Method and system for detecting fastening force of fuel cell stack in real time

Abstract

The invention relates to a fuel cell stack fastening force real-time detection method and system, and the method comprises the steps: collecting the contact resistance data of a plurality of detection points on a fuel cell stack in real time, the plurality of detection points being arranged along the axial and circumferential directions of the fuel cell stack; and based on a preset fastening force mapping model about the contact resistance values and the contact resistance value corresponding to each detection point, obtaining fastening force values of the plurality of detection points on the fuel cell stack. The multiple detection points are arranged in the axial direction and the circumferential direction of the fuel cell stack, so that the system can cover a key monitoring layer of the fuel cell stack, and comprehensive and multi-point monitoring of the fastening force in the fuel cell stack is achieved; through a preset contact resistance-fastening force mapping model, contact resistance data collected in real time can be converted into a fastening force value, and direct measurement of the actual stress state in the fuel cell stack is realized; the system can continuously work in the whole life cycle of fuel cell stack manufacturing, testing, loading and use and the like, and continuous and real-time detection is achieved.

Description

Technical Field

[0001] This application relates to the field of fuel cells, specifically to a method and system for real-time detection of fuel cell stack fastening force. Background Technology

[0002] As a highly efficient and clean energy conversion device, the performance and safety of fuel cells are highly dependent on the sealing and internal structural integrity of the fuel cell stack. Among these, the press-fitting of the fuel cell stack is a critical step, and appropriate clamping force is essential to ensuring the sealing performance and overall performance of the fuel cell stack.

[0003] Currently, the pressing force of fuel cell stacks is mainly determined indirectly by press pressure sensors or stack height, but there are obvious drawbacks: the pressure sensors measure the applied force and cannot reflect the actual force inside the stack; changes in stack height are affected by manufacturing tolerances and assembly errors, and are not sensitive to small changes in fastening force.

[0004] More importantly, as the usage time increases, the stack components will undergo permanent deformation, resulting in a decrease in fastening force. Existing technology cannot monitor this degradation process in real time, and cannot detect and resolve fastening force problems in a timely manner during assembly, long-term use, and harsh environments, thus affecting the performance, efficiency, and safety of fuel cells. Summary of the Invention

[0005] This application provides a method and system for real-time detection of fastening force of fuel cell stacks, which can solve the problem that the existing technology lacks an effective method for continuous and real-time detection of fastening force throughout the entire life cycle, and cannot timely detect and solve fastening force problems that occur during assembly, after long-term use, and in harsh environments.

[0006] In a first aspect, embodiments of this application provide a method for real-time detection of the fastening force of a fuel cell stack, comprising: Real-time acquisition of contact resistance data from multiple detection points on the fuel cell stack, with multiple detection points set along the axial and circumferential directions of the fuel cell stack; Based on a preset fastening force mapping model for contact resistance values ​​and the contact resistance value corresponding to each detection point, the fastening force values ​​of multiple detection points on the fuel cell stack are obtained.

[0007] In conjunction with the first aspect, in one implementation, a resistance monitoring system constructed on the fuel cell stack is used to collect contact resistance data at multiple detection points on the fuel cell stack in real time.

[0008] Before acquiring contact resistance data at multiple detection points on the fuel cell stack in real time, the method further includes the step of constructing a resistance monitoring system on the fuel cell stack: Detection layers are allocated based on the total number of layers in the fuel cell stack; Along the circumference of the fuel cell stack, multiple detection points are arranged at intervals on the bipolar plate of the fuel cell stack corresponding to each detection layer, and probes are set at the detection points. Connect the top of the probe to the detection circuit for use with an external acquisition module to complete the construction of the resistance monitoring system.

[0009] In conjunction with the first aspect, in one embodiment, multiple detection points are arranged at intervals on the bipolar plates of the fuel cell stack corresponding to each detection layer, along the circumference of the fuel cell stack, specifically including: Grooves of a set size are machined at intervals along the circumference of the fuel cell stack bipolar plate corresponding to each detection layer. Deposit a conductive substrate within the groove; An anti-oxidation layer is electroplated onto a conductive substrate; Embed the probe into the groove; An insulating layer is coated on the anti-oxidation layer, and the insulating layer covers the probe. A hole is made in the insulating layer to expose the top of the probe, thus completing the arrangement of the detection points.

[0010] In conjunction with the first aspect, in one implementation, a resistance monitoring system constructed on the fuel cell stack is used to collect contact resistance data at multiple detection points on the fuel cell stack in real time, specifically including: During the rest period of the fuel cell stack, a constant current is simultaneously injected into multiple detection points of the resistance monitoring system to obtain the voltage of each detection point; Based on Ohm's law, constant current, and voltage at each detection point, contact resistance data at multiple detection points on the fuel cell stack are obtained.

[0011] In conjunction with the first aspect, in one implementation, contact resistance data at multiple detection points on the fuel cell stack are acquired based on Ohm's law, constant current, and the voltage at each detection point, specifically including: Based on Ohm's law, constant current, and voltage at each detection point, the raw contact resistance data of multiple detection points on the fuel cell stack are obtained. Temperature compensation is performed on the original contact resistance data according to the preset temperature-resistance compensation model to obtain contact resistance data decoupled from temperature.

[0012] In conjunction with the first aspect, in one embodiment, before obtaining the fastening force values ​​of multiple detection points on the fuel cell stack based on a preset fastening force mapping model for contact resistance values ​​and the contact resistance value corresponding to each detection point, the method further includes: Establish a pre-defined fastening force mapping model for contact resistance values.

[0013] In conjunction with the first aspect, in one implementation, a preset fastening force mapping model regarding the contact resistance value is established, specifically including: At a fixed temperature, a nonlinear relationship was established between contact resistance data and raw fastening force data. Under constant pressure, the relationship between temperature and contact resistance data was established. Based on the influence of temperature on contact resistance data, a temperature compensation coefficient is obtained, and a temperature-resistance compensation model is established. Based on the temperature-resistance compensation model, temperature compensation is performed on the contact resistance data in the nonlinear relationship between contact resistance data and fastening force to obtain the contact resistance value after temperature compensation. The contact resistance value after temperature compensation and the original fastening force data are nonlinearly fitted to obtain a preset fastening force mapping model with respect to the contact resistance value. The preset fastening force mapping model with respect to the contact resistance value satisfies the following: the determination coefficient of the preset fastening force mapping model with respect to the contact resistance value is greater than a first set value, and the residual standard deviation is less than a second set value.

