A self-adaptive dynamic surface control method for air supply system of hydrogen fuel cell

Abstract

The application relates to a hydrogen fuel cell air supply system adaptive dynamic surface control method, aiming at the peroxide ratio control problem of a proton exchange membrane fuel cell under the air compressor fault condition, first, considering the high temperature, friction and pipeline air leakage fault of the air compressor, a dynamic model of the proton exchange membrane fuel cell air supply system is established; secondly, a specified performance function is designed and error conversion is carried out to obtain an unconstrained error model; thirdly, a virtual control quantity is designed, a first-order filter is introduced to replace the derivative of the virtual control quantity; finally, an air compressor input voltage adaptive control instruction is designed to complete the adaptive dynamic surface control of the proton exchange membrane fuel cell air supply system, the peroxide ratio control of the proton exchange membrane fuel cell under the variable load and air compressor fault condition is realized, the method has the characteristics of high control precision, strong stability and easy engineering implementation, and is suitable for the peroxide ratio tracking control of the proton exchange membrane fuel cell air supply system.

Description

Technical Field

[0001] This invention belongs to the field of hydrogen fuel cell air supply system control, specifically relating to an adaptive dynamic surface control method for a hydrogen fuel cell air supply system, which solves the problem of oxygen ratio tracking control in a proton exchange membrane fuel cell air supply system with air compressor high temperature, friction, and pipeline air leakage faults. Background Technology

[0002] Proton exchange membrane fuel cells (PEMFCs) generate electricity through the reaction of hydrogen and oxygen on a proton exchange membrane. They are highly efficient power generation devices characterized by high power density, zero pollution, and low operating temperature, and are commonly used in systems such as electric vehicles. The oxygen permeate ratio (OPR) is a crucial indicator for evaluating system performance. PEMFCs are prone to oxygen starvation and oxygen saturation under varying load conditions. Oxygen starvation leads to voltage drop, flooding of the stack, and shortened battery life; oxygen saturation causes membrane dryness, reducing PEMFC efficiency. Therefore, controlling the OPR to track the optimal reference value is key to improving the performance and efficiency of PEMFC systems. The air compressor is a core component of the air supply system. Its speed can be adjusted by controlling the input voltage, thereby changing the output airflow and achieving the desired OPR. However, the air compressor is a complex, highly nonlinear system, and in actual operation, it is susceptible to faults including high temperature, friction, and air leaks in the pipeline. Prolonged high-temperature operation caused by factors such as clogged air filters and radiators can increase the resistance of the air compressor motor. Increased friction in the mechanical components of an air compressor leads to an increase in the motor's back electromotive force constant. Furthermore, the frictional force continues to increase as the motor rotors mesh, causing surge in the air compressor. Under high pressure, frequent air leaks in the pipeline reduce the compressor's speed and affect its outlet gas flow rate. Therefore, adaptive dynamic surface control of the hydrogen fuel cell air supply system is a crucial technology, significantly contributing to improving the tracking performance of the air-to-oxygen ratio.

[0003] Currently, research on the oxygen ratio control problem in proton exchange membrane fuel cell (PEMFC) air supply systems mainly focuses on oxygen ratio tracking under varying loads, with limited research on technologies considering air compressor failures and ensuring specified oxygen ratio tracking performance. The paper "Adaptive High-Order Sliding Mode Control of Oxygen Ratio in PEMFC Systems" uses feedback linearization of the nonlinear model of the air supply system, and ensures that the system tracking error converges to the neighborhood of the origin through online adaptive control gain. However, feedback linearization needs to consider the stability of the system's internal dynamics and is highly dependent on the accuracy of the nonlinear model. The paper "Hierarchical Control of Oxygen Ratio in PEMFC Systems Based on Maximum Net Power" addresses the oxygen ratio control problem by designing an upper-level controller based on an offline + online oxygen ratio optimization algorithm and a lower-level controller using super-spiral sliding mode control, achieving oxygen ratio control under dynamic load current changes. However, for offline data optimization, this method suffers from the problem of optimal performance point offset; while some online optimization methods have problems such as complex algorithms and large computational loads. Chinese patent application CN201811345140.7 utilizes an extended state observer to adjust the oxygen ratio output estimation error in real time, converging the output error to zero within a finite time while simultaneously achieving accurate estimation of the unmeasurable cathode pipe pressure. However, this method simplifies uncertainties and external disturbances in the system to lumped disturbances primarily affecting the gas supply manifold, resulting in low accuracy in system modeling and oxygen ratio output calculation. Chinese patent application CN202211000472.8 achieves coordinated control and rapid response of intake flow and intake pressure by establishing and training a fuzzy neural network model. However, this method does not consider the impact of gas supply system failures and suffers from a large computational burden.

