EIS online detection system and method for high-power fuel cell system

Abstract

The invention provides an EIS on-line detection system and method for a high-power fuel cell system, a main loop of the EIS on-line detection system adopts two Boost type converter topological structures which are connected in parallel, and small signal excitation injection is carried out through an auxiliary DC / DC converter 2; the control system comprises a sampling system, a PWM drive circuit, a communication module and an embedded controller. Establishing a battery equivalent circuit model; when a high-power fuel cell system runs, a multi-frequency sine periodic signal optimized by a non-dominated sorting genetic algorithm improved based on Tent chaotic mapping is injected into an output current of an auxiliary DC / DC converter 2; sampling the voltage of a single body or a galvanic pile, measuring current data, calculating the voltage drop corresponding to the internal resistance of the current data, converting a time domain signal into a frequency domain signal, and obtaining the battery impedance information in the corrected Foouquet equivalent circuit model; importing the characteristic parameter data into a least square support vector machine (LSSVM) for training and testing, and performing in-situ detection and fault diagnosis on line; according to the invention, a new way can be provided for active diagnosis and early warning of faults.

Description

Technical Field

[0001] The present invention relates to the technical field of fuel cell systems, and in particular to an EIS online detection system and method for a high-power fuel cell system. Background Art

[0002] The widespread application of hydrogen in the fields of transportation and energy provides an effective way to achieve the recycling of carbon-free energy. Among them, fuel cells, as the core technology in the development and application of hydrogen energy, play a key role. Fuel cells have the advantages of high energy efficiency, good environmental friendliness, and high dynamic response. However, under different operating conditions, fuel cells often face frequent start-stop and large discharge depth, which puts high demands on their performance, especially in terms of fast start-up and dynamic response capabilities. Frequent start-stop and drastic working changes can easily lead to failures such as membrane drying and flooding in fuel cells, affecting their stability and life. High-power fuel cells have a large output current, so small signal excitation currents are easily submerged by high DC bias and difficult to detect. Therefore, it is necessary to accurately detect the operating state parameters of the fuel cell stack and each single chip to provide data support for the fault diagnosis system. However, due to the multi-physical field coupling characteristics and material anisotropy of fuel cells, their internal state parameters are difficult to obtain directly. Electrochemical impedance spectroscopy (EIS) can superimpose various disturbance excitation signals on the fuel cell, and use the electrochemical equivalent circuit model to extract and calculate features, thereby obtaining information on the internal state of the fuel cell, which can effectively monitor and perform fault diagnosis. Summary of the invention

[0003] The present invention proposes an EIS online detection system and method for a high-power fuel cell system, proposes a modified Fouquet equivalent circuit model, and adopts a two-way parallel Boost converter topology structure to monitor and calculate the internal state information of the fuel cell stack during operation, and injects a multi-frequency sinusoidal periodic signal optimized by a non-dominated sorting genetic algorithm based on an improved Tent chaotic map into an auxiliary DC / DC converter, and then uses EIS characteristic data to realize the characterization of partial faults, providing a new way for active diagnosis and early warning of faults.

[0004] The present invention adopts the following technical solutions.

[0005] An EIS online detection system for a high-power fuel cell system includes a main circuit and a control system; the main circuit adopts a two-way parallel Boost converter topology structure, and performs small signal excitation injection through an auxiliary DC / DC converter; the control system includes a sampling system, a PWM drive circuit, a communication module and an embedded controller; a battery equivalent circuit model is established;

[0006] The main circuit includes a fuel cell, a DC / DC converter and a DC load connected in sequence; the control system includes a sampling system, a PWM drive circuit, a communication module and an embedded controller; the sampling system includes a CVM patroller, a voltage sensor, and a high-precision current Hall sensor; wherein the CVM patroller integrated in the sampling system is used to process and calculate the collected fuel cell voltage in real time, and calculate the data of the single cell voltage consistency state, average voltage, and total voltage; the voltage sensor measures the total output voltage of the fuel cell; the current sensor measures the output current of the fuel cell; the communication module is connected to the host computer;

[0007] The main circuit adopts a two-way parallel Boost converter topology structure, one way is for output power transmission to meet the load demand, and the other way is for injection of the multi-frequency sinusoidal periodic excitation signal, and its DC bias is low to avoid the detection difficulty caused by being submerged by the high DC bias of the high-power fuel cell, thereby improving the detection accuracy of small signal excitation.

[0008] A detection method for an EIS online detection system for a high-power fuel cell system, when the high-power fuel cell system is running, a multi-frequency sinusoidal periodic signal optimized by a non-dominated sorting genetic algorithm based on an improved Tent chaotic map is injected into the output current of an auxiliary DC / DC converter; the cell or stack voltage is sampled by a fuel cell voltage inspection CVM, and current data is measured in combination with components such as a Hall current sensor, the voltage drop corresponding to the internal resistance is calculated according to the open circuit voltage and output voltage of the fuel cell, and the time domain signal is converted into a frequency domain signal by a fast Fourier transform; the battery impedance information in a modified Fouquet equivalent circuit model is obtained according to the amplitude and phase fitting at a specific frequency point; the characteristic parameter data is imported into a least squares support vector machine LSSVM for training and testing, and in-situ detection and fault diagnosis are performed online.

[0009] The multi-frequency sinusoidal periodic excitation signal in the diagnostic system is optimized by a non-dominated sorting genetic algorithm based on an improved Tent chaotic map. The phases of sinusoidal signals of different frequencies are optimized and the crest factor of the combined multi-frequency sinusoidal signal is reduced to avoid the problem of reduced EIS measurement accuracy caused by excessive superposition of excitation signal amplitudes.