[0014] In conjunction with the first aspect, in one embodiment, after obtaining the fastening force values ​​of multiple detection points on the fuel cell stack based on a preset fastening force mapping model for contact resistance values ​​and the contact resistance value corresponding to each detection point, the method further includes: By analyzing fuel cell stack operation data in real time using a preset fastening force attenuation prediction model, the percentage of resistance deviation caused by changes in fuel cell stack fastening force in the next time period can be predicted. The corresponding early warning mechanism is triggered by comparing the predicted percentage of resistance deviation with the early warning threshold.

[0015] In conjunction with the first aspect, in one implementation, a corresponding early warning mechanism is triggered based on a comparison between the predicted resistance deviation percentage and an early warning threshold, specifically including: triggering a first early warning mechanism when the resistance deviation percentage is in a first early warning threshold range, triggering a second early warning mechanism when the resistance deviation percentage is in a second early warning threshold range, and triggering a third early warning mechanism when the resistance deviation percentage is in a third early warning threshold range. The warning threshold is dynamically adjusted based on the current ambient temperature and the fuel cell stack load current.

[0016] Secondly, embodiments of this application provide a real-time detection system for the fastening force of a fuel cell stack, comprising: a resistance monitoring system and a fastening force calculation module. The resistance monitoring system is used to collect contact resistance data of multiple detection points on the fuel cell stack in real time, with the multiple detection points arranged along the axial and circumferential directions of the fuel cell stack. The fastening force calculation module is used to obtain the fastening force values ​​of multiple detection points on the fuel cell stack based on a preset fastening force mapping model for contact resistance values ​​and the contact resistance value corresponding to each detection point.

[0017] The beneficial effects of the technical solutions provided in this application include: This application provides a method and system for real-time detection of fastening force in fuel cell stacks. By setting multiple detection points along the axial and circumferential directions of the fuel cell stack, the system can cover the key monitoring layers of the fuel cell stack, achieving comprehensive and multi-point monitoring of the internal fastening force. Through a preset contact resistance-fastening force mapping model, the system can convert real-time collected contact resistance data into fastening force values, enabling direct measurement of the actual stress state inside the fuel cell stack. This allows the system to operate continuously throughout the entire lifecycle of the fuel cell stack, including manufacturing, testing, and vehicle installation, achieving continuous and real-time detection, and promptly identifying fastening force problems during assembly, fastening force attenuation after prolonged use, and changes in fastening force under harsh environments. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the real-time detection method for fuel cell stack fastening force in this application; Figure 2 This is a schematic diagram of the bipolar plate of the real-time detection system for the fastening force of the fuel cell stack in this application; Figure 3 This is a schematic diagram of the detection circuit of the real-time detection system for the fastening force of the fuel cell stack in this application; Figure 4 This is a schematic diagram of the RF calibration curve of the real-time detection system for the fastening force of the fuel cell stack in this application; Figure 5 This is a schematic diagram of the bipolar plate of the real-time detection system for the fastening force of the fuel cell stack in this application; Figure 6 This is a schematic diagram showing the distribution of detection points on the bipolar plates of the fuel cell stack in the real-time detection system for the fastening force of the fuel cell stack according to this application.

[0019] In the diagram: 1. Bipolar plate; 2. Embedded detection electrode; 3. Battery stack end plate; 4. Flexible circuit; 5. Layered detection unit; 6. Groove; 7. Insulating layer; 8. Probe; 9. Connection point. Detailed Implementation

[0020] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present application.

[0021] This application provides a method and system for real-time detection of fastening force of fuel cell stacks, which can solve the problem that the existing technology lacks an effective method for continuous and real-time detection of fastening force throughout the entire life cycle, and cannot timely detect and solve fastening force problems that occur during assembly, after long-term use, and in harsh environments.

[0022] To make the objectives, technical solutions, and advantages of this application clearer, the embodiments of this application will be described in further detail below with reference to the accompanying drawings.

[0023] In a first aspect, embodiments of this application provide a method for real-time detection of the fastening force of a fuel cell stack, comprising: 101: Real-time acquisition of contact resistance data at multiple detection points on the fuel cell stack, with multiple detection points set along the axial and circumferential directions of the fuel cell stack; 102: Based on the preset fastening force mapping model of contact resistance value and the contact resistance value corresponding to each detection point, obtain the fastening force value of multiple detection points on the fuel cell stack.

[0024] In this application, by setting multiple detection points along the axial and circumferential directions of the fuel cell stack, the system can cover the key monitoring layer of the fuel cell stack, realizing comprehensive and multi-point monitoring of the internal fastening force of the fuel cell stack. Through the preset contact resistance-fastening force mapping model, the real-time collected contact resistance data can be converted into fastening force values, realizing direct measurement of the actual stress state inside the fuel cell stack. This enables the system to work continuously throughout the entire life cycle of the fuel cell stack, including manufacturing, testing, and vehicle use, achieving continuous and real-time detection, and timely detection of fastening force problems during assembly, fastening force decay after long-term use, and changes in fastening force under harsh environments.

[0025] It should be noted that this application uses a resistance monitoring system built on the fuel cell stack to collect contact resistance data at multiple detection points on the fuel cell stack in real time.

[0026] Therefore, before acquiring contact resistance data at multiple detection points on the fuel cell stack in real time, the method further includes the step of constructing a resistance monitoring system on the fuel cell stack: First, the detection layers are allocated based on the total number of layers in the fuel cell stack: the detection layers are dynamically allocated axially according to the total number of layers in the stack. In this embodiment, one set of detection points is set for every 3 layers of bipolar plates 1, covering the key monitoring layers.