[0004] In summary, existing methods lack effective oxygen ratio tracking and control capabilities under conditions of high temperature in air compressors, friction, and air leakage in pipelines. There is an urgent need to develop an oxygen ratio control method for hydrogen fuel cell air supply systems based on adaptive fault-tolerant technology. Summary of the Invention

[0005] To address the oxygen ratio control problem in proton exchange membrane fuel cell air supply systems with issues such as high temperature and friction from the air compressor, as well as air leakage in the pipeline, this invention overcomes the shortcomings of existing technologies by providing an adaptive dynamic surface control method for hydrogen fuel cell air supply systems. This method enables adaptive adjustment of controller parameters under varying loads and air compressor failures. By designing a specified performance function and performing error transformation, it ensures that the oxygen ratio of the proton exchange membrane fuel cell air has a specified tracking performance, thereby improving the working efficiency of the fuel cell and the stability of the system.

[0006] To achieve the above objectives, the present invention adopts the following technical solution:

[0007] An adaptive dynamic surface control method for a hydrogen fuel cell air supply system, characterized by comprising the following steps:

[0008] The first step is to establish a dynamic model of the air supply system for a proton exchange membrane fuel cell, considering the high temperature, friction, and air leakage in the pipeline of the air compressor. The high temperature affects the motor resistance of the air compressor, the friction affects the back EMF constant of the motor, and the air leakage in the pipeline affects the speed of the air compressor.

[0009] The second step is to define the tracking error of the system based on the dynamic model of the proton exchange membrane fuel cell air supply system in the first step, design a specified performance function and perform error transformation to obtain the first unconstrained error model.

[0010] The third step is to design a virtual control quantity based on the unconstrained error model in the second step, and introduce a first-order filter to replace the derivative of the virtual control quantity.

[0011] The fourth step involves defining a second error model based on the first-order filter from the third step, designing an adaptive control command for the air compressor input voltage, and completing the adaptive dynamic surface control of the proton exchange membrane fuel cell air supply system.

[0012] The advantages of this invention compared to the prior art are as follows:

[0013] This invention relates to an adaptive dynamic surface control method for a hydrogen fuel cell air supply system. Addressing the shortcomings of existing methods in effectively tracking the oxygen ratio under conditions of high compressor temperature, friction, and pipeline air leakage, this method establishes a dynamic model of the proton exchange membrane fuel cell air supply system considering compressor failures. A first-order filter is introduced to replace the derivative of the virtual control quantity, simplifying the controller structure and avoiding the differential explosion phenomenon in the backstepping method. A specified performance function is designed and error transformation is performed. Through adaptive adjustment of the controller parameters, a specified tracking performance of the proton exchange membrane fuel cell air oxygen ratio is achieved under varying loads and time-varying compressor failures, ensuring system efficiency and stability. This method features high control accuracy and ease of engineering implementation, and is suitable for oxygen ratio control in proton exchange membrane fuel cell air supply systems. Attached Figure Description

[0014] Figure 1 This is a flowchart illustrating the design of an adaptive dynamic surface control method for a hydrogen fuel cell air supply system according to the present invention. Detailed Implementation

[0015] The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0016] like Figure 1As shown, this invention relates to an adaptive dynamic surface control method for a hydrogen fuel cell air supply system, comprising the following steps:

[0017] The first step is to establish a dynamic model of the air supply system for a proton exchange membrane fuel cell, considering the high temperature, friction, and air leakage in the pipeline of the air compressor. The high temperature affects the motor resistance of the air compressor, the friction affects the back EMF constant of the motor, and the air leakage in the pipeline affects the speed of the air compressor.