[0010] The high-power fuel cell is modeled by an equivalent circuit, and the ohmic internal resistance R ohm , activation internal resistance R ct 、Warburg diffusion resistance Z w and constant phase Z CPE The Fouquet model consisting of four components is improved by using a series Foster circuit to replace the Warburg diffusion resistor Z wTo further accurately simulate the low-frequency and high-frequency characteristics of high-power fuel cells, a double-layer distributed capacitor C dl Replace CPE components to reduce the difficulty of fitting, forming an ohmic internal resistance R ohm , activation internal resistance R ct , concentration resistance R w , double layer distributed capacitance C dl and concentration capacitance C w Modified Fouquet equivalent circuit model composed of series and parallel connections; where the ohmic internal resistance R ohm , activation internal resistance R ct Indicates the degree of membrane dryness failure, concentration resistance R w Indicates the degree of flooding fault, double-layer distributed capacitance C dl Represents the electrical characteristics generated by the bipolar plates and electrolyte of the battery stack.

[0011] The online detection method is used to identify the three states of normal, membrane dry and flooded fuel cell stacks;

[0012] The data used for LSSVM model training and model verification comes from experimental samples with clear conclusions and contains state labels; the training set data and test set data are randomly divided into 3:1 ratio;

[0013] The least squares support vector machine (LSSVM) uses the battery ohmic resistance, activation resistance, and impedance at 2, 4, 8, 16, 32, 64, 128, 256, and 1024 frequency points in three states as feature vectors, and the label values ​​are normal, dry membrane, and flooded, corresponding to the numbers 1, 2, and 3 respectively.

[0014] In the detection method, the sampling system collects voltage and current information, and converts the analog signal into a digital signal in real time through an A / D converter. One signal enters the communication module and is input into the embedded controller for pre-data processing and calculation, and the other signal is input into the DC / DC controller for PID calculation and voltage or current stabilization control through the PWM drive circuit.

[0015] The communication module uses the CAN field bus protocol to transmit real-time data to the host computer and information storage medium for display and storage respectively.

[0016] The fuel cell voltage monitoring CVM interacts directly with the communication module without going through an A / D converter.

[0017] The method comprises the following steps:

[0018] Step S1, assigning values ​​to the initial population by using the NSGA-II genetic algorithm improved based on Tent chaotic mapping to improve the convergence ability of the algorithm; using the improved algorithm to optimize the phase of the sinusoidal excitation signal at different frequencies; wherein the optimization target is the crest factor, that is, the ratio of the signal peak value to twice the effective value;

[0019] The frequency values ​​of each sinusoidal component of the sinusoidal excitation signal are 2, 4, 8, 16, 32, 64, 128, 256, and 1024 respectively, the amplitude value of each sinusoidal component is 4A, and the phase value defaults to 0rad;

[0020] Step S2, establishing a modified Fouquet equivalent circuit model that describes the Nyquist characteristics of the fuel cell; wherein the internal resistance of the fuel cell is simplified to be composed of the ohmic internal resistance R ohm , activation internal resistance R ct , concentration resistance R w , double layer distributed capacitance C dl and concentration capacitance C w Circuits composed of series and parallel connections;

[0021] Step S3, superimposing the optimized multi-frequency sinusoidal excitation signal on the output current of the auxiliary DC / DC converter of the fuel cell system in stable operation through the PWM drive circuit;

[0022] Step S4, after the excitation signal is injected, the sampling system synchronously collects and obtains a discrete sequence of response voltage and excitation current, and then transforms them into a frequency domain voltage complex sequence and a frequency domain current complex sequence through fast Fourier transform; the impedance corresponding to the excitation frequency is calculated through the impedance spectrum, that is, the sinusoidal components of each frequency of the excitation current signal and the response voltage of the internal resistance under the sinusoidal excitation of each frequency, and the amplitude and phase information of the impedance corresponding to the frequency point are obtained, and the Nyquist curve is fitted; the single cell voltage signal collected by the CVM directly enters the communication module and is uploaded to the host computer;

[0023] Step S5, the calculated impedance information is used as the original data set for fault diagnosis training and testing, which includes 20 types of feature data and 1 label column; the data set is input into the LSSVM vector machine to complete the fault diagnosis.

[0024] Step S1 specifically includes the following steps:

[0025] The function of the multi-frequency sinusoidal excitation signal is:

[0026]

[0027] Where: f(t) is the frequency of the excitation model at time t.

[0028] The initialization population of the NSGA-II genetic algorithm is composed of Tent chaotic mapping.

[0029] The optimization objective of the NSGA-II genetic algorithm is defined as the crest factor, which represents the uniformity of the signal amplitude distribution in the entire time domain, and its calculation formula is:

[0030]

[0031] Where: I ac,max is the maximum value of the multi-frequency sinusoidal excitation signal; I ac,min is the minimum value of the multi-frequency sinusoidal excitation signal; I eff is the effective value of the multi-frequency sinusoidal excitation signal. eff The expression is:

[0032]

[0033] Furthermore, step S2 specifically includes the following steps:

[0034] The modified Fouquet equivalent circuit model impedance can be expressed as:

[0035]

[0036] The real and imaginary parts of the impedance are calculated according to the above formula, and the Nyquist curve is drawn in the complex plane according to the relationship between the real and imaginary parts of the modified Fouquet equivalent circuit model at various frequencies to represent the electrochemical impedance spectrum of the fuel cell.

[0037] Step S4 specifically includes the following steps:

[0038] The sampling system includes the main components of CVM patroller, Hall current sensor, voltage sensor and precision resistor, analog-to-digital converter ADC and other components required for sampling; wherein, the CVM patroller can monitor the inside of the battery stack, measure the current operating status of a single battery and transmit real-time data to the main controller through the field bus protocol. The detailed voltage data measured by the CVM patroller can be used for fault diagnosis. At the same time, when problems such as voltage imbalance and unit failure occur, a fault alarm can be directly issued to trigger and record the voltage data at the time of the fault; step S5 specifically includes the following steps:

[0039] The LSSVM simplifies the standard SVM by converting the constraints into a square loss function to simplify the optimization process;

[0040] According to the SRM criterion, the objective function and constraints of LSSVM are:

[0041]

[0042] In the formula, ξ iis the prediction error; c is the regularization parameter or penalty parameter used to control the prediction error ξ i The degree of sample penalty. LSSVM handles errors by minimizing the square loss, avoiding the introduction of slack variables; by constructing the Lagrange transformation polynomial, the following transformation is performed:

[0043]

[0044] For the above formula w, b, e k , α k By taking partial derivatives and setting the derivatives to zero, we can get a system of linear equations. The calculation formula of the LSSVM algorithm is:

[0045]

[0046] In the formula, K(x,x i ) is the kernel function.