[0027] Then, along the circumference of the fuel cell stack, multiple detection points are arranged at intervals on the bipolar plate 1 of the fuel cell stack corresponding to each detection layer. These detection points are located in the non-reaction zone at the outer edge of the flow channel area of ​​the bipolar plate 1, avoiding interference with the normal power generation of the fuel cell stack. Probes 8 are installed at each detection point. Specifically, the bipolar plate 1 where the probes 8 are to be placed is processed. Along the circumference of the fuel cell stack, grooves 6 of a predetermined size (e.g., diameter 0.5-1mm) are processed at intervals along the edge of the bipolar plate 1 corresponding to each detection layer. In this embodiment, laser engraving technology is used to process grooves 6 with a diameter of 0.8mm and a depth of 0.3mm on the edge of the bipolar plate 1. The non-contact processing of laser engraving avoids mechanical stress damage, ensures the smoothness of the groove 6 edges, and improves the stability of the probe 8 embedding. Then, a conductive substrate is deposited in the groove 6. In this embodiment, a 10μm thick nickel layer is deposited as the conductive substrate. The nickel layer provides excellent conductivity and adhesion, reducing contact resistance drift. Next, an anti-oxidation layer is electroplated on the conductive substrate. In this embodiment, a 5μm thick gold layer is electroplated. The gold layer effectively inhibits oxidation on the surface of the bipolar plate 1 and extends the service life of the probe 8. Then, the probe 8 is embedded in the groove 6. In this embodiment, the probe 8 is a gold-plated copper probe with a thickness of 50-100μm. An insulating layer 7 is coated on the anti-oxidation layer, and the insulating layer 7 covers the probe 8. In this embodiment, a 20μm thick polyimide insulating layer 7 is spin-coated. Finally, a hole is drilled in the insulating layer 7 using laser drilling technology to expose the top of the probe 8. It should be noted that in this embodiment, only a contact point with a top diameter of 0.2mm is exposed, thereby completing the arrangement of the detection point.

[0028] In this embodiment, three sets of probes 8 are arranged radially at 120° intervals along the circumference of the fuel cell stack to ensure monitoring at any angle, thus forming redundant monitoring. The purpose of setting three sets is to make the three sets of probes 8 backups for each other; if one set fails (e.g., due to contact point oxidation), the other sets can still work.

[0029] Finally, the probe 8 is connected to the detection circuit via connection point 9 on its top for connection with the external acquisition module, completing the construction of the resistance monitoring system. The detection circuit uses a PI-based flexible circuit 4 (FPC, 100μm linewidth, 150μm spacing) to connect each detection point. Leads are routed through pre-drilled holes in the fuel cell stack end plate 3 to the external acquisition module, ensuring stable and accurate signal transmission. Furthermore, anti-interference design is implemented: a magnetic isolation relay is installed between the measuring electrode and the main power generation circuit to prevent the fuel cell operating current from interfering with the detection signal. The fuel cell stack end plate 3 refers to the outermost supporting structural component of the fuel cell stack, typically made of high-strength metal (such as stainless steel) or composite materials, located at both ends of the stack (positive and negative end plates).

[0030] Based on the above embodiments, in this embodiment, after the preprocessing of the bipolar plate 1 is completed, the contact resistance data of multiple detection points on the fuel cell stack are collected in real time through a resistance monitoring system built on the fuel cell stack, specifically including: First, during the rest period of the fuel cell stack, a constant current (i.e., a DC microcurrent) is simultaneously injected into multiple detection points of the resistance monitoring system to obtain the voltage at each detection point. The fuel cell stack rest period refers to the state where the stack is not powered on, effectively avoiding electromagnetic interference from the operating current and ensuring noise-free voltage measurement. Then, based on Ohm's law, the constant current, and the voltage at each detection point, the contact resistance data of multiple detection points on the fuel cell stack are obtained.

[0031] In this embodiment, the ADuCM360 low-power ADC is used. Its built-in high-precision analog-to-digital converter and low-noise circuit design effectively suppress electromagnetic interference in the fuel cell stack operating environment, ensuring the purity of the measurement signal. During the fuel cell stack rest period (two-second interval, avoiding the fuel cell stack working cycle to eliminate working current noise and avoid measurement data drift), a constant current of 0.5-2mA is injected. The current value is calibrated by the thermal characteristics of the fuel cell stack material to prevent overheating and oxidation of the contact points, ensuring long-term measurement stability. The measurement voltage resolution reaches 1μV, achieving a resistance detection resolution of 0.1mΩ, capturing minute fluctuations in fastening force; the corresponding resistance detection accuracy is ±0.1mΩ. This level of accuracy can reliably identify assembly deviations and fastening force attenuation during long-term use, significantly improving the input data quality of the mapping model.

[0032] By measuring and calculating the contact resistance, data support is provided for the next step of establishing a fastening force mapping model of the contact resistance. High-precision data directly reduces the model calibration error, improves the matching degree between the fastening force estimation and the actual working conditions, and supports real-time monitoring and fault early warning functions.

[0033] It should also be noted that during constant current injection, a constant current of 0.5–2mA is injected into the resistor being measured (such as the test point of a fuel cell stack) through two current leads (HI, LO), which is achieved by the programmable excitation current source inside the ADuCM360. The current value is dynamically adjusted through a real-time feedback loop. For example, the system first performs a fast initial resistance estimation, and then calculates the optimal current value based on the estimated value to ensure that the voltage signal amplitude is stable within the stable range of the ADC, avoiding signal overload or signal-to-noise ratio degradation due to resistance fluctuations. This current value is dynamically adjusted according to the range of the resistor being measured to ensure that the voltage signal amplitude is within the ADC range.

[0034] During voltage measurement, two additional voltage leads (SENSE+, SENSE-) are directly connected to the two ends of the resistor under test, and high-precision voltage sampling is performed using the dual-channel 24-bit Σ-Δ ADC of the ADuCM360. In this embodiment, since the input impedance of the voltage measurement terminal is as high as MΩ, the current flowing through the voltage leads is negligible, so the lead resistance does not introduce errors. Specifically, a four-wire Kelvin connection is adopted, and the input impedance of the voltage measurement terminal is designed to be above 100MΩ, making the current flowing through the voltage leads less than 0.1nA. Therefore, the interference of lead resistance and contact resistance on the voltage reading is negligible, and measurement consistency is maintained even under stack vibration or long lead scenarios. Through the synergistic optimization of four-wire high-precision measurement and anti-interference design, the noise and error problems in micro-resistance detection of fuel cell stacks are solved.

