Method for determining reusable simulation pyrofuse board card based on multi-parameter joint

Abstract

The invention provides a method for determining a reusable simulation Pyrofuse board card based on multi-parameter joint, belongs to the technical field of automobile and energy storage part test verification, and aims to solve the technical problems of high test cost and poor flexibility caused by disposable use of traditional Pyrofuse. According to the method, loop current signals are collected in real time through a high-speed ADC sampling module of a board card, a CPU adopts multi-parameter logic of energy accumulation integral judgment or overcurrent duration and threshold current combined judgment, whether the fusing condition is met or not is accurately judged, an output module is controlled to simulate the fusing state, and repeated use can be achieved through recovery of a reset module after testing. The board card integrates a variable resistor, a communication module, a power supply module and other modules, and key parameters can be configured by software. High-fidelity simulation of Pyrofuse electrical characteristics is achieved, the Pyrofuse electrical characteristic simulation device can be repeatedly used, the research and development cost is reduced, anti-interference performance and adaptability are improved through multi-parameter judgment, and the Pyrofuse electrical characteristic simulation device is suitable for test verification scenes of various automobiles and energy storage parts.

Description

Technical Field

[0001] This invention relates to the field of automotive and energy storage component testing and verification technology, and in particular to a method for determining reusable simulated pyrofuse boards based on multi-parameter joint determination. Background Technology

[0002] The core function of a pyrofuse is as a safety switch for circuits. When a vehicle or energy storage system experiences a collision, short circuit, overcurrent, or other safety malfunction, the explosive force of the detonator quickly disconnects the circuit, reducing the probability of danger. Currently, pyrofuses are widely used in electric vehicles and energy storage applications as a safety mechanism to prevent power outages in case of high-voltage leakage and emergencies.

[0003] In the testing and verification of automotive and energy storage components, multiple detonations of Pyrofuse are required for parameter verification. However, traditional Pyrofuses are disposable devices; once detonated, they cannot be restored and must be replaced to continue testing, significantly increasing R&D and testing costs. Furthermore, the fixed parameters of traditional Pyrofuses, such as current threshold and trip time, cannot be flexibly adjusted for different testing scenarios, resulting in poor adaptability. In addition, traditional Pyrofuses are triggered by a single voltage or current threshold, making detonation conditions uncontrollable and explosion curves difficult to predict, increasing the risk of accidental detonation or missed detonation, thus affecting the accuracy of testing and verification.

[0004] Existing technologies lack solutions that can accurately simulate the electrical characteristics of Pyrofuse, are reusable, and have flexible parameter configuration and precise trigger determination.

[0005] Therefore, this invention proposes a method for determining reusable simulated pyrofuse boards based on multi-parameter joint determination. Summary of the Invention

[0006] This invention provides a method for determining reusable simulated pyrofuse boards based on multiple parameters, in order to solve the aforementioned technical problems.

[0007] This invention provides a method for determining reusable simulated pyrofuse boards based on multi-parameter joint determination, comprising: Step 1: Configure test parameters based on the communication module. The test parameters include the current threshold ITH, time threshold Ts, energy threshold Eth, and the target resistance value of the variable resistor module. Step 2: The high-speed ADC sampling module measures the voltage across the low-temperature drift SHUNT resistor connected in series in the controlled loop, and calculates the loop current signal in real time based on Ohm's law. This signal is then processed by a 24-bit... After the chip performs digital processing, the data is transmitted to the CPU via the SPI interface; Step 3: The CPU processes the digitized current signal in real time and uses multi-parameter joint judgment logic to determine whether the fuse condition is met. The multi-parameter joint judgment logic is: parallel execution of energy accumulation integral judgment and overcurrent duration-threshold current joint judgment. If any judgment condition is met, the fuse is triggered. Step 4: If the fuse condition is met, the CPU drives the relay of the output module to disconnect, making the output module present a high impedance state, simulating the Pyrofuse fuse action; Step 5: After the test is completed, a reset operation is performed through the reset module. The relay of the output module closes and restores the low-resistance state, enabling the board to be reused.

[0008] Preferably, the energy accumulation integral determination includes: Using the trapezoidal integral method according to the formula E = Calculate the cumulative energy E, where, The equivalent resistance of the melt is denoted as I(t); the real-time current is denoted as t; and t is a time variable, ranging from t1 to t2. When both conditions are met and When the circuit breaker condition is met, it is determined that the circuit breaker condition has been met. The joint determination of overcurrent duration and threshold current includes: Real-time statistics of overcurrent duration T, when simultaneously satisfying and When the condition is met, it is determined that the circuit breaker condition is met.

