on the predicted sequence:

[0109] (10)

[0110] Among them, u comp,af,k and u comp,bef,k and u comp,cor,k These are the operating voltages of the air compressor before, after, and after correction, respectively. bpv,af,k and u bpv,bef,k and u bpv,cor,k These represent the opening degree of the back pressure valve after correction, before correction, and after correction.

[0111] See Figure 5 As shown, step S104 specifically includes:

[0112] The sliding variables are directly calculated using the anode-cathode pressure difference and the hydrogen permeation ratio deviation. Specifically, the sliding mode variable s1(x) is defined as the difference between the target cathode pressure and the anode pressure, and the sliding mode variable s2(x) is defined as the difference between the target hydrogen permeation ratio and the actual feedback hydrogen permeation ratio of the system. The control law u can be designed as follows:

[0113] (11)

[0114] Among them, u=[u fcv , u pump ] T , s=[s1, s2] T , γ=[γ1, γ2] T u2=[u2_1, u 2_2 ] T α = [α1, α2] T .

[0115] Specifically, step S105 is as follows:

[0116] Step S51: Comprehensive Impact Analysis of Anode Exhaust Strategy

[0117] Anode system efficiency, as an important evaluation indicator of exhaust strategy, comprehensively considers the net power of the anode system and hydrogen emission loss. Its calculation formula is as follows:

[0118] (12)

[0119] Where, η an For the efficiency of the anode system, P pump For the power consumption of the hydrogen circulation pump, LHV H2 It has the lowest calorific value of hydrogen.

[0120] The venting interval and duration of the venting valve both affect the stability of the anode pressure, especially in high-altitude environments, such as... Figure 6 As shown. Therefore, pressure fluctuation is also an important evaluation indicator for exhaust strategies.

[0121] (13)

[0122] Where, p ref This is a pressure reference value.

[0123] Step S52: Optimal exhaust strategy based on entropy-weighted TOPSIS evaluation system

[0124] First, five evaluation indicators were designed for the anode system: average efficiency, maximum efficiency, average single-cell voltage of the stack, anode pressure fluctuation error, and nitrogen concentration.

[0125] Then, the data is standardized according to the direction of the indicator attributes, and the information entropy value and weight value are calculated. The weight value is multiplied by the data to obtain new data, which is then calculated using the TOPSIS method.

[0126] Finally, a segmented analysis of the current density was conducted, with 15 exhaust strategies set for each current range. After comprehensive evaluation, the optimal exhaust strategy for different current densities and altitude environments was obtained.

[0127] Specifically, the load current and altitude ranges are divided into 5 regions, with the load current range divided into (0, 100] A, (100, 200] A, (200, 300] A, (300, 400] A, and (400, 450] A, and the altitude range divided into (0, 800] m, (800, 1600] m, (1600, 2400] m, (2400, 3200] m, and (3200, 4000] m.

[0128] Specifically, the controller adjusts the current air compressor voltage and back pressure valve opening based on the predicted sequence using a 110% surge line as the surge warning line. That is, the adjustment of the control variables is based on the difference between 110% × surge value and the predicted air mass flow / pressure value.

[0129] Specifically, in the segmented analysis of the current density, the current density is divided into (0, 0.7] A / cm. 2 (0.7, 1.0] A / cm 2 (1.0, 1.3] A / cm 2 (1.3, 1.6] A / cm 2 There are 4 sections in total.

[0130] Specifically, 15 exhaust strategies are set for each current range, including 3 exhaust intervals and 5 exhaust durations. These 15 exhaust strategies are obtained by arranging and combining different exhaust intervals and durations; where (0, 0.7] A / cm 2 The exhaust intervals at different current densities are relatively long, including 10 s, 20 s, and 30 s, with exhaust durations of 0.4 s, 0.6 s, 0.8 s, 1.0 s, and 1.2 s; (0.7, 1.0] A / cm 2 The exhaust intervals at different current densities included 7 s, 14 s, and 18 s, and the exhaust durations included 0.4 s, 0.6 s, 0.8 s, 1.0 s, and 1.2 s; while at higher current densities such as >1.0 A / cm 2 The exhaust strategy settings include short time intervals of 4 s, 8 s, and 12 s, and exhaust durations of 0.4 s, 0.6 s, 0.8 s, 1.0 s, and 1.2 s.