[0035] The resistance value is calculated using Ohm's law (R=V / I) and combined with the on-chip digital filters of the ADUCM360 (such as the sinc3 filter and the sinc4 filter; the sinc4 filter provides 40dB noise suppression in the 50Hz-1kHz power frequency noise band, which is suitable for the slow stress change characteristics of fuel cell stacks, while the sinc3 is used for fast transient signal processing. The system automatically selects the filtering mode according to the real-time signal frequency) to eliminate noise and achieve a detection accuracy of ±0.1mΩ.

[0036] Based on this, to improve the accuracy of contact resistance data, a temperature compensation strategy is incorporated. Therefore, based on Ohm's law, constant current, and the voltage at each detection point, contact resistance data at multiple detection points on the fuel cell stack are obtained, specifically including: First, based on Ohm's law, constant current, and voltage at each detection point, the raw contact resistance data of multiple detection points on the fuel cell stack are obtained. Then, temperature compensation is performed on the original contact resistance data according to the preset temperature-resistance compensation model to obtain contact resistance data decoupled from temperature.

[0037] The pre-defined temperature-resistance compensation model is a key component in achieving high-precision fastening force detection in this application. Its establishment process includes steps such as temperature-resistance relationship calibration, temperature compensation coefficient calculation, compensation formula construction, verification, and optimization. First, under fixed pressure conditions (e.g., 100kN), the fuel cell stack is placed in a high-low temperature shock test chamber for temperature cycling testing: the temperature is increased from 25℃ to 80℃ at a rate of 5℃ / min and held for 30 minutes, then decreased to -30℃ at the same rate and held for 30 minutes. The resistance values ​​at each temperature point (every 10℃ interval) are recorded. Using these data, the temperature coefficient β = (1 / R) is calculated. 25 )×(ΔR / ΔT), where R 25 The reference resistance is at 25℃, and ΔR / ΔT is the linear rate of change of resistance with temperature.

[0038] Then, based on the calibration data, the temperature compensation formula is established: Rc_corrected = Rc_measured / [1 + β(T - T0) + γ·RH + δ·P], where Rc_corrected is the corrected contact resistance data, Rc_measured is the measured original contact resistance data, T0 is the reference temperature (usually 25℃), RH is the relative humidity, P is the fuel cell stack operating pressure, and β, γ, and δ are coefficients obtained through environmental chamber calibration (γ = 2.3 × 10⁻⁶). -4 / %, δ=1.7×10 -5 / kPa). This formula eliminates the interference of temperature, humidity and fuel cell operating pressure on contact resistance through three key terms: β(T-T0) compensates for temperature changes, γ·RH compensates for humidity effects, and δ·P compensates for fuel cell operating pressure effects.

[0039] To verify the compensation effect, a cross-validation method was used: the same pressure (e.g., 120kN) was applied at -30℃, 25℃, and 80℃, and the resistance deviation before and after compensation was compared. The goal was to make the resistance deviation after compensation ≤ ±1% (which can reach ±8% without compensation).

[0040] In addition, long-term stability testing is required after the model is established. It must operate continuously for 1000 hours at 80℃ and 80% humidity, with data collected every 24 hours to assess model parameter drift, ensuring an annual drift of <2%. In practical applications, when three consecutive measurements deviate from the model's predicted value by more than 5%, the system automatically initiates an online calibration procedure. This involves applying a reference pressure (e.g., 100kN) and measuring the resistance, dynamically adjusting the model coefficients based on the measured values ​​to ensure the long-term accuracy of the compensation model. Through this series of calibration, verification, and optimization processes, a temperature-resistance compensation model capable of adapting to the entire temperature range of -30℃ to 80℃ and effectively eliminating environmental interference is finally constructed, providing a reliable guarantee for the accurate monitoring of fuel cell stack fastening forces.

[0041] In fuel cell stack fastening force testing, to address the contact resistance fluctuation problem (resistance drift caused by surface oxidation or contamination), gold plating of probe 8 is used to enhance oxidation resistance, combined with a mechanism of periodically injecting microcurrents to clean the contacts, effectively suppressing contact surface deterioration. To address the effects of material creep (material deformation altering the contact state after long-term pressure), a creep correction term ΔRcreep=γ·ln(t) is embedded in the temperature-resistance compensation model, where the creep coefficient γ is precisely calibrated through accelerated aging experiments. Simultaneously, to address local resistance deviations caused by uneven temperature distribution within the fuel cell stack, an innovative array of temperature sensors is deployed at multiple locations within the fuel cell stack, using a weighted average algorithm to dynamically compensate for temperature gradient interference. Through the implementation of this systematic solution, the model achieves a detection accuracy of ±3.5% (significantly improved compared to the traditional pressure sensor method), reliable compensation capability covering the entire temperature range from -30℃ to 80℃, and maintains parameter drift below 1.5% after 1000 hours of continuous aging testing, significantly improving the accuracy, environmental adaptability, and long-term stability of fuel cell stack fastening force monitoring.

[0042] Based on the above embodiments, in this embodiment, before obtaining the fastening force values ​​of multiple detection points on the fuel cell stack based on a preset fastening force mapping model for contact resistance values ​​and the contact resistance value corresponding to each detection point, the method further includes: Establish a pre-defined fastening force mapping model for contact resistance values.

[0043] The process of establishing a preset fastening force mapping model for contact resistance values ​​specifically includes: first, establishing a nonlinear relationship between contact resistance data and original fastening force data at a fixed temperature; then, establishing the influence of temperature on contact resistance data under a fixed pressure; next, obtaining a temperature compensation coefficient based on the influence of temperature on contact resistance data and establishing a temperature-resistance compensation model; then, based on the temperature-resistance compensation model, performing temperature compensation on the contact resistance data in the nonlinear relationship between contact resistance data and fastening force to obtain the temperature-compensated contact resistance value; finally, performing nonlinear fitting on the temperature-compensated contact resistance value and the original fastening force data to obtain a preset fastening force mapping model for contact resistance values, wherein the preset fastening force mapping model for contact resistance values ​​satisfies the following: the determination coefficient of the preset fastening force mapping model for contact resistance values ​​is greater than a first set value, and the residual standard deviation is less than a second set value.