[0018] The second step is to define the tracking error of the system based on the dynamic model of the proton exchange membrane fuel cell air supply system in the first step, design a specified performance function and perform error transformation to obtain the first unconstrained error model.

[0019] The third step is to design a virtual control quantity based on the unconstrained error model after performance conversion specified in the second step, and introduce a first-order filter to replace the derivative of the virtual control quantity.

[0020] The fourth step involves defining a second error model based on the first-order filter from the third step, designing an adaptive control command for the air compressor input voltage, and completing the adaptive dynamic surface control of the proton exchange membrane fuel cell air supply system.

[0021] This invention employs an adaptive dynamic surface control method with specified performance constraints, avoiding the differential explosion phenomenon in the backstepping method. It can ensure that the system tracking error has specified convergence performance under variable load and air compressor failure conditions, and has the characteristics of high stability and easy engineering implementation. It is suitable for the oxygen ratio control of proton exchange membrane fuel cell gas supply systems.

[0022] Specifically, in the first step, based on the physical models of the air compressor and intake manifold, and grounded in the ideal gas law and the law of conservation of mass, a third-order nonlinear dynamic model of the proton exchange membrane fuel cell air supply system is established:

[0023]

[0024] Where, x1 = p ca The initial value of the cathode gas pressure for the air supply system is 105000 Pa; x2 = p sm The initial gas pressure in the gas supply manifold is 110,000 Pa; x3 = ω cp The angular velocity of the air compressor is initially 4000 rad / s; as well as Let x1, x2, and x3 represent the derivatives, respectively; I st The load current of the fuel cell is h(x2,x3) = W. cp τ is the outlet gas flow rate of the air compressor; cm The driving torque of the air compressor; a iFor system model coefficients, i = 1, 2, ..., 11.

[0025] Considering three fault scenarios for the air compressor—high temperature, friction, and air leakage in the pipeline—the air compressor driving torque τ cm The expression is:

[0026]

[0027] Where, η cm The efficiency of the drive motor is set to 98%; k t k v For motor constants, ... cm The resistance of the motor is 1.3Ω; v cm This refers to the input voltage for the air compressor. and These represent the changes in air compressor dynamic parameters caused by high temperature, friction, and air leakage in the pipeline.

[0028] Specifically, in the second step, the unconstrained error model is:

[0029]

[0030] Among them, a 12 and a 13 It is a constant; This represents the actual oxygen ratio of the system. For reference, the peroxide ratio is set to 2; e y (t) represents the system output tracking error; p(t) is the specified performance function; z1 is the transformed unconstrained error model; p0 > p ∞ >0, m>0, and All are preset constants, with values ​​of 14000, 750, 2, 1 and 1 respectively. and These represent the upper and lower bounds of the error, respectively. The steady-state value of error convergence is limited, and the exponent m determines the convergence rate of the error.

[0031] Specifically, in the third step, based on the unconstrained error model after performance conversion specified in the second step, a virtual control quantity is designed, and a first-order filter is introduced to replace the derivative of the virtual control quantity:

[0032]

[0033] Among them, B, S1, η and h is an intermediate variable defined. *c1 is the virtual control variable; c1 > 0 is the design parameter with a value of 10; ι is the time constant of the first-order filter with a value of 7; ε is the output of the filter.

[0034] Specifically, the fourth step includes:

[0035] Define the second error model and the parameters to be estimated:

[0036]

[0037] Where z2 is the second error model; S2, H(t), G(t), φ(t) and g0 are defined intermediate variables; θ and β are parameters to be estimated; and Let H(t) and φ(t) be the supremum of their respective values. It represents the infimum of |G(t)|.