[0047] The present invention provides an EIS online detection system and method for a high-power fuel cell system, including a main circuit and a control system. The main circuit adopts a two-way parallel Boost converter topology structure, and performs small signal excitation injection through an auxiliary DC / DC converter-2. The control system includes a sampling system, a PWM drive circuit, a communication module and an embedded controller; a battery equivalent circuit model is established; when the high-power fuel cell system is running, a multi-frequency sinusoidal periodic signal optimized by a non-dominated sorting genetic algorithm based on an improved Tent chaotic map is injected into the output current of the auxiliary DC / DC converter-2; the cell or stack voltage is sampled by a fuel cell voltage inspection CVM, and the current data is measured in combination with components such as a Hall current sensor, and the voltage drop corresponding to the internal resistance is calculated according to the open circuit voltage and output voltage of the fuel cell, and the time domain signal is converted into a frequency domain signal by a fast Fourier transform. According to the amplitude and phase fitting at a specific frequency point, the battery impedance information in the modified Fouquet equivalent circuit model is obtained; the characteristic parameter data is imported into the least squares support vector machine LSSVM for training and testing, and in-situ detection and fault diagnosis are performed online; compared with the prior art, the present invention has the following beneficial effects:

[0048] 1. This scheme can realize the functional integration of signal optimization, equivalent circuit parameter identification and fuel cell EIS fault diagnosis, and provides a high-precision high-power fuel cell EIS fault diagnosis method, opening up new ideas for in-situ detection of high-power fuel cells.

[0049] 2. The method described in the present invention adopts a two-way parallel Boost converter topology structure and injects an excitation signal through an auxiliary DC / DC converter, thereby avoiding the influence of a high-power fuel cell DC bias on a small current excitation signal.

[0050] 3. The method described in the present invention can optimize the excitation signal, obtain impedance information of multiple frequencies at one time, improve the speed of EIS time domain measurement, and effectively avoid the influence of excessive amplitude caused by signal superposition.

[0051] 4. The system of the present invention integrates a CVM patroller, which can process and calculate the voltage data of the fuel cell in real time, and obtain parameters such as the single cell voltage consistency state, average voltage and total voltage, providing strong technical support for the in-situ monitoring of high-power fuel cells, EIS online rapid measurement and fault diagnosis technology research and development.

[0052] 5. The present invention measures the EIS of a high-power fuel cell online, monitors and calculates the internal state information of the fuel cell stack during operation, and then uses the impedance characteristic data to characterize some faults, providing a new way for active diagnosis and early warning of faults. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] The present invention is further described in detail below with reference to the accompanying drawings and specific embodiments:

[0054] Attached Figure 1 A schematic diagram of the hardware structure of a preferred embodiment of the present invention;

[0055] Attached Figure 2 A schematic diagram of the control system circuit principle of a preferred embodiment of the present invention;

[0056] Attached Figure 3 This is a schematic diagram of the EIS of a fuel cell in a healthy state obtained by simulation in the embodiment;

[0057] Attached Figure 4 The EIS obtained by simulating the change of ohmic resistance in the embodiment. The solid line part is the EIS of a normal fuel cell, the dotted line part is the EIS of a faulty fuel cell, and the circle dotted line part is the schematic diagram of the EIS of a faulty fuel cell estimated by the FFT algorithm in this solution;

[0058] Attached Figure 5 This is a schematic diagram of EIS obtained by simulating the change of activation resistance in the embodiment;

[0059] Attached Figure 6 Schematic diagram of the fault diagnosis results of the LSSVM support vector machine in the embodiment (the sample category labels 1, 2, and 3 in the figure represent the three states of normal, membrane dryness, and flooding, respectively). DETAILED DESCRIPTION

[0060] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0061] It should be noted that the following detailed descriptions are illustrative and are intended to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present application belongs.

[0062] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application; as used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, they indicate the presence of features, steps, operations, devices, components and / or their combinations.

[0063] As shown in the figure, an EIS online detection system for a high-power fuel cell system includes a main circuit and a control system; the main circuit adopts a two-way parallel Boost converter topology structure, and performs small signal excitation injection through an auxiliary DC / DC converter; the control system includes a sampling system, a PWM drive circuit, a communication module and an embedded controller; a battery equivalent circuit model is established;

[0064] The main circuit includes a fuel cell, a DC / DC converter and a DC load connected in sequence; the control system includes a sampling system, a PWM drive circuit, a communication module and an embedded controller; the sampling system includes a CVM patroller, a voltage sensor, and a high-precision current Hall sensor; wherein the CVM patroller integrated in the sampling system is used to process and calculate the collected fuel cell voltage in real time, and calculate the data of the single cell voltage consistency state, average voltage, and total voltage; the voltage sensor measures the total output voltage of the fuel cell; the current sensor measures the output current of the fuel cell; the communication module is connected to the host computer;

[0065] The main circuit adopts a two-way parallel Boost converter topology structure, one way is for output power transmission to meet the load demand, and the other way is for injection of the multi-frequency sinusoidal periodic excitation signal, and its DC bias is low to avoid the detection difficulty caused by being submerged by the high DC bias of the high-power fuel cell, thereby improving the detection accuracy of small signal excitation.