[0044] The establishment of the dynamic contact resistance-fastening force mapping model should follow laboratory calibration specifications, and a highly reliable model should be constructed through equipment configuration and systematic data acquisition.

[0045] The experimental design and equipment configuration employ a four-fold collaborative system: the pressure application device uses a universal testing press with a range of 0–300 kN and an accuracy of ±0.5%, ensuring precise control of the clamping force; the resistance measurement system is equipped with a four-wire micro-ohmmeter (such as Keysight 34420A) with a resolution of 0.1 μΩ, achieving micro-ohm level measurement of contact resistance through an embedded gold-plated copper probe 8 (0.5 mm in diameter, with a polytetrafluoroethylene insulating layer covering the surface and only 0.2 mm of contact exposed); the temperature control device uses a high and low temperature shock test chamber (temperature range -40℃~150℃, temperature change rate ≥5℃ / min) to accurately simulate the full temperature range working conditions; and the data acquisition system synchronously records pressure, resistance, and temperature data, with a sampling frequency of up to 10 Hz to capture transient changes. In the sample preparation stage, a micro-groove 6 (0.8 mm in diameter and 0.3 mm in depth) is processed on the graphite bipolar plate 1, gold-plated probe 8 is embedded and covered with an insulating layer 7. During the assembly of the stack, the detection probe 8 is embedded in groups of 3 layers to form a small test stack of 10–20 layers, ensuring that the contact points are evenly distributed and repeatable.

[0046] The data acquisition process is executed in three steps: First, the fuel cell stack is pre-pressurized to 10kN at room temperature (25±2℃) to eliminate assembly gaps and avoid interference from poor initial contact; second, the pressure is gradually increased from 50kN to 200kN in increments of 10kN. After each pressure level is stabilized for 30 seconds, the original resistance value (Rmeasured) is recorded. Simultaneously, the depressurization process (200kN→50kN) is performed to verify the data hysteresis error (must be <1%).

[0047]

[0048] Finally, the verification was further deepened through a pressure calibration process—using a hydraulic / servo motor testing machine to apply axial pressure of 5–200 kN (5 kN step, 40 data points in total), with a holding time of ≥10 seconds to eliminate creep effects. During the stack rest period, the ADuCM360 measurement circuit was triggered to inject a constant current of 1 mA, and the voltage signal was collected to calculate Rc. Each pressure group was repeated 3 times and the average value was taken. This process, through high-precision equipment, standardized sample preparation, and a closed-loop verification mechanism, ensured that the calibration data met the stringent requirements of model construction, laying a reliable foundation for subsequent temperature compensation and nonlinear fitting.

[0049] The process of establishing a pre-defined fastening force mapping model for contact resistance begins with applying multiple levels of fastening force (range 50kN–150kN) under fixed temperature conditions (e.g., 25℃), collecting raw contact resistance data at corresponding detection points, and establishing a nonlinear relationship model between contact resistance and fastening force (e.g., using cubic polynomial fitting). Simultaneously, under fixed pressure conditions (e.g., 100kN), temperature gradient tests are conducted (-30℃ to 80℃, in 10℃ intervals), measuring the change in contact resistance with temperature, fitting a temperature-resistance influence curve, and extracting the temperature compensation coefficient β (β=(1 / R)). 25 )×(ΔR / ΔT), R 25 (The reference resistance is 25℃). Based on this, a temperature-resistance compensation model Rc_corrected=Rc_measured / [1+β(T-T0)] (T0 is the reference temperature 25℃) is constructed to eliminate the interference of temperature fluctuations on the contact resistance. Subsequently, the original contact resistance data is temperature compensated using this model to obtain the temperature-decoupled contact resistance value.

[0050] When generating the preset fastening force mapping model by performing a nonlinear fit between the compensated contact resistance value (Rc_corrected) and the original fastening force data, "Rc=KF" is used. -a+C is the nonlinear relationship between the coefficients (k, α, C) determined through experimental calibration. The coefficients "k, α, C" are determined by pressure calibration on the testing machine (5-200kN, 5kN step), and the fitted curve R²>0.99. The physical meaning of coefficient k (proportional factor) is: characterizing the initial conductivity of the material contact interface, related to the material surface roughness, conductive layer characteristics, and contact area; its influence is: the larger the k value, the higher the contact resistance Rc under the same clamping force, indicating lower material contact efficiency (e.g., thicker surface oxide layer or sparse contact points); experimental correlation is: in the low pressure range (when F is small), k dominates the trend of Rc change, reflecting the contact instability of the material under weak pressure. The physical meaning of the exponent α (attenuation factor) is: describing the rate of attenuation of contact resistance with increasing pressure, related to the material's elastic modulus, yield strength, and interface... The deformation characteristics are relevant; the influence is as follows: the larger the α value, the faster the Rc decreases with increasing F, indicating that the material is more likely to form a low-resistance path under pressure (such as soft materials or highly elastic sealing layers); the typical range is: α is usually 0.3~0.7. If α is close to 1, it indicates that the contact resistance is extremely sensitive to pressure, suitable for high-precision assembly scenarios. The physical meaning of the constant C (base resistance) is: it represents the limiting contact resistance when the pressure approaches infinity, including the material volume resistance, the base resistance of the measurement system, and the resistance of unavoidable interface defects; the influence is as follows: the smaller the C value, the better the material contact performance (such as uniform coating of bipolar plate 1 and smooth surface of the membrane electrode); if C is significantly greater than zero, assembly process or material defects need to be investigated. The experimental correlation is: in the high pressure range (F>150kN), Rc approaches C, reflecting the saturation state of the material contact area.

[0051] The fitting process uses the Levenberg-Marquardt algorithm to minimize the sum of squared residuals. The objective function is constrained to have a coefficient of determination R² > 0.99 (measured R² = 0.993), and a residual standard deviation < 0.1 mΩ. (Example of fitting results: coefficient k = 2.5 × 10⁻⁶) -3 Ω·kN a (Characterizes the initial conductivity of the material; the larger the k value, the lower the contact efficiency), α=0.45 (attenuation factor, reflects the resistance attenuation rate under pressure; α=0.45 indicates that the material has moderate sensitivity under pressure), C=0.02mΩ (ground resistance, represents the limiting resistance under high pressure saturation).