[0038] The design parameters, adaptive law, and air compressor input voltage control commands are as follows:

[0039]

[0040] Here, α and ζ are defined intermediate variables; and θ and β are estimated values; Υ>0, γ1>0, γ2>0, σ1>0 and σ2>0 are design parameters, with values ​​of 3000, 0.3, 0.001, 350 and 1000 respectively.

[0041] Using the method of this invention for oxygen ratio control can ensure that the tracking error of the system has a specified convergence performance under variable load and air compressor failure conditions, which is beneficial to improving system stability.

[0042] The contents not described in detail in this specification are existing technologies known to those skilled in the art.

Claims

1. An adaptive dynamic surface control method for a hydrogen fuel cell air supply system, characterized in that, Includes the following steps: The first step involves establishing a dynamic model of the proton exchange membrane fuel cell air supply system, considering factors such as high temperature in the air compressor, friction, and air leakage in the pipeline. High temperature affects the motor resistance of the air compressor, friction affects the back electromotive force constant of the motor, and air leakage in the pipeline affects the air compressor speed. (1) Based on the physical models of the air compressor and intake manifold, and on the ideal gas law and the law of conservation of mass, a third-order nonlinear dynamic model of the air supply system of the proton exchange membrane fuel cell is established: ； in, The cathode gas pressure of the air supply system; The gas pressure in the gas supply manifold; The angular velocity of the air compressor; , as well as They represent , and The derivative; This refers to the fuel cell load current. This refers to the outlet gas flow rate of the air compressor. This refers to the driving torque of the air compressor. For system model coefficients, ; (2) Considering the three fault scenarios of high temperature, friction, and air leakage in the pipeline, the driving torque of the air compressor is... The expression is: ； in, For the efficiency of the drive motor; and For motor constants; The motor resistance; This refers to the input voltage for the air compressor. , and These are the changes in air compressor dynamic parameters caused by high temperature, friction, and air leakage in the pipeline. The second step involves defining the system's tracking error based on the proton exchange membrane fuel cell air supply system dynamics model from the first step, designing a specified performance function, and performing error transformation to obtain the first unconstrained error model. The first unconstrained error model is as follows: ； in, and It is a constant; and These are the actual oxygen ratio and the reference oxygen ratio of the system, respectively. The system outputs the tracking error; Specify the performance function; This is the first unconstrained error model; , , and All are preset constants; and These represent the upper and lower bounds of the error, respectively. The steady-state value of error convergence is limited, and the exponential value is also limited. This determines the convergence rate of the error; The third step is to design a virtual control quantity based on the unconstrained error model in the second step, and introduce a first-order filter to replace the derivative of the virtual control quantity. The fourth step involves defining a second error model based on the first-order filter from the third step, designing an adaptive control command for the air compressor input voltage, and completing the adaptive dynamic surface control of the proton exchange membrane fuel cell air supply system.

2. The adaptive dynamic surface control method for a hydrogen fuel cell air supply system according to claim 1, characterized in that: In the third step, based on the unconstrained error model after performance conversion specified in the second step, a virtual control quantity is designed, and a first-order filter is introduced to replace the derivative of the virtual control quantity: ； in, , , and For the definition of intermediate variables; This is a virtual control variable; For design parameters; The time constant of the first-order filter; This is the output of the filter.

3. The adaptive dynamic surface control method for a hydrogen fuel cell air supply system according to claim 1, characterized in that: The fourth step includes: (1) Define the second error model and the parameters to be estimated: ； in, This is the second error model; , , , and For the definition of intermediate variables; and The parameters to be estimated; and They represent and The upper limit, express The infimum; (2) The design parameters, adaptive law, and air compressor input voltage control command are as follows: ； in, and For the definition of intermediate variables; and for and The estimated value; , , , and These are design parameters.

Images