[0131] Specifically, the selected fuel cell system has a rated power of 100 kW, a rated current of 400 A, and 450 individual cells.

[0132] Specifically, the prediction time domain in the model predictive controller is set to 15, the control time domain is 5, and the sampling period is 10ms.

[0133] Specifically, Figure 4 (a) to (e) show the anti-surge results of air compressors at altitudes of 0 m, 1000 m, 2000 m, 3000 m, and 4000 m, respectively. Figure 4Figure (f) compares the anti-surge results under different control methods. A value of 1 indicates successful surge suppression, and a value of 0 indicates failure. The figure shows that the designed AMPC effectively constrains the surge phenomenon of the air compressor, keeping the operating trajectory within a safe envelope. Other MPC strategies without adaptive model updates all exhibit air compressor surge problems. This comparison fully verifies the important role of adaptive model update strategies in improving the operational reliability of air compressors.

[0134] Specifically, Figure 7 (a) shows the anode pressure control results under the CHTC operating condition at an altitude of 4000m. Figure 7 (b) Changes in anode efficiency and nitrogen concentration under operating conditions at an altitude of 4000m & CHTC. Figure 7 (c) shows the anode hydrogen ratio control results under the 4000m altitude & CHTC condition. Simulation data shows that the anode pressure tracking absolute error of the high-altitude strategy is reduced to 3.9183 kPa / cycle, a decrease of 4.21% compared to the plain strategy. In terms of system efficiency, the high-altitude strategy outperforms the plain strategy overall, with an average efficiency improvement of 0.20%. Although emissions are reduced, the accelerated exhaust gas discharge in the high current density region keeps the nitrogen concentration stable at 3.10%. The study also found that the larger internal and external pressure difference of the system at high altitudes interferes with the exhaust strategy's regulation of HER, but the HER control AAE of the high-altitude strategy is still reduced to 0.0105, a decrease of 5.41% compared to the plain strategy.

[0135] As can be seen, this invention addresses the significant problems of compressor surge and anode pressure fluctuation in variable altitude environments by proposing a method for coordinated control of cathode surge suppression and anode dynamic exhaust in variable altitude environments, thereby improving the operational reliability of fuel cell systems in high-altitude environments.

[0136] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for optimizing the anode and cathode gas control of a variable altitude fuel cell, characterized in that, The method comprises the following steps: constructing a fuel cell stack model, an air supply system model, a hydrogen supply system model and an atmospheric environment parameter model; the air supply system model comprises an air compressor MAP, an intake / exhaust manifold, a cathode flow channel and a back pressure valve; the hydrogen supply system model comprises a solenoid valve, a proportional valve, an ejector, a hydrogen supply manifold, an anode flow channel, a return manifold, an exhaust valve and a hydrogen circulation pump; and the atmospheric environment parameter model comprises an environmental pressure, an environmental temperature and an air density; the air supply system model and the hydrogen supply system model are both nonlinear models; performing adaptive predictive control on a cathode over-oxygen ratio and a cathode pressure of the fuel cell according to changes of a safe working area of the air compressor in a variable altitude environment and based on a load current and an altitude range; setting a double surge judgment mechanism for the air compressor, and correcting control variables of the air supply system in combination with working states of the air compressor in a prediction time domain; designing coordinated control on an anode over-hydrogen ratio and an anode pressure of the fuel cell based on a super-spiral algorithm with the cathode pressure as a reference; taking an average efficiency, a maximum efficiency, an average single-cell voltage of the fuel cell stack, an anode pressure fluctuation error and a nitrogen concentration as evaluation indexes, and selecting an optimal anode exhaust strategy in the variable altitude environment by using an entropy weight TOPSIS evaluation system.