[0009] Preferably, the pyrofuse board includes a high-speed ADC sampling module, a variable resistor module, a CPU, an output module, a reset module, a communication module, and a power supply module. The high-speed ADC sampling module is connected to the CPU via an SPI interface, the variable resistor module, the output module, and the reset module are connected to the CPU via GPIO interfaces, the communication module is connected to the CPU via a CAN / EtherCAT interface, and the power supply module provides stable 5V and 12V power to each module. The output module uses an electromagnetic relay connected in series in the controlled circuit to switch between high and low impedance states.

[0010] Preferred, by 24 bits Before the chip undergoes digital processing, it includes: A test circuit including the chip is constructed, and different test signals are sent to the test circuit. The output signals of the chip under different sampling frequencies of different test signals are captured respectively. A signal-time matrix based on each test signal is constructed. In the signal matrix, each element in the first row is a reference pair consisting of the test signal value and the standard test time, and each element in the remaining rows is an actual pair consisting of the output signal value and the signal test time. The maximum temperature value of the chip under different sampling frequencies of the same test signal is measured to construct a temperature vector and obtain a temperature matrix; The signal-time matrix is ​​analyzed to obtain a first eigenvector, and the first eigenvector is compared with the first row of the signal matrix to obtain a first difference vector. At the same time, the matrix after removing the first row vector of the signal-time matrix is ​​used to obtain a second eigenvector, and the second eigenvector is compared with the first row of the signal matrix to obtain a second difference vector. Based on the first difference vector and the second difference vector, the first reference delay and the first reference noise of the corresponding test signal are determined. At the same time, the third difference vector of the second feature vector and each remaining row of the corresponding signal matrix (excluding the first row) are obtained and compared with the first difference vector to obtain the corresponding second actual delay and the second actual noise. The delay vector is formed by the first reference delay under the same test signal and all the second actual delays. At the same time, the noise vector is formed by the first reference noise under the same test signal and all the second actual noises. The delay vector noise vector Temperature vector A fitting analysis was performed to determine the effect function of temperature on delay and noise, where, This represents the delay value at the i-th sampling frequency; This represents the noise value at the i-th sampling frequency. Let be the maximum temperature value at the i-th sampling frequency; n is the number of sampling frequencies. A comprehensive analysis of the influence functions of all test signals is performed to determine the temperature influence range. The qualified elements of the temperature matrix are then calibrated according to the temperature influence range to obtain the recommendation coefficient for each sampling frequency. The sampling frequency corresponding to the highest coefficient among all recommendation coefficients is selected as the subsequent operating frequency of the chip.

[0011] Preferably, determining the effect function of temperature on delay and noise includes: Construct sampling frequency vector ,in, Let i be the i-th sampling frequency; Constructing a vector-level frequency-temperature coupled fitting function ,in, This is the frequency-temperature coupling weight matrix, with elements... This represents the weight of the effect of temperature on the delay corresponding to the i-th sampling frequency at the j-th sampling frequency; For Hadamah accumulation; This is the delay reference vector, corresponding to the basic delay value of each sampling frequency under rated temperature conditions; It is the adaptive residual correction vector, and ,and This is the correction factor for delayed residuals; It is an n-order identity matrix; This is the delay residual vector obtained from the previous fitting; Constructing nonlinear dimensional coupling functions ,in, This is the frequency-temperature coupling weight matrix corresponding to the noise; Frequency sensitivity index; , These are respectively the sampling frequency vectors The power-extended vector, the quadratic extension vector of the temperature vector; This is the noise reference vector, corresponding to the basic noise value of each sampling frequency under rated temperature conditions; This is the noise adaptive residual correction vector, and ,and This is the noise residual correction factor; It is a symbolic function; It is an L1 norm; This is the noise residual vector obtained from the previous fitting.

[0012] Preferably, a comprehensive analysis of the influence functions of all test signals is performed to determine the temperature influence range, including: Based on the maximum allowable delay threshold and maximum noise threshold when the board is working, the effective temperature range corresponding to each group of test signals is selected. The effective temperature range is the temperature range in which the temperature-delay influence function calculation result of the corresponding group of test signals does not exceed the maximum delay threshold and the temperature-noise influence function calculation result does not exceed the maximum noise threshold. Based on the similarity between each group of test signals and the actual working signals of the board, the weight coefficients of each group of test signals are quantified, and the sum of all weight coefficients is 1. Calculate the weighted average of the lower limit values ​​of the effective temperature range corresponding to all test signals, and subtract the confidence correction value determined by the variance of the fitting residuals of each group of test signals to obtain the lower limit value of the comprehensive temperature influence range. Calculate the weighted average of the upper limits of the effective temperature ranges corresponding to all test signals, and add the confidence correction value to obtain the upper limit of the comprehensive temperature influence range; The interval formed by the lower and upper limits of the comprehensive temperature influence range is determined as the final temperature influence range.