[0066] A detection method for an EIS online detection system for a high-power fuel cell system, when the high-power fuel cell system is running, a multi-frequency sinusoidal periodic signal optimized by a non-dominated sorting genetic algorithm based on an improved Tent chaotic map is injected into the output current of an auxiliary DC / DC converter; the cell or stack voltage is sampled by a fuel cell voltage inspection CVM, and current data is measured in combination with components such as a Hall current sensor, the voltage drop corresponding to the internal resistance is calculated according to the open circuit voltage and output voltage of the fuel cell, and the time domain signal is converted into a frequency domain signal by a fast Fourier transform; the battery impedance information in a modified Fouquet equivalent circuit model is obtained according to the amplitude and phase fitting at a specific frequency point; the characteristic parameter data is imported into a least squares support vector machine LSSVM for training and testing, and in-situ detection and fault diagnosis are performed online.

[0067] The multi-frequency sinusoidal periodic excitation signal in the diagnostic system is optimized by a non-dominated sorting genetic algorithm based on an improved Tent chaotic map. The phases of sinusoidal signals of different frequencies are optimized and the crest factor of the combined multi-frequency sinusoidal signal is reduced to avoid the problem of reduced EIS measurement accuracy caused by excessive superposition of excitation signal amplitudes.

[0068] The high-power fuel cell is modeled by an equivalent circuit, and the ohmic internal resistance R ohm , activation internal resistance R ct 、Warburg diffusion resistance Z w and constant phase Z CPE The Fouquet model consisting of four components is improved by using a series Foster circuit to replace the Warburg diffusion resistor Z w To further accurately simulate the low-frequency and high-frequency characteristics of high-power fuel cells, a double-layer distributed capacitor C dl Replace CPE components to reduce the difficulty of fitting, forming an ohmic internal resistance R ohm , activation internal resistance R ct , concentration resistance R w , double layer distributed capacitance C dl and concentration capacitance C w Modified Fouquet equivalent circuit model composed of series and parallel connections; where the ohmic internal resistance R ohm , activation internal resistance R ct Indicates the degree of membrane dryness failure, concentration resistance R w Indicates the degree of flooding fault, double-layer distributed capacitance C dl Represents the electrical characteristics generated by the bipolar plates and electrolyte of the battery stack.

[0069] The online detection method is used to identify the three states of normal, membrane dry and flooded fuel cell stacks;

[0070] The data used for LSSVM model training and model verification comes from experimental samples with clear conclusions and contains state labels; the training set data and test set data are randomly divided into 3:1 ratio;

[0071] The least squares support vector machine (LSSVM) uses the battery ohmic resistance, activation resistance, and impedance at 2, 4, 8, 16, 32, 64, 128, 256, and 1024 frequency points in three states as feature vectors, and the label values ​​are normal, dry membrane, and flooded, corresponding to the numbers 1, 2, and 3 respectively.

[0072] In the detection method, the sampling system collects voltage and current information, and converts the analog signal into a digital signal in real time through an A / D converter. One signal enters the communication module and is input into the embedded controller for pre-data processing and calculation, and the other signal is input into the DC / DC controller for PID calculation and voltage or current stabilization control through the PWM drive circuit.

[0073] The communication module uses the CAN field bus protocol to transmit real-time data to the host computer and information storage medium for display and storage respectively.

[0074] The fuel cell voltage monitoring CVM interacts directly with the communication module without going through an A / D converter.

[0075] The method comprises the following steps:

[0076] Step S1, assigning values ​​to the initial population by using the NSGA-II genetic algorithm improved based on Tent chaotic mapping to improve the convergence ability of the algorithm; using the improved algorithm to optimize the phase of the sinusoidal excitation signal at different frequencies; wherein the optimization target is the crest factor, that is, the ratio of the signal peak value to twice the effective value;

[0077] The frequency values ​​of each sinusoidal component of the sinusoidal excitation signal are 2, 4, 8, 16, 32, 64, 128, 256, and 1024 respectively, the amplitude value of each sinusoidal component is 4A, and the phase value defaults to 0rad;

[0078] Step S2, establishing a modified Fouquet equivalent circuit model that describes the Nyquist characteristics of the fuel cell; wherein the internal resistance of the fuel cell is simplified to be composed of the ohmic internal resistance R ohm , activation internal resistance R ct , concentration resistance R w , double layer distributed capacitance C dl and concentration capacitance C w Circuits composed of series and parallel connections;

[0079] Step S3, superimposing the optimized multi-frequency sinusoidal excitation signal on the output current of the auxiliary DC / DC converter of the fuel cell system in stable operation through the PWM drive circuit;

[0080] Step S4, after the excitation signal is injected, the sampling system synchronously collects and obtains a discrete sequence of response voltage and excitation current, and then transforms them into a frequency domain voltage complex sequence and a frequency domain current complex sequence through fast Fourier transform; the impedance corresponding to the excitation frequency is calculated through the impedance spectrum, that is, the sinusoidal components of each frequency of the excitation current signal and the response voltage of the internal resistance under the sinusoidal excitation of each frequency, and the amplitude and phase information of the impedance corresponding to the frequency point are obtained, and the Nyquist curve is fitted; the single cell voltage signal collected by the CVM directly enters the communication module and is uploaded to the host computer;

[0081] Step S5, the calculated impedance information is used as the original data set for fault diagnosis training and testing, which includes 20 types of feature data and 1 label column; the data set is input into the LSSVM vector machine to complete the fault diagnosis.

[0082] Step S1 specifically includes the following steps:

[0083] The function of the multi-frequency sinusoidal excitation signal is:

[0084]

[0085] Where: f(t) is the frequency of the excitation model at time t.

[0086] The initialization population of the NSGA-II genetic algorithm is composed of Tent chaotic mapping.

[0087] The optimization objective of the NSGA-II genetic algorithm is defined as the crest factor, which represents the uniformity of the signal amplitude distribution in the entire time domain, and its calculation formula is:

[0088]

[0089] Where: I ac,max is the maximum value of the multi-frequency sinusoidal excitation signal; I ac,min is the minimum value of the multi-frequency sinusoidal excitation signal; I eff is the effective value of the multi-frequency sinusoidal excitation signal. eff The expression is:

[0090]

[0091] Furthermore, step S2 specifically includes the following steps:

[0092] The modified Fouquet equivalent circuit model impedance can be expressed as:

[0093]

[0094] The real and imaginary parts of the impedance are calculated according to the above formula, and the Nyquist curve is drawn in the complex plane according to the relationship between the real and imaginary parts of the modified Fouquet equivalent circuit model at various frequencies to represent the electrochemical impedance spectrum of the fuel cell.