2. The method of claim 1, wherein, The adaptive predictive control on the cathode over-oxygen ratio and the cathode pressure of the fuel cell according to changes of the safe working area of the air compressor in the variable altitude environment and based on the load current and the altitude range comprises the following steps: obtaining a safe working boundary of the air compressor in the variable altitude environment, which is expressed as: ； where W surge,h and W choke,h are the surge and choke flow rates at an altitude of h for an air compressor outlet pressure p comp,h c 00,h , c 01,h , c 02,h , c 10,h , c 11,h , c 12,h are fitting parameters; dividing the load current and the altitude into intervals, performing Taylor series expansion on the air supply system model in each interval, establishing a linearized approximate model by truncating high-order components, and expressing the linearized approximate model as: ； wherein A, B u , B d , C, D u , D d , B w , D w are Jacobian matrices; X(t) is a system state variable, is a system state variable differential, is a system output variable, is a system control variable, is a load disturbance, is a linearization error disturbance, and t is a time variable. wherein Δ represents an increment, k+1 and k represent a k+1 time and a k time respectively, and dis represents a discretization process; ； constructing a prediction equation based on output variables in a prediction time domain according to the linear discrete-time state space model in the increment form, so as to establish a target function for regulating the over-oxygen ratio and the cathode pressure, which is expressed as: The interval division of the load current and the altitude is specifically as follows: ； where Γ y is a prediction weighting factor, Γ u is a control weighting factor, S u is a system state coefficient matrix, E p is an error in the prediction step, denotes the square of the norm.

3. The method of claim 2, wherein, the load current is divided into five intervals, including (0, 100] A, (100, 200] A, (200, 300] A, (300, 400] A and (400, 450] A; the altitude is divided into five intervals, including (0, 800] m, (800, 1600] m, (1600, 2400] m, (2400, 3200] m and (3200, 4000] m. The double surge judgment mechanism for the air compressor is set, and the control variables of the air supply system are corrected in combination with the working states of the air compressor in the prediction time domain, which comprises the following steps:

4. The method for co-optimization of anode and cathode gas of a variable-altitude fuel cell according to claim 1, characterized in that, a first surge early warning detector is set, which is expressed as: a second surge early warning detector is set, which is expressed as: ； wherein J 1,surge,k,i is the decision signal of the first surge detector of the i-th prediction step in the k-th control step, W comp,k,i and p comp,k,i are the flow rate and pressure of the predicted air compressor output of the i-th prediction step in the k-th control step, W surge,k,i is the surge decision signal corresponding to the air compressor output pressure of the i-th prediction step in the k-th control step; ​ ； ； wherein J 2,surge,k is the decision signal of the second surge detector at the kth control step, NC Wcomp,low is the low threshold of the normalized change rate of the air compressor flow; NC Wcomp,k represents the normalized change rate of the air compressor flow, the change rate of the air compressor flow is C Wcomp,k , the normalized result of C s is the sampling time, W comp,k, is the predicted flow of the air compressor output at the kth control step; In the subsequent kth control step, when J 1,surge,k,i = 1 and J 2,surge,k = 1, the controller adjusts both control variables, the current air compressor voltage and the back pressure valve opening, according to the predicted sequence: ； wherein u comp,af,k and u comp,bef,k and u comp,cor,k are the corrected working voltage, the uncorrected working voltage and the working voltage correction amount, respectively, of the air compressor u bpv,af,k and u bpv,bef,k and u bpv,cor,k are the corrected back pressure valve opening, the uncorrected back pressure valve opening and the back pressure valve opening correction amount, respectively.

5. The method of claim 4, wherein, The adjustment of the two control variables, the current air compressor voltage and the back pressure valve opening, based on the predicted sequence is as follows: the surge line of 110% is used as the surge warning line, that is, the value of the control variable = 110% × |surge value - predicted control output value|.