[0013] Preferably, before the relay of the CPU drive output module is closed, the following steps are included: An auxiliary circuit for relay sensitivity testing is constructed. The auxiliary circuit includes an indicator light circuit, a multi-dimensional signal sampling circuit, a drive closed-loop adjustment unit, and a freewheeling diode performance monitoring unit. The auxiliary circuit is connected in series with the relay and the CPU output drive circuit. The multi-dimensional signal sampling circuit is used to collect the actual current signal of the relay coil, the contact action feedback signal, and the coil magnetic field strength signal. The freewheeling diode performance monitoring unit is used to collect the reverse current signal of the freewheeling diode connected in parallel across the coil. Based on the multi-dimensional signal sampling circuit and the freewheeling diode performance monitoring unit, the signal data corresponding to each pre-signal is collected. Combined with the status signal of the indicator light circuit, the critical drive parameter set corresponding to the relay contact trigger action is recorded. The critical drive parameter set includes the critical drive current, the critical signal slope, the critical magnetic field strength, and the critical remaining energy value calculated based on the reverse current of the freewheeling diode. Based on the set of critical driving parameters and the rated operating parameters of the relay, the sensitivity adaptation coefficient is calculated; Determine whether the sensitivity adaptation coefficient is within the preset qualified range. If not, adaptively adjust the gain of the CPU output drive signal and the output level of the relay drive chip. If the sensitivity matching coefficient is still not within the acceptable range after adjustment, replace the relay and the freewheeling diode with matching parameters, and retest. If so, in the scenario of simulating the repeated use of the pyrofuse board, after performing a preset number of closing-opening actions, the decay rate of the reverse current of the freewheeling diode and the stability of the coil magnetic field strength are monitored simultaneously. If the sensitivity adaptation coefficient is still within the acceptable range after performing a preset number of closing-opening actions, and the stability is greater than the preset value, it is determined that the CPU is allowed to drive the relay to close. Otherwise, replace the relay.

[0014] Preferably, the sensitivity adaptation coefficient is the normalized deviation weighted sum of the critical drive current, critical signal slope, critical residual energy value and corresponding rated parameters, multiplied by the correlation coefficient between coil magnetic field strength and contact action, and the weight of the deviation weighted sum is jointly determined by the reusability requirements of the pyrofuse board and the reverse current decay characteristics of the freewheeling diode.

[0015] Compared with the prior art, the beneficial effects of this application are: The single amplitude triggering concept has been upgraded to a multi-dimensional collaborative judgment concept of amplitude, time, and energy, which has strong anti-interference ability: it can effectively resist the interference of instantaneous surges and greatly reduce the false alarm rate; precise action: it ensures that the protection device operates reliably when a real continuous fault occurs; and high controllability: by setting different current thresholds, time thresholds, or energy thresholds, it can achieve precise matching with different protection scenarios and realize the simulation and reproduction of the real explosion scene.

[0016] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description and the accompanying drawings.

[0017] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0018] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1 This is a flowchart of a method for determining a reusable simulated pyrofuse board based on multiple parameters in an embodiment of the present invention; Figure 2 This is a graph related to energy accumulation integral in an embodiment of the present invention; Figure 3 This is a graph showing the relationship between overcurrent duration and threshold current in an embodiment of the present invention; Figure 4 This is a circuit connection diagram of the pyrofuse board in an embodiment of the present invention; Detailed Implementation

[0019] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0020] This invention provides a method for determining reusable simulated pyrofuse boards based on multiple parameters, such as... Figure 1 As shown, it includes: Step 1: Configure test parameters based on the communication module. The test parameters include the current threshold ITH, time threshold Ts, energy threshold Eth, and the target resistance value of the variable resistor module. Step 2: The high-speed ADC sampling module measures the voltage across the low-temperature drift SHUNT resistor connected in series in the controlled loop, and calculates the loop current signal in real time based on Ohm's law. This signal is then processed by a 24-bit... After the chip performs digital processing, the data is transmitted to the CPU via the SPI interface; Step 3: The CPU processes the digitized current signal in real time and uses multi-parameter joint judgment logic to determine whether the fuse condition is met. The multi-parameter joint judgment logic is: parallel execution of energy accumulation integral judgment and overcurrent duration-threshold current joint judgment. If any judgment condition is met, the fuse is triggered. Step 4: If the fuse condition is met, the CPU drives the relay of the output module to disconnect, making the output module present a high impedance state, simulating the Pyrofuse fuse action; Step 5: After the test is completed, a reset operation is performed through the reset module. The relay of the output module closes and restores the low-resistance state, enabling the board to be reused.