[0052] To address long-term environmental disturbances, a creep correction term ΔRcreep = γ˙ln(t) was added to the model, where γ was calibrated through aging experiments. Multiple temperature sensors were placed at different locations on the fuel cell stack, and a weighted average was used to compensate for temperature gradient disturbances. The creep correction term γ was calibrated through a 1000-hour aging test (80℃ / 80%RH) (γ = 1.2 × 10⁻⁶). -4This addresses resistance drift caused by material creep (e.g., a 15% reduction in contact area after prolonged pressure); the weighted average algorithm of the temperature sensor array (5 points) dynamically compensates for the internal temperature gradient of the fuel cell (e.g., when the temperature difference between the center and the edge reaches 10℃, the resistance deviation decreases from ±8% to ±1%).

[0053] Finally, its statistical performance was verified: ensuring a coefficient of determination R² > 0.99 (first set value), a residual standard deviation < 0.1 mΩ (second set value), and ensuring reliability through cross-validation (reserving 10% of data points) and a dynamic calibration mechanism (automatically adjusting coefficients when the deviation between three consecutive measurements and model predictions exceeds 5%). Implementation results show that the model achieves a detection accuracy of ±3.5% (compared to the pressure sensor method) across the entire temperature range of -30℃ to 80℃, with parameter drift < 1.5% after 1000 hours of aging. Furthermore, through an online calibration process—applying a reference pressure (e.g., 100 kN) and measuring resistance—the model coefficients are dynamically adjusted to maintain long-term stability. This process transforms the theoretical model into an engineering closed-loop system, providing high-precision and robust support for the online real-time monitoring of fuel cell stack fastening forces.

[0054] Based on the above embodiments, in this embodiment, after obtaining the fastening force values ​​of multiple detection points on the fuel cell stack based on a preset fastening force mapping model for contact resistance values ​​and the contact resistance value corresponding to each detection point, the method further includes: First, a pre-set fastening force attenuation prediction model is used to analyze the fuel cell stack operating data in real time to predict the percentage of resistance deviation caused by changes in the fuel cell stack fastening force in the next time period. Then, based on the comparison between the predicted resistance deviation percentage and the warning threshold, a corresponding warning mechanism is triggered. Specifically, when the resistance deviation percentage is within the first warning threshold range, the first warning mechanism is triggered; when the resistance deviation percentage is within the second warning threshold range, the second warning mechanism is triggered; and when the resistance deviation percentage is within the third warning threshold range, the third warning mechanism is triggered. The warning thresholds are dynamically adjusted based on the current ambient temperature and the fuel cell stack load current.

[0055] Specifically, after acquiring the fastening force values ​​at multiple detection points on the fuel cell stack, this embodiment introduces an intelligent early warning mechanism. The intelligent early warning mechanism realizes a complete chain from data acquisition to closed-loop intervention: First, a time alignment layer integrates asynchronous data from multiple sources such as resistance, temperature, and load in real time, eliminating data timing differences and significantly improving operational adaptability; subsequently, a fastening force attenuation prediction model based on an LSTM neural network automatically analyzes input parameters every 5 minutes—including the real-time resistance change rate (ΔR / R0=(R_current-R0) / R0, where R0 is the reference resistance value reflecting the current contact state), temperature gradient (… It captures the rate of change of ambient / operating temperature, historical decay slope (Slope=linear_fit(ΔR / R0,10min), quantifying the recent decay trend), load current (I_load, from the BMS system) and operating time (t_operation, cumulative hours) to accurately predict the trend of changes in the tightness of the fuel cell stack.

[0056] In the dynamic threshold adjustment stage, the system calculates the threshold correction coefficient η=1+0.02(T-25)+0.05(I / I_rated) in real time based on the ambient temperature T and the load current I. For example, when the ambient temperature T=40℃ and the load current I=0.8I_rated (I_rated is the rated current), the threshold is automatically increased by 12%, which effectively solves the problem of false triggering of traditional fixed thresholds under high temperature and high load conditions.

[0057] The predicted results are compared with the dynamic threshold to trigger a graded early warning mechanism: when the resistance deviation percentage is in the mild abnormal range of 5%-10%, the system automatically records the problem but does not interrupt operation; when it is in the medium risk range of 10%-15%, a real-time reminder is sent to maintenance personnel via SMS or APP; when it is in the severe fault range of >15%, the emergency device is immediately activated - mechanical compensation is performed by pushing the disc spring through the stepper motor (stroke range 0.1-0.3mm), while reducing the power generation to ensure safety.

[0058] This process, through a closed loop of "predictive early warning + dynamic adjustment," upgrades abnormal fastening force from a passive response to an active intervention, ensuring the safe and stable operation of the fuel cell throughout its entire life cycle and significantly improving system reliability and maintenance efficiency.

[0059] This application addresses the shortcomings of existing fuel cell fastening force detection methods by introducing a series of innovations and improvements. Firstly, it abandons the method of indirectly determining fastening force by simply relying on compressor pressure sensors or fuel cell height. Instead, it cleverly utilizes the strong correlation between the internal contact resistance of the fuel cell stack and the fastening force. By monitoring the resistance changes at key contact points in real time, it indirectly reflects the fastening force status; fundamentally solving the problem that existing methods cannot directly and accurately detect the internal fastening force of the fuel cell stack.

[0060] Secondly, to achieve this detection method, an embedded detection electrode 2 was designed: a micro-conductive probe was embedded in the edge or non-reactive area of ​​the bipolar plate 1 of the fuel cell stack to form a distributed resistance monitoring network. This design not only avoids interference with the normal power generation of the fuel cell stack, but also enables comprehensive monitoring of the fastening force at different locations inside the fuel cell stack, greatly improving the accuracy and reliability of the detection.

[0061] Furthermore, a low-power dynamic measurement circuit was adopted, employing a four-wire micro-current measurement method. A constant micro-current was injected into the detection electrode during non-power generation periods, and the voltage drop was measured to calculate the contact resistance. Simultaneously, an integrated temperature sensor compensated for the effect of temperature on material resistance, further improving the accuracy and stability of the detection.