[0095] Step S4 specifically includes the following steps:

[0096] The sampling system includes the main components of CVM patroller, Hall current sensor, voltage sensor and precision resistor, analog-to-digital converter ADC and other components required for sampling; wherein, the CVM patroller can monitor the inside of the battery stack, measure the current operating status of a single battery and transmit real-time data to the main controller through the field bus protocol. The detailed voltage data measured by the CVM patroller can be used for fault diagnosis. At the same time, when problems such as voltage imbalance and unit failure occur, a fault alarm can be directly issued to trigger and record the voltage data at the time of the fault; step S5 specifically includes the following steps:

[0097] The LSSVM simplifies the standard SVM by converting the constraints into a square loss function to simplify the optimization process;

[0098] According to the SRM criterion, the objective function and constraints of LSSVM are:

[0099]

[0100] In the formula, ξ i is the prediction error; c is the regularization parameter or penalty parameter used to control the prediction error ξ i The degree of sample penalty. LSSVM handles errors by minimizing the square loss, avoiding the introduction of slack variables; by constructing the Lagrange transformation polynomial, the following transformation is performed:

[0101]

[0102] For the above formula w, b, e k , α k By taking partial derivatives and setting the derivatives to zero, we can get a system of linear equations. The calculation formula of the LSSVM algorithm is:

[0103]

[0104] In the formula, K(x,x i ) is the kernel function.

[0105] Embodiment 1

[0106] This example proposes an EIS online detection system and method for a high-power fuel cell system. Figures 1 to 6, including the main circuit, control system, PWM drive circuit, sampling module; establish a battery equivalent circuit model; when the high-power fuel cell system is running, the multi-frequency sinusoidal periodic signal optimized by the non-dominated sorting genetic algorithm based on the improved Tent chaotic map is superimposed on the output current of the fuel cell, and the fuel cell to be tested is stimulated by the PWM drive circuit; the voltage data of each unit of the fuel cell is obtained by online sampling through the fuel cell voltage patrol CVM, and the current data is measured by combining the Hall current sensor and other components, and the voltage drop corresponding to the internal resistance is calculated according to the open circuit voltage and output voltage of the fuel cell, and the time domain signal is converted into a frequency domain signal through the fast Fourier transform. According to the amplitude and phase fitting at a specific frequency point, the battery impedance information is obtained; the characteristic parameter data is imported into the least squares support vector machine LSSVM for training and testing, and in-situ detection and fault diagnosis are performed online.

[0107] The main circuit includes a fuel cell, a main DC / DC converter-1, an auxiliary DC / DC converter-2 and a DC load connected in sequence; the control system includes a sampling system, a PWM drive circuit, a communication module and an embedded controller; the sampling system includes a CVM patroller, a voltage sensor, and a high-precision current Hall sensor; the input end of the PWM drive circuit is connected to the DC / DC controller, and the output end of the PWM drive circuit is connected to the MOS tube in the DC / DC converter; the input ends of the voltage sensor and the high-precision current Hall sensor of the sampling system are both connected to the DC / DC converter, and their output ends are connected to the A / D conversion module.

[0108] The main circuit consists of a fuel cell, two parallel Boost converters and a load. The system adopts an integrated method of DC / DC converter and fuel cell internal resistance test. The two DC / DC converters are used to realize the transmission of fuel cell output power and the superposition of excitation signals. By applying high-frequency and medium-low frequency excitation signals to the voltage loop and current loop of the auxiliary DC / DC converter-2, the output voltage of the PEMFC is collected and compared with the set expected output voltage. The output end of the PWM drive circuit is connected to the gate of the switch tube of the DC / DC converter, and the PWM duty cycle of the MOS tube is adjusted according to the comparison result, thereby adjusting the output signal of the DC / DC converter. The main DC / DC converter-1 ensures the stability of the system output voltage, and the auxiliary DC / DC converter-2 provides the necessary excitation signal for the battery.

[0109] The communication module uses the CAN field bus protocol to transmit real-time data to the host computer and information storage medium for display and storage. The fuel cell voltage inspection CVM can directly interact with the communication module without going through an A / D converter. The communication module is connected to the embedded controller.

[0110] In the main loop, the DC / DC converter is connected to the fuel cell group; the voltage and current output ends of the fuel cell group are directly connected to the sampling system; the output end of the excitation signal module is connected to the input end of the PWM drive circuit; and the output end of the PWM drive circuit is connected to the switch tube of the DC / DC converter.

[0111] The sampling system in the system mainly realizes the function of collecting the voltage and current signals of the fuel cell without the need for additional hardware circuits. Considering the sampling accuracy and the frequency range of the disturbance signal, the battery management system sampling system should use an analog front-end chip with a higher sampling frequency.

[0112] The control system module is used for system control, signal processing, impedance spectrum calculation, fault diagnosis, etc., and can adopt the digital controller or microcomputer in the battery management system.

[0113] The information storage device is used to store the impedance spectrum information obtained from the initial and subsequent measurements of each battery, providing historical information as a basis for battery diagnosis. It can take the form of various products that can implement computer programs, such as disk storage, CD-ROM, flash memory, optical storage, etc.

[0114] In Simulink software or other fitting software, a DC / DC converter circuit and a fuel cell modified Fouquet equivalent circuit model are constructed, and multi-frequency sinusoidal superposition signals with frequencies of 2, 4, 8, 16, 32, 64, 128, 256, and 1024 Hz are added. According to the method described in the invention, the voltage and current data of the battery stack and the single cell are obtained through a sampling system with a sampling frequency of 2500 Hz.