6. The method of co-optimized control of anode and cathode gases for a variable-altitude fuel cell of claim 1, wherein, The method for coordinated control of the anode hydrogen ratio and anode pressure in a fuel cell, using the cathode pressure as a reference and based on a superspiral algorithm, includes the following steps: The sliding variables, including sliding mode variables s1(x) and s2(x), are directly calculated using the anode-cathode pressure difference and the hydrogen permeation ratio deviation. Sliding mode variable s1(x) represents the difference between the target cathode pressure and the anode pressure, and sliding mode variable s2(x) represents the difference between the target hydrogen permeation ratio and the actual feedback hydrogen permeation ratio of the system. The control variable u of the anode system is designed based on the sliding variable, and is expressed as: ； wherein u = [u fcv , u pump ] T , s = [s1, s2] T , γ = [γ1, γ2] T , α = [α1, α2] T , u fcv is the proportional valve opening, u pump is the hydrogen circulation pump speed, γ and α are both convergence coefficients, γ1 and α1 are the convergence coefficients of the sliding mode variable s1(x), γ2 and α2 are the convergence coefficients of the sliding mode variable s2(x); sign represents the sign function.

7. The method of claim 1, wherein, The method uses the average efficiency, maximum efficiency, average single-cell voltage of the fuel cell stack, anode pressure fluctuation error, and nitrogen concentration as evaluation indicators, and employs the entropy-weighted TOPSIS evaluation system to select the optimal anode venting strategy under varying altitude environments, including the following steps: Evaluation metrics for designing an anode system include average efficiency, maximum efficiency, average single-cell voltage of the fuel cell stack, anode pressure fluctuation error, and nitrogen concentration. Data is standardized according to the direction of indicator attributes, and information entropy and weight values ​​are calculated. The weight values ​​are multiplied by the data to obtain new data, which is then calculated using the TOPSIS method. A segmented analysis of the current density was performed, and several exhaust strategies were set for each current range. After comprehensive evaluation, the optimal exhaust strategy for different current densities and altitude environments was obtained.

8. The method for co-optimization of anode and cathode gas of a variable-altitude fuel cell according to claim 7, characterized in that, The formula for calculating the efficiency of the anode system is as follows: ； wherein η an is the anode system efficiency, P st is the stack power, P pump is the power consumption of the hydrogen circulation pump, LHV H2 is the lower heating value of hydrogen, is the outlet hydrogen flow of the anode exhaust valve, cycle is the hydrogen flow consumed by the reaction, cycle.

9. The method of claim 7, wherein, The anode pressure fluctuation error is expressed as: ； wherein, is the anode pressure fluctuation error, p an is the anode pressure, p ref is the pressure reference value.

10. The method of claim 7, wherein, The process involves segmented analysis of current density, setting several exhaust strategies for each current range, and comprehensively evaluating the optimal exhaust strategy for different current densities and altitude environments. This includes the following steps: The current density in the segmented analysis is divided into four intervals, including (0, 0.7] A / cm 2 ,(0.7, 1.0] A / cm 2 ,(1.0, 1.3] A / cm 2 ,(1.3, 1.6] A / cm 2 ; 15 exhaust strategies are set under each current interval, wherein, (0, 0.7] A / cm 2 The exhaust interval under the current density includes 10 s, 20 s and 30 s, and the exhaust duration includes 0.4 s, 0.6 s, 0.8 s, 1.0 s and 1.2 s; (0.7, 1.0] A / cm 2 The exhaust interval under the current density includes 7 s, 14 s and 18 s, and the exhaust duration includes 0.4 s, 0.6 s, 0.8 s, 1.0 s and 1.2 s; >1.0 A / cm 2 The exhaust interval under the current density includes 4 s, 8 s and 12 s, and the exhaust duration includes 0.4 s, 0.6 s, 0.8 s, 1.0 s and 1.2 s.

Images