[0062] Finally, an intelligent data processing and early warning system was established. By establishing a resistance-fastening force calibration curve and using advanced algorithms to automatically identify abnormal trends, it can promptly trigger early warning or automatic adjustment mechanisms, effectively ensuring the safety and reliability of the fuel cell.

[0063] This solution not only enables real-time detection of the fastening force of fuel cell stacks, but also comprehensively covers the detection of fastening force parameters throughout the entire lifecycle of the stack, including manufacturing, testing, and vehicle use. This allows for timely and accurate monitoring of fastening force changes throughout the entire usage process, enabling early detection of potential problems. Furthermore, by transforming the inherent characteristics of fuel cell stacks through IoT technology, it directly and continuously monitors the fastening force, balancing cost and reliability. This provides support for large-scale production lines and in-service stack health management, significantly reducing detection costs by eliminating the need for expensive sensors. The implementation plan relies on bipolar plate 1 pre-processing and flexible circuit 4 integration, making the detection device easy to install and maintain. Multi-point detection accurately locates local loosening problems, improving fault diagnosis efficiency. At the same time, the adaptive compensation mechanism and disc spring assembly work together to achieve dynamic force adjustment, further enhancing the stability and reliability of fuel cells under different operating conditions.

[0064] Secondly, embodiments of this application provide a real-time detection system for the fastening force of a fuel cell stack, comprising: a resistance monitoring system and a fastening force calculation module. The resistance monitoring system is used to collect contact resistance data of multiple detection points on the fuel cell stack in real time, with the multiple detection points arranged along the axial and circumferential directions of the fuel cell stack. The fastening force calculation module is used to obtain the fastening force values ​​of multiple detection points on the fuel cell stack based on a preset fastening force mapping model for contact resistance values ​​and the contact resistance value corresponding to each detection point.

[0065] In this application, by setting multiple detection points along the axial and circumferential directions of the fuel cell stack, the system can cover the key monitoring layer of the fuel cell stack, realizing comprehensive and multi-point monitoring of the internal fastening force of the fuel cell stack. Through the preset contact resistance-fastening force mapping model, the real-time collected contact resistance data can be converted into fastening force values, realizing direct measurement of the actual stress state inside the fuel cell stack. This enables the system to work continuously throughout the entire life cycle of the fuel cell stack, including manufacturing, testing, and vehicle use, achieving continuous and real-time detection, and timely detection of fastening force problems during assembly, fastening force decay after long-term use, and changes in fastening force under harsh environments.

[0066] The functions of each module in the above-mentioned real-time detection system for fuel cell stack fastening force correspond to the steps in the above-mentioned real-time detection method for fuel cell stack fastening force, and their functions and implementation processes will not be described in detail here.

[0067] Thirdly, embodiments of this application provide a real-time detection device for the fastening force of a fuel cell stack. The real-time detection device for the fastening force of a fuel cell stack can be a device with data processing capabilities, such as a personal computer (PC), a laptop computer, or a server.

[0068] In this embodiment of the application, the real-time detection device for the fastening force of the fuel cell stack may include a processor, a memory, a communication interface, and a communication bus.

[0069] The communication bus can be of any type and is used to interconnect the processor, memory, and communication interface.

[0070] The communication interface includes input / output (I / O) interfaces, physical interfaces, and logical interfaces for interconnecting internal components of the real-time fuel cell stack clamping force monitoring device, as well as interfaces for interconnecting the device with other devices (such as other computing devices or user equipment). Physical interfaces can be Ethernet interfaces, fiber optic interfaces, ATM interfaces, etc.; user equipment can be displays, keyboards, etc.

[0071] Memory can be various types of storage media, such as random access memory (RAM), read-only memory (ROM), non-volatile RAM (NVRAM), flash memory, optical storage, hard disk, programmable ROM (PROM), erasable PROM (EPROM), electrically erasable PROM (EEPROM), etc.

[0072] The processor can be a general-purpose processor, which can call the real-time detection program for fuel cell stack fastening force stored in the memory and execute the real-time detection method for fuel cell stack fastening force provided in the embodiments of this application. For example, the general-purpose processor can be a central processing unit (CPU). The method executed when the real-time detection program for fuel cell stack fastening force is called can refer to the various embodiments of the real-time detection method for fuel cell stack fastening force of this application, and will not be repeated here.

[0073] Fourthly, embodiments of this application also provide a computer-readable storage medium.

[0074] The present application stores a real-time detection program for fuel cell stack fastening force on a computer-readable storage medium, wherein when the real-time detection program for fuel cell stack fastening force is executed by a processor, it implements the steps of the real-time detection method for fuel cell stack fastening force as described above.

[0075] The method implemented when the fuel cell stack fastening force real-time detection program is executed can be referred to in various embodiments of the fuel cell stack fastening force real-time detection method of this application, and will not be repeated here.

[0076] It should be noted that the sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0077] The terms "comprising" and "having," and any variations thereof, in the specification, claims, and accompanying drawings of this application are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to such process, method, product, or apparatus. The terms "first," "second," and "third," etc., are used to distinguish different objects, etc., and do not indicate a sequence, nor do they limit "first," "second," and "third" to different types.

[0078] In the description of the embodiments of this application, terms such as "exemplary," "for example," or "for instance" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as "exemplary," "for example," or "for instance" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of terms such as "exemplary," "for example," or "for instance" is intended to present the relevant concepts in a concrete manner.

[0079] In the description of the embodiments of this application, unless otherwise stated, " / " means "or". For example, A / B can mean A or B. The "and / or" in the text is merely a description of the relationship between related objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist simultaneously, and B exists alone. In addition, in the description of the embodiments of this application, "multiple" means two or more.

[0080] In some processes described in the embodiments of this application, multiple operations or steps are included in a specific order. However, it should be understood that these operations or steps may not be executed in the order they appear in the embodiments of this application, or they may be executed in parallel. The sequence number of the operation is only used to distinguish different operations, and the sequence number itself does not represent any execution order. In addition, these processes may include more or fewer operations, and these operations or steps may be executed sequentially or in parallel, and these operations or steps may be combined.