[0115] The operating parameters of the fuel cell stack are adjusted to make it appear membrane dry and water flooded. The same detection conditions and methods as the EIS of a healthy fuel cell are used to obtain the EIS information of the battery in the two states of membrane dry and water flooded. After data preprocessing, curve feature point and feature value extraction and other processes, the changes in the spectrum can be clearly observed by comparing with the normal battery EIS.

[0116] like Figures 4 to 5 As shown in the figure, by setting the test operating conditions under different fault conditions, the parameters of each component of the fuel cell equivalent model can be obtained, and the parameter values ​​of the Fouquet equivalent circuit model can be corrected in the simulation. By simulating the changes in the battery ohmic resistance, activation resistance, and impedance at each frequency point, and using the same detection conditions and methods, the EIS information estimated by the algorithm after the impedance changes can be obtained. By intuitively comparing the changes in the battery EIS after the ohmic resistance and charge transfer resistance are changed, the impact of these parameters on the battery performance can be clearly observed.

[0117] In addition, the impedance information at different frequency points in the impedance spectrum can effectively characterize different fault processes, especially the changes in water content inside the fuel cell stack. This enables us to deeply understand the performance of the battery under different working conditions by comparing and analyzing the impedance spectra of fuel cells under different water content states, providing strong data support and basis for fault diagnosis.

[0118] like Figure 4 As shown in the figure, when the ohmic resistance increases, the EIS image shifts to the right along the real axis relative to the EIS image of a healthy battery, which may indicate that a membrane dry failure has occurred in the battery. The membrane dry failure hinders the conduction ability of water molecules inside the battery stack as proton carriers, which in turn significantly increases the ohmic impedance. Due to the small time scale of the ohmic conduction process, this change is usually manifested as an increase in high-frequency impedance in the impedance spectrum.

[0119] like Figure 5 As shown in the figure, when the activation resistance increases, the height of the semicircle of the EIS image rises along the imaginary axis, and the diameter of the semicircle increases, which may mean that the battery has a flooding failure. In the case of a flooding failure, liquid water hinders the effective supply of reactants, resulting in a significant increase in mass transfer losses. Since the time scale of the mass transfer process is long, this change is usually reflected in the low-frequency band of the impedance spectrum, manifested as an increase in low-frequency impedance, which in turn leads to an expansion of the impedance spectrum radius.

[0120] like Figure 6 As shown, the voltage and current signals under three states, normal, membrane dry and flooded, were obtained respectively, and 40 sets of data were collected under each state. After the voltage and current signals were processed by the digital controller, the ohmic resistance, polarization internal resistance and impedance information of each frequency point were obtained respectively. The collected impedance information and the corresponding characteristic data, a total of 20 types of characteristic data, were used as the original data for health state identification. At the same time, a label matrix was established based on 120 sets of data, and labels 1, 2 and 3 were respectively represented as normal, membrane dry and flooded states. The LSSVM method used the Gaussian radial basis kernel function to calculate the Euclidean distance between the input data points, and through high-dimensional space mapping, the data can be classified through a hyperplane segmentation to improve the fault diagnosis ability. Then, 90 sets of training samples and 30 sets of test samples were randomly generated, which were used for the training and testing of the LSSVM model, respectively, to complete the diagnosis of fuel cell faults. In this way, the performance of the fuel cell under different fault states can be effectively identified, providing reliable data support for fault diagnosis.

[0121] Embodiment 2

[0122] This example provides an EIS online detection system and method for a high-power fuel cell system, proposes a modified Fouquet equivalent circuit model, and adopts a two-way parallel Boost converter topology to monitor and calculate the internal state information of the fuel cell stack during operation, and injects a multi-frequency sinusoidal periodic signal optimized by a non-dominated sorting genetic algorithm based on an improved Tent chaotic map into the auxiliary DC / DC converter, and then uses EIS characteristic data to characterize some faults, providing a new way for active fault diagnosis and early warning.

[0123] To achieve the above purpose, this example adopts the following technical solutions: Provide an EIS online detection system and method for a high-power fuel cell system, including a main circuit and a control system. Among them, the main circuit adopts a two-way parallel Boost converter topology structure, and performs small signal excitation injection through the auxiliary DC / DC converter-2. The control system includes a sampling system, a PWM drive circuit, a communication module and an embedded controller; establish a battery equivalent circuit model; when the high-power fuel cell system is running, inject a multi-frequency sinusoidal periodic signal optimized by a non-dominated sorting genetic algorithm based on the improved Tent chaotic map into the output current of the auxiliary DC / DC converter-2; sample the single cell or stack voltage through the fuel cell voltage patrol CVM, and measure the current data in combination with components such as the Hall current sensor, calculate the voltage drop corresponding to the internal resistance according to the open circuit voltage and output voltage of the fuel cell, and convert the time domain signal into a frequency domain signal through a fast Fourier transform. According to the amplitude and phase fitting at a specific frequency point, the battery impedance information is obtained; the characteristic parameter data is imported into the least squares support vector machine LSSVM for training and testing, and in-situ detection and fault diagnosis are performed online.

[0124] The main circuit includes a fuel cell, a main DC / DC converter-1, an auxiliary DC / DC converter-2 and a DC load connected in sequence; the control system includes a sampling system, a PWM drive circuit, a communication module and an embedded controller; the sampling system includes a CVM patroller, a voltage sensor, and a high-precision current Hall sensor; the input end of the PWM drive circuit is connected to the DC / DC controller, and the output end of the PWM drive circuit is connected to the MOS tube in the DC / DC converter; the input ends of the voltage sensor and the high-precision current Hall sensor of the sampling system are both connected to the DC / DC converter, and their output ends are connected to the A / D conversion module; the fuel cell voltage patrol CVM in the sampling system can directly interact with the communication module without going through an A / D converter. The communication module is connected to the embedded controller.