[0021] Preferably, the energy accumulation integral determination includes: Using the trapezoidal integral method according to the formula Calculate the cumulative energy E, where, The equivalent resistance of the melt is denoted as I(t); the real-time current is denoted as t; and t is a time variable, ranging from t1 to t2. When both conditions are met and When the circuit breaker condition is met, it is determined that the circuit breaker condition has been met. The joint determination of overcurrent duration and threshold current includes: Real-time statistics of overcurrent duration T, when simultaneously satisfying and When the condition is met, it is determined that the circuit breaker condition is met.

[0022] Preferably, the pyrofuse board includes a high-speed ADC sampling module, a variable resistor module, a CPU, an output module, a reset module, a communication module, and a power supply module. The high-speed ADC sampling module is connected to the CPU via an SPI interface; the variable resistor module, output module, and reset module are connected to the CPU via GPIO interfaces; the communication module is connected to the CPU via a CAN / EtherCAT interface; and the power supply module provides stable 5V and 12V power to each module. The output module uses an electromagnetic relay connected in series in the controlled circuit to switch between high and low impedance states, such as... Figure 4 As shown.

[0023] This invention is a high-end electronic testing and simulation device that integrates high-speed sampling, real-time processing, precise control, and flexible simulation. It is not a physical fuse that blows once, but a highly intelligent virtual fuse system designed specifically for design verification, performance testing, and system integration scenarios. Its aim is to safely, reliably, and repeatably simulate the behavior of a real pyrofuse under various operating conditions.

[0024] The variable resistor module employs a high-performance relay matrix architecture, achieving precise resistance output within the range of 0.5Ω to 5Ω through program control. The module utilizes resistor units with high stability and high power tolerance, ensuring it can carry the nominal current at different resistance values, demonstrating excellent flexibility and reliability, and providing high-precision dynamic resistive loads for various testing and simulation applications.

[0025] In this embodiment, a complete current monitoring solution is built based on a high-speed sampling module. Its working principle is based on Ohm's law, precisely measuring the voltage across the cryogenic drift SHUNT resistor connected in series in the detonation circuit to calculate the high-precision circuit current value in real time. This analog voltage signal is digitized by a 24-bit high-precision Δ-Σ ADC, ensuring distortion-free capture of transient signals with a sampling rate of up to 80 kSPS. The digitized data seamlessly interacts with the main controller via a high-speed SPI interface, forming a high-fidelity link from signal sensing to data processing. This configuration not only provides extremely high DC accuracy over a wide dynamic range but also possesses excellent transient response capabilities, fully capturing instantaneous changes in current and complex waveforms, providing a reliable data foundation for accurately reconstructing the actual detonation process and deeply analyzing its electrical characteristics.

[0026] The controller (CPU) core is a 600MHz high-speed processor with powerful real-time data processing capabilities. It continuously processes the data stream from the high-speed sampling module and executes configurable trigger condition decisions. Once the target pulse is captured, the system immediately activates two parallel actions: First, the key current waveforms before and after the trigger point are rapidly recorded at 10-microsecond intervals using an integrated high-speed cache (200kB). Second, the relay is immediately driven to simulate the fuse-breaking action of Pyrofuse. All action parameters, such as trigger threshold, waveform amplitude, and duration, can be flexibly configured via software, achieving a high degree of adaptability and control precision.

[0027] The analog output module uses a fast-acting relay as its core actuator. By connecting the relay in series with the controlled circuit, it achieves accurate simulation of two states: when the relay is closed, the module presents a low-impedance state, simulating a fully functional fuse; when the relay is open, the module presents a high-impedance state, simulating a blown fuse with high fidelity. This provides a safe, reliable, and repeatable solution for testing and verifying circuit protection functions.