[0081] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes several instructions to cause a terminal device to execute the methods described in the various embodiments of this application.

[0082] The above are merely preferred embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. A method for real-time detection of the fastening force of a fuel cell stack, characterized in that, It includes: Real-time acquisition of contact resistance data from multiple detection points on the fuel cell stack, with multiple detection points set along the axial and circumferential directions of the fuel cell stack; Based on a preset fastening force mapping model for contact resistance values ​​and the contact resistance value corresponding to each detection point, the fastening force values ​​of multiple detection points on the fuel cell stack are obtained.

2. The method for real-time detection of fuel cell stack fastening force as described in claim 1, characterized in that, A resistance monitoring system built on the fuel cell stack is used to collect contact resistance data at multiple detection points on the fuel cell stack in real time. Before acquiring contact resistance data at multiple detection points on the fuel cell stack in real time, the method further includes the step of constructing a resistance monitoring system on the fuel cell stack: Detection layers are allocated based on the total number of layers in the fuel cell stack; Along the circumference of the fuel cell stack, multiple detection points are arranged at intervals on the bipolar plate (1) of the fuel cell stack corresponding to each detection layer, and probes (8) are provided at the detection points. Connect the top of the probe (8) to the detection circuit for use with the external acquisition module to complete the construction of the resistance monitoring system.

3. The method for real-time detection of fuel cell stack fastening force as described in claim 2, characterized in that, Along the circumference of the fuel cell stack, multiple detection points are arranged at intervals on the bipolar plate (1) of the fuel cell stack corresponding to each detection layer, specifically including: Along the circumference of the fuel cell stack, grooves (6) of a set size are machined at intervals along the edge of the bipolar plate (1) of the fuel cell stack corresponding to each detection layer. A conductive substrate is deposited within the groove (6); An anti-oxidation layer is electroplated onto a conductive substrate; Embed the probe (8) into the groove (6); An insulating layer (7) is coated on the anti-oxidation layer, and the insulating layer (7) covers the probe (8); Make a hole in the insulating layer (7) to expose the top of the probe (8) and complete the arrangement of the detection points.

4. The method for real-time detection of fuel cell stack fastening force as described in claim 2, characterized in that, A resistance monitoring system built on the fuel cell stack is used to collect contact resistance data at multiple detection points on the fuel cell stack in real time, specifically including: During the rest period of the fuel cell stack, a constant current is simultaneously injected into multiple detection points of the resistance monitoring system to obtain the voltage of each detection point; Based on Ohm's law, constant current, and voltage at each detection point, contact resistance data at multiple detection points on the fuel cell stack are obtained.

5. The method for real-time detection of fuel cell stack fastening force as described in claim 4, characterized in that, Based on Ohm's law, constant current, and voltage at each detection point, contact resistance data at multiple detection points on the fuel cell stack are obtained, specifically including: Based on Ohm's law, constant current, and voltage at each detection point, the raw contact resistance data of multiple detection points on the fuel cell stack are obtained. Temperature compensation is performed on the original contact resistance data according to the preset temperature-resistance compensation model to obtain contact resistance data decoupled from temperature.

6. The method for real-time detection of fuel cell stack fastening force as described in claim 1, characterized in that, Before obtaining the fastening force values ​​at multiple detection points on the fuel cell stack based on a preset fastening force mapping model for contact resistance values ​​and the contact resistance value corresponding to each detection point, the method further includes: Establish a pre-defined fastening force mapping model for contact resistance values.

7. The method for real-time detection of fuel cell stack fastening force as described in claim 6, characterized in that, Establish a pre-defined fastening force mapping model for contact resistance values, specifically including: At a fixed temperature, a nonlinear relationship was established between contact resistance data and raw fastening force data. Under constant pressure, the relationship between temperature and contact resistance data was established. Based on the influence of temperature on contact resistance data, a temperature compensation coefficient is obtained, and a temperature-resistance compensation model is established. Based on the temperature-resistance compensation model, temperature compensation is performed on the contact resistance data in the nonlinear relationship between contact resistance data and fastening force to obtain the contact resistance value after temperature compensation. The contact resistance value after temperature compensation and the original fastening force data are nonlinearly fitted to obtain a preset fastening force mapping model with respect to the contact resistance value. The preset fastening force mapping model with respect to the contact resistance value satisfies the following: the determination coefficient of the preset fastening force mapping model with respect to the contact resistance value is greater than a first set value, and the residual standard deviation is less than a second set value.

8. The method for real-time detection of fuel cell stack fastening force as described in claim 1, characterized in that, Based on a preset fastening force mapping model for contact resistance values ​​and the contact resistance value corresponding to each detection point, after obtaining the fastening force values ​​of multiple detection points on the fuel cell stack, the method further includes: By analyzing fuel cell stack operation data in real time using a preset fastening force attenuation prediction model, the percentage of resistance deviation caused by changes in fuel cell stack fastening force in the next time period can be predicted. The corresponding early warning mechanism is triggered by comparing the predicted percentage of resistance deviation with the early warning threshold.

9. The method for real-time detection of fuel cell stack fastening force as described in claim 8, characterized in that: Based on the comparison between the predicted percentage of resistance deviation and the warning threshold, a corresponding warning mechanism is triggered, specifically including: When the percentage of resistance deviation is within the first warning threshold range, the first warning mechanism is triggered; when the percentage of resistance deviation is within the second warning threshold range, the second warning mechanism is triggered; when the percentage of resistance deviation is within the third warning threshold range, the third warning mechanism is triggered. The warning threshold is dynamically adjusted based on the current ambient temperature and the fuel cell stack load current.

10. A real-time detection system for the fastening force of a fuel cell stack, characterized in that, It includes: A resistance monitoring system is used to collect contact resistance data at multiple detection points on the fuel cell stack in real time. These multiple detection points are set along the axial and circumferential directions of the fuel cell stack. The fastening force calculation module is used to obtain the fastening force values ​​of multiple detection points on the fuel cell stack based on a preset fastening force mapping model of contact resistance value and the contact resistance value corresponding to each detection point.

Images