[0125] In a preferred embodiment; the main DC / DC converter-1 and the auxiliary DC / DC converter-2 are both composed of a filter capacitor, an inductor, and a MOS tube to form a basic Boost type boost circuit, which is used for fuel cell output power transmission or excitation signal superposition; the communication module adopts the CAN field bus protocol, which can transmit real-time data to an embedded controller or other host computer, information storage medium, etc., to perform data output, calculation, display and storage functions.

[0126] Step S1, by selecting the phase of each single-frequency sinusoidal signal through the NSGA-II genetic algorithm improved based on Tent chaotic mapping, a time domain signal with a lower crest factor than that under random phase can be obtained. Among them, the optimization target is the crest factor (the ratio of the signal peak value to twice the effective value). The frequency values ​​of each sinusoidal component of the sinusoidal excitation signal are 2, 4, 8, 16, 32, 64, 128, 256, and 1024, respectively, the amplitude value of each sinusoidal component is 4A, and the phase value defaults to 0rad;

[0127] Step S2, establish a modified Fouquet equivalent circuit model that describes the Nyquist characteristics of the fuel cell. The internal resistance of the fuel cell is simplified to the ohmic internal resistance R ohm , activation internal resistance R ct , concentration resistance R w , double layer distributed capacitance C dl and concentration capacitance C w Circuits composed of series and parallel connections;

[0128] Step S3, superimposing the optimized multi-frequency sinusoidal excitation signal on the output current of the PEMFC system in stable operation through a PWM drive circuit;

[0129] Step S4, after the excitation signal is injected, the sampling system synchronously collects and obtains a discrete sequence of response voltage and excitation current, and then transforms them into a frequency domain voltage complex sequence and a frequency domain current complex sequence through fast Fourier transform; the impedance corresponding to the excitation frequency is calculated through the impedance spectrum, that is, the sinusoidal components of each frequency of the excitation current signal and the response voltage of the internal resistance under the sinusoidal excitation of each frequency, and the amplitude and phase information of the impedance corresponding to the frequency point are obtained, and the Nyquist curve is fitted; the single cell voltage signal collected by the CVM directly enters the communication module and is uploaded to the host computer;

[0130] Step S5, the calculated impedance information is used as the original data set for fault diagnosis training and testing, which includes 20 types of feature data and 1 label column; the data set is input into the LSSVM vector machine to complete the fault diagnosis.

Claims

1. An EIS online detection system for a high-power fuel cell system, characterized in that: It includes a main circuit and a control system. The main circuit adopts a two-way parallel Boost converter topology and injects small signal excitation through an auxiliary DC / DC converter. The control system includes a sampling system, a PWM drive circuit, a communication module and an embedded controller. A battery equivalent circuit model is established. The main circuit includes a fuel cell, a DC / DC converter and a DC load connected in sequence; the control system includes a sampling system, a PWM drive circuit, a communication module and an embedded controller; the sampling system includes a CVM patroller, a voltage sensor, and a high-precision current Hall sensor; wherein the CVM patroller integrated in the sampling system is used to process and calculate the collected fuel cell voltage in real time, and calculate the data of the single cell voltage consistency state, average voltage, and total voltage; the voltage sensor measures the total output voltage of the fuel cell; the current sensor measures the output current of the fuel cell; the communication module is connected to the host computer; the main circuit adopts a two-way parallel Boost converter topology structure, one way is for output power transmission to meet the load demand, and the other way is for the multi-frequency sinusoidal periodic excitation signal injection, and its DC bias is low to avoid the detection difficulty caused by being submerged by the high DC bias of the high-power fuel cell, thereby improving the detection accuracy of small signal excitation.

2. A detection method for an EIS online detection system for a high-power fuel cell system, characterized in that: When the high-power fuel cell system is running, a multi-frequency sinusoidal periodic signal optimized by a non-dominated sorting genetic algorithm based on an improved Tent chaotic map is injected into the output current of the auxiliary DC / DC converter; the cell or stack voltage is sampled by the fuel cell voltage inspection CVM, and the current data is measured in combination with components such as the Hall current sensor, and the voltage drop corresponding to the internal resistance is calculated according to the open circuit voltage and output voltage of the fuel cell, and the time domain signal is converted into a frequency domain signal through a fast Fourier transform; the battery impedance information in the modified Fouquet equivalent circuit model is obtained according to the amplitude and phase fitting at a specific frequency point; The characteristic parameter data is imported into the least squares support vector machine LSSVM for training and testing, and in-situ detection and fault diagnosis are performed online.

3. The detection method of the EIS online detection system for a high-power fuel cell system according to claim 2, characterized in that: The multi-frequency sinusoidal periodic excitation signal in the diagnostic system is optimized by a non-dominated sorting genetic algorithm based on an improved Tent chaotic map. The phases of sinusoidal signals of different frequencies are optimized and the crest factor of the combined multi-frequency sinusoidal signal is reduced to avoid the problem of reduced EIS measurement accuracy caused by excessive superposition of excitation signal amplitudes.

4. The detection method of the EIS online detection system for a high-power fuel cell system according to claim 3, characterized in that: The high-power fuel cell is modeled by an equivalent circuit, and the ohmic internal resistance R ohm , activation internal resistance R ct 、Warburg diffusion resistance Z w and constant phase Z CPE The Fouquet model consisting of four components is improved by using a series Foster circuit to replace the Warburg diffusion resistor Z w To further accurately simulate the low-frequency and high-frequency characteristics of high-power fuel cells, a double-layer distributed capacitor C dl Replace CPE components to reduce the difficulty of fitting, forming an ohmic internal resistance R ohm , activation internal resistance R ct , concentration resistance R w , double layer distributed capacitance C dl and concentration capacitance C w Modified Fouquet equivalent circuit model composed of series and parallel connections; where the ohmic internal resistance R ohm , activation internal resistance R ct Indicates the degree of membrane dryness failure, concentration resistance R w Indicates the degree of flooding fault, double-layer distributed capacitance C dl Represents the electrical characteristics generated by the bipolar plates and electrolyte of the battery stack.