[0028] The reset module provides a dual reset mechanism—external hardware and internal software—ensuring reliable system recovery under any conditions. External reset is achieved by applying a 12V high-level pulse to the Rset pin; a buzzer sounds upon successful reset. This hardware-level operation can force a system restart, even in extreme cases such as software crashes. Internal reset is triggered by software commands, providing the system with flexible and convenient remote or logical control restart capabilities.

[0029] The communication module offers a dual-mode interconnect solution, supporting both the classic CAN bus and the high-performance EtherCAT bus. The module employs a highly efficient real-time processing architecture, ensuring deterministic microsecond-level response capabilities in demanding industrial applications, meeting the highest requirements for real-time performance and reliability in scenarios such as motion control and real-time data acquisition.

[0030] This device uses a wide input range of 9-36V DC power supply, providing excellent power adaptability. Internally, it employs a high-efficiency isolated DC-DC power module, which not only effectively suppresses noise interference, ensuring a clean and stable power supply to the internal circuits, but also provides reliable electrical isolation, significantly improving the system's anti-interference capability and long-term operational reliability.

[0031] like Figure 2 As shown, in region P1 (blue): although the current I3 lasts for a relatively long time t3, its value is far lower than the threshold current I(TH), and the energy P2 generated is insufficient to cause the melt to explode.

[0032] Region P0 (gray): Although the current I1 briefly exceeded the threshold current ITH, its duration t1 was too short, and the energy accumulation P1 did not reach the critical point. It is a recoverable overshoot and should not trigger the explosion.

[0033] Region P1 (yellow): The current I2 not only significantly exceeds the threshold current ITH, but its duration t2 is long enough to make the accumulated energy P0 far exceed the safety threshold. At this time, it is determined to be an effective blasting condition.

[0034] Therefore, the complete triggering logic of this method requires two conditions to be met simultaneously: Current condition: I>ITH, which indicates a real risk of overcurrent; Energy condition: E>Eth, the accumulated thermal stress has reached the melting critical point; This energy integration method can effectively distinguish between surge current and continuous fault current, avoiding malfunctions caused by normal transient processes such as equipment startup.

[0035] like Figure 3As shown, the overcurrent events within the gray area are ignored because the duration T1 < Ts; and the current I2 is excluded because it never reaches the threshold current ITH. Only when both the amplitude and duration of the current break through their respective thresholds simultaneously does the system confirm an irrecoverable permanent fault and execute the blasting instruction.

[0036] The beneficial effects of the above technical solution are as follows: Strong anti-interference ability: It can effectively resist the interference of instantaneous surges and greatly reduce the false alarm rate; Precise action: Ensure that the protection device operates reliably when a real continuous fault occurs; High controllability: By setting different current thresholds, time thresholds or energy thresholds, it can be accurately matched with different protection scenarios to simulate and restore the real explosion scene.

[0037] The present invention provides a method for a reusable analog pyrofuse board based on multi-parameter joint determination. Before digital processing by a 24-bit chip, it includes: Build a test circuit including the chip, and send different test signals to the test circuit. Capture the output signals of the chip at different sampling frequencies for different test signals respectively, and construct a signal-time matrix based on each test signal. Among them, each element in the first row of the signal matrix is a reference pair composed of the test signal value and the standard test time, and each element in the remaining rows is an actual pair composed of the output signal value and the signal test time; Measure the maximum temperature values of the chip at different sampling frequencies for the same test signal to construct a temperature vector, and obtain a temperature matrix; Analyze the signal-time matrix to obtain a first eigenvector, and perform a first difference analysis on the first eigenvector and the first row of the signal matrix to obtain a first difference vector. At the same time, obtain a second eigenvector for the matrix after removing the first row vector of the signal-time matrix, and perform a second difference analysis on the second eigenvector and the first row of the signal matrix to obtain a second difference vector; According to the first difference vector and the second difference vector, determine the first reference delay and the first reference noise of the corresponding test signal. At the same time, obtain the third difference vector of the second eigenvector and each remaining row except the first row in the corresponding signal matrix, and compare it with the first difference vector to obtain the corresponding second actual delay and the second actual noise; Based on the first reference delay under the same test signal and all the second actual delays, form a delay vector. At the same time, based on the first reference noise under the same test signal and all the second actual noises, form a noise vector; The delay vector 、noise vector 、temperature vector A fitting analysis was performed to determine the effect function of temperature on delay and noise, where, This represents the delay value at the i-th sampling frequency; This represents the noise value at the i-th sampling frequency. Let be the maximum temperature value at the i-th sampling frequency; n is the number of sampling frequencies. A comprehensive analysis of the influence functions of all test signals is performed to determine the temperature influence range. The qualified elements of the temperature matrix are then calibrated according to the temperature influence range to obtain the recommendation coefficient for each sampling frequency. The sampling frequency corresponding to the highest coefficient among all recommendation coefficients is selected as the subsequent operating frequency of the chip.

[0038] In this embodiment, the test circuit is a dedicated circuit for testing the performance of a 24-bit Δ-Σ ADC chip. It includes a signal source, chip power supply, load, and temperature monitoring components. Its function is to simulate the chip's actual operating environment and obtain chip performance parameters under different operating conditions. For example, the test circuit consists of a function signal generator, an ADC chip power supply circuit, a 10kΩ load resistor, and a temperature sensor. The function signal generator outputs different types of test signals to the ADC chip input, and the temperature sensor is used to monitor the chip's operating temperature. Specifically, the test circuit is built on a breadboard, with test probes soldered to key nodes for easy voltage and temperature measurement; the circuit grounding is consistent with the board grounding to reduce grounding loop interference and ensure the accuracy of the test data.

[0039] Test signals are input signals used to test the performance of ADC chips. They need to cover signal types with different amplitudes, frequencies, and waveforms to comprehensively evaluate the chip's performance under different operating conditions. For example, three sets of test signals can be set: the first set is a sine wave with an amplitude of 1Vpp and a frequency of 1kHz; the second set is a square wave with an amplitude of 2Vpp and a frequency of 10kHz; and the third set is a triangle wave with an amplitude of 1.5Vpp and a frequency of 5kHz, covering common signal types and frequency ranges.

[0040] Sampling frequency refers to the rate at which an ADC chip converts an analog signal into a digital signal. Different sampling frequencies result in different delay and noise characteristics for the chip; the optimal value is selected through testing. For example, if n=5 sampling frequencies are set: =16kSPS =32kSPS =64kSPS =128kSPS =256kSPS, covering low, medium and high rate ranges. Specifically, configuration commands are sent to the ADC chip through the CPU's SPI interface to adjust the chip's internal sampling rate control register, enabling rapid switching between different sampling frequencies. After each switch, the chip works stably for 5 minutes before data acquisition.

[0041] In this embodiment, the signal-time matrix is ​​a two-dimensional matrix recording the test signal value, standard test time, ADC output signal value, and signal test time. LabVIEW software is used to collect the values ​​and corresponding times of the test and output signals in real time, arranging them row-wise to generate the matrix. The first row is the reference pair (test signal value and standard test time), and the remaining rows are the actual pairs (output signal value and signal test time), used for subsequent difference analysis and feature extraction. Taking a sine wave test signal as an example, the first row of the signal-time matrix is ​​[(1V, 0ms), (0V, 0.25ms), (-1V, 0.5ms)]. The second line is [(0.998V, 0.001ms), (-0.002V, 0.251ms), (-0.997V, 0.501ms)]; The third line is [(0.999V, 0.002ms), (-0.001V, 0.252ms), (-0.998V, 0.502ms)].

[0042] In this implementation, the temperature vector G is a one-dimensional vector recording the maximum chip temperature values ​​corresponding to different sampling frequencies under the same test signal, used to analyze the impact of temperature on chip performance. For example, under a sine wave test signal, the maximum temperature values ​​corresponding to the five sampling frequencies are as follows: =45℃ =50℃ =58℃ Given g = 65℃ and g5 = 72℃, the temperature vector G = [45, 50, 58, 65, 72], with units of ℃. The temperature sensor is attached to the surface of the ADC chip and connected to the data acquisition card via an I2C interface. After stabilizing at each sampling frequency for 5 minutes, the maximum temperature value of the chip is recorded to ensure that the temperature data reflects the chip's heating characteristics at that sampling frequency.

[0043] The temperature matrix is ​​a two-dimensional matrix composed of the temperature vectors corresponding to all test signals. Each row corresponds to a set of test signals, and each column corresponds to a sampling frequency. It is used to comprehensively analyze the relationship between temperature and sampling frequency under different test signals. For example, if the temperature vectors of the three test signals are G1=[45,50,58,65,72], G2=[46,51,59,66,73], and G3=[44,49,57,64,71], then the temperature matrix is ​​[[45,50,58,65,72],[46,51,59,66,73],[44,49,57,64,71]]. The unit of the matrix elements is ℃, and the accuracy is 0.1℃.

[0044] In this embodiment, the first eigenvector is a one-dimensional vector obtained by feature extraction from the complete signal-time matrix. It can reflect the overall characteristics of the matrix and facilitate subsequent difference analysis. Principal component analysis is used to extract features from the signal-time matrix of the sinusoidal test signal, resulting in the first eigenvector V1=[0.98,0.02,-0.97]. Its dimension is consistent with the number of columns in the matrix, and the contribution rate of the first principal component is ≥95%, which can effectively characterize the core features of the matrix.

[0045] Difference analysis refers to calculating the difference between corresponding elements of two vectors to obtain a vector reflecting the difference between them. This vector is used to quantify the difference between the reference signal and the output signal, and between the eigenvector and the reference vector. For example, if the first eigenvector is V1=[0.98,0.02,-0.97], and the signal value vector of the reference pair in the first row of the signal matrix is ​​S0=[1,0,-1], then the first difference vector is D1=V1-S0=[-0.02,0.02,0.03]. This vector can reflect the degree of deviation between the first eigenvector and the reference vector.

[0046] The first reference delay is calculated based on the first difference vector and reflects the baseline delay between the test signal and the standard time. It serves as the reference benchmark for subsequent comparisons of actual delays. For example, if the first difference vector D1 = [-0.02, 0.02, 0.03], the corresponding time differences are 0.001ms, 0.001ms, and 0.001ms, respectively. Taking the average value yields a first reference delay of 0.001ms.

[0047] The first reference noise is a parameter that is calculated based on the first difference vector and reflects the reference noise level of the test signal. It serves as a reference for subsequent comparison of actual noise. For example, if the first difference vector D1 = [-0.02, 0.02, 0.03], the first reference noise calculated using the root mean square is 0.024. This value can reflect the noise level under the reference state.

[0048] The second eigenvector is a one-dimensional vector obtained by extracting features from the matrix after removing the first row vector of the signal-time matrix. It is used to analyze the characteristics of the actual output signal. Its extraction method is the same as that of the first eigenvector, and will not be repeated here.

[0049] The second difference vector is a vector obtained by performing difference analysis between the second eigenvector and the first row of the signal matrix. It can reflect the deviation between the second eigenvector and the reference vector. The third difference vector is a vector obtained by obtaining the difference between the second eigenvector and each of the remaining rows in the corresponding signal matrix except for the first row. It is used to calculate the actual delay and actual noise.

[0050] The second actual delay and the second actual noise are obtained based on the difference analysis between the second eigenvector and the first and remaining rows of the signal matrix, reflecting the actual delay and noise level of the chip at different sampling frequencies. For example, the second eigenvector V2=[0.99,0.01,-0.98] and the second difference vector D2=[-0.01,0.01,0.02] of the first row of the signal matrix, the calculated second actual delay is 0.0008ms; the third difference vector D3=[0.001,0.011,0.002] of the third row of the signal matrix, the calculated second actual noise is 0.0066.

[0051] The delay vector and noise vector are one-dimensional vectors composed of the first reference delay and all the second actual delays / noise, which can comprehensively reflect the delay and noise distribution of the chip at different sampling frequencies.

[0052] Fitting analysis uses a mathematical model to fit the relationship between the delay vector, noise vector, and temperature vector, obtaining the influence function of temperature on delay and noise, and quantifying the combined impact of temperature and sampling frequency on chip performance.

[0053] The temperature influence range is determined by a comprehensive function of all test signals, defining the temperature range within which the impact of temperature on delay and noise during board operation falls within the allowable range. This range serves as the basis for subsequent qualification element calibration. For example, if the maximum allowable delay threshold for the board is 0.005ms and the maximum noise threshold is 0.05ms, after considering all test signals, the temperature influence range is determined to be -10℃ to 50℃. Within this range, both delay and noise meet the board's operational requirements.

[0054] The calibration of qualified elements is based on the temperature influence range. Elements in the temperature matrix that fall within this range are marked as "qualified," while those that do not are marked as "unqualified." This process is used to screen sampling frequencies that meet the temperature requirements. For example, if the temperature influence range is -10℃ to 50℃, elements in the temperature matrix with temperatures of 42℃, 43℃, and 31℃ are all within this range and are marked as "qualified." If an element has a temperature of 50℃, which exceeds the temperature influence range, it is marked as "unqualified."

[0055] The recommendation coefficient is a coefficient assigned to each sampling frequency based on the calibration results of qualified elements, reflecting its suitability. The larger the coefficient, the better the suitability of the sampling frequency. For example, if the number of qualified elements corresponding to the 5 sampling frequencies are 3, 3, 3, 3, and 3 respectively, and considering the magnitude of delay and noise, the calculated recommendation coefficients are 0.8, 0.85, 0.95, 0.9, and 0.88 respectively. The recommendation coefficient = percentage of qualified elements × (1 - normalized delay value) × (1 - normalized noise value), with a coefficient range of 0 to 1. The normalized delay value and normalized noise value are obtained by dividing the delay and noise by their respective maximum thresholds, ensuring that the coefficient comprehensively reflects the suitability and performance of the sampling frequency.

[0056] The optimal sampling frequency is selected by choosing the sampling frequency corresponding to the highest coefficient from all recommended coefficients. This frequency serves as the subsequent operating frequency of the ADC chip, ensuring that the chip operates at its optimal performance. For example, in the recommended coefficients mentioned above, 0.95 corresponds to a sampling frequency of 64kSPS, so 64kSPS is chosen as the operating frequency of the ADC chip. Specifically, the software iterates through the array of recommended coefficients, finds the index corresponding to the maximum value, and the sampling frequency corresponding to this index is the optimal sampling frequency. This is then written to the ADC chip's sampling rate control register via CPU configuration instructions.

[0057] The beneficial effects of the above technical solution are as follows: through test circuit construction, signal acquisition, feature extraction and fitting analysis, the optimal operating frequency of the ADC chip is accurately selected, the influence of temperature on delay and noise is fully considered, a reasonable temperature influence range is determined, the working stability of the chip under different operating conditions is ensured, the quantitative evaluation of the recommendation coefficient provides an objective basis for frequency selection, effectively improves the digital processing accuracy of the ADC chip, provides a high-quality data foundation for subsequent multi-parameter joint judgment, and ensures the stability and reliability of the overall performance of the board.

[0058] This invention provides a method for determining reusable simulated Pyrofuse boards based on multi-parameter joint determination, including: [The method is described in the original text, but the provided text is incomplete and cannot be translated accurately.] Construct sampling frequency vector ,in, Let i be the i-th sampling frequency; Constructing a vector-level frequency-temperature coupled fitting function ,in, This is the frequency-temperature coupling weight matrix, with elements... This represents the weight of the effect of temperature on the delay corresponding to the i-th sampling frequency at the j-th sampling frequency; For Hadamah accumulation; This is the delay reference vector, corresponding to the basic delay value of each sampling frequency under rated temperature conditions; It is the adaptive residual correction vector, and ,and This is the correction factor for delayed residuals; It is an n-order identity matrix; This is the delay residual vector obtained from the previous fitting; Constructing nonlinear dimensional coupling functions ,in, This is the frequency-temperature coupling weight matrix corresponding to the noise; Frequency sensitivity index; , These are respectively the sampling frequency vectors The power-extended vector, the quadratic extension vector of the temperature vector; This is the noise reference vector, corresponding to the basic noise value of each sampling frequency under rated temperature conditions; This is the noise adaptive residual correction vector, and ,and This is the noise residual correction factor; It is a symbolic function; It is an L1 norm; This is the noise residual vector obtained from the previous fitting.

[0059] Frequency-temperature coupling weight matrix , It reflects the degree to which temperature affects delay / noise at different sampling frequencies. A matrix, where elements are influence weights, quantifies the coupling effect between frequency and temperature. For example, the weight matrix corresponding to delay. .

[0060] In this embodiment, the delay reference vector This is a vector composed of the base delay values ​​corresponding to each sampling frequency under the rated temperature of 25℃, serving as a benchmark reference for delay fitting. For example, at the rated temperature of 25℃, the base delay values ​​for each sampling frequency are as follows: =0.0005ms =0.0006ms =0.0007ms =0.0008ms =0.0009ms, then the vector =[0.0005,0.0006,0.0007,0.0008,0.0009] (unit: ms).

[0061] Adaptive residual correction vector This is a vector used to correct the error of the fitted model. It is calculated based on the previous fitting residuals and the weight matrix, and can improve the accuracy and robustness of the fitted model. For example, the delay residual correction coefficient. =0.05, n=3rd order identity matrix The previous fitted delay residual vector =[0.0001,-0.0002,0.0001], then we can get .

[0062] Delay residual correction factor This is a constant used to adjust the residual correction strength. It is set according to the magnitude of the fitting error to ensure the appropriateness of the residual correction. For example, based on multiple fitting experiments... Set it to 0.05. If the fitting error is large (MSE ≥ 