5. The detection method of the EIS online detection system for a high-power fuel cell system according to claim 4, characterized in that: The online detection method is used to identify the three states of normal, membrane dry and flooded fuel cell stacks; The data used for LSSVM model training and model verification comes from experimental samples with clear conclusions and contains state labels; the training set data and test set data are randomly divided into 3:1 ratio; The least squares support vector machine (LSSVM) uses the battery ohmic resistance, activation resistance, and impedance at 2, 4, 8, 16, 32, 64, 128, 256, and 1024 frequency points in three states as feature vectors, and the label values ​​are normal, dry membrane, and flooded, corresponding to the numbers 1, 2, and 3 respectively.

6. The detection method of the EIS online detection system for a high-power fuel cell system according to claim 4, characterized in that: In the detection method, the sampling system collects voltage and current information, and converts the analog signal into a digital signal in real time through an A / D converter. One signal enters the communication module and is input into the embedded controller for pre-data processing and calculation, and the other signal is input into the DC / DC controller for PID calculation and voltage or current stabilization control through the PWM drive circuit. The communication module uses the CAN field bus protocol to transmit real-time data to the host computer and information storage medium for display and storage respectively.

7. The detection method of the EIS online detection system for a high-power fuel cell system according to claim 6, characterized in that: The fuel cell voltage monitoring CVM interacts directly with the communication module without going through an A / D converter.

8. The detection method of the EIS online detection system for a high-power fuel cell system according to claim 4, characterized in that: The method comprises the following steps: Step S1, assigning values ​​to the initial population by using the NSGA-II genetic algorithm improved based on Tent chaotic mapping to improve the convergence ability of the algorithm; using the improved algorithm to optimize the phase of the sinusoidal excitation signal at different frequencies; wherein the optimization target is the crest factor, that is, the ratio of the signal peak value to twice the effective value; The frequency values ​​of each sinusoidal component of the sinusoidal excitation signal are 2, 4, 8, 16, 32, 64, 128, 256, and 1024 respectively, the amplitude value of each sinusoidal component is 4A, and the phase value defaults to 0rad; Step S2, establishing a modified Fouquet equivalent circuit model that describes the Nyquist characteristics of the fuel cell; wherein the internal resistance of the fuel cell is simplified to be composed of the ohmic internal resistance R ohm , activation internal resistance R ct , concentration resistance R w , double layer distributed capacitance C dl and concentration capacitance C w Circuits composed of series and parallel connections; Step S3, superimposing the optimized multi-frequency sinusoidal excitation signal on the output current of the auxiliary DC / DC converter of the fuel cell system in stable operation through the PWM drive circuit; Step S4, after the excitation signal is injected, the sampling system synchronously collects and obtains a discrete sequence of response voltage and excitation current, and then transforms them into a frequency domain voltage complex sequence and a frequency domain current complex sequence through fast Fourier transform; the impedance corresponding to the excitation frequency is calculated through the impedance spectrum, that is, the sinusoidal components of each frequency of the excitation current signal and the response voltage of the internal resistance under the sinusoidal excitation of each frequency, and the amplitude and phase information of the impedance corresponding to the frequency point are obtained, and the Nyquist curve is fitted; the single cell voltage signal collected by the CVM directly enters the communication module and is uploaded to the host computer; Step S5, the calculated impedance information is used as the original data set for fault diagnosis training and testing, which includes 20 types of feature data and 1 label column; the data set is input into the LSSVM vector machine to complete the fault diagnosis.

9. The detection method of the EIS online detection system for a high-power fuel cell system according to claim 8, characterized in that: Step S1 specifically includes the following steps: The function of the multi-frequency sinusoidal excitation signal is: Where: f(t) is the frequency of the excitation model at time t. The initialization population of the NSGA-II genetic algorithm is composed of Tent chaotic mapping. The optimization objective of the NSGA-II genetic algorithm is defined as the crest factor, which represents the uniformity of the signal amplitude distribution in the entire time domain, and its calculation formula is: Where: I ac,max is the maximum value of the multi-frequency sinusoidal excitation signal; I ac,min is the minimum value of the multi-frequency sinusoidal excitation signal; I eff is the effective value of the multi-frequency sinusoidal excitation signal. eff The expression is: Furthermore, step S2 specifically includes the following steps: The modified Fouquet equivalent circuit model impedance can be expressed as: The real and imaginary parts of the impedance are calculated according to the above formula, and the Nyquist curve is drawn in the complex plane according to the relationship between the real and imaginary parts of the modified Fouquet equivalent circuit model at various frequencies to represent the electrochemical impedance spectrum of the fuel cell.

10. The detection method of the EIS online detection system for a high-power fuel cell system according to claim 9, characterized in that: Step S4 specifically includes the following steps: The sampling system includes the main components of CVM patroller, Hall current sensor, voltage sensor and precision resistor, analog-to-digital converter ADC and other components required for sampling; wherein, the CVM patroller can monitor the inside of the battery stack, measure the current operating status of a single battery and transmit real-time data to the main controller through the field bus protocol. The detailed voltage data measured by the CVM patroller can be used for fault diagnosis. At the same time, when problems such as voltage imbalance and unit failure occur, a fault alarm can be directly issued to trigger and record the voltage data at the time of the fault; step S5 specifically includes the following steps: The LSSVM simplifies the standard SVM by converting the constraints into a square loss function to simplify the optimization process; According to the SRM criterion, the objective function and constraints of LSSVM are: In the formula, ξ i is the prediction error; c is the regularization parameter or penalty parameter used to control the prediction error ξ i The degree of sample penalty. LSSVM handles errors by minimizing the square loss, avoiding the introduction of slack variables; By constructing the Lagrange transformation polynomial, the following transformation is performed: For the above formula w, b, e k , α k By taking partial derivatives and setting the derivatives to zero, we can get a system of linear equations. The calculation formula of the LSSVM algorithm is: In the formula, K(x,x i ) is the kernel function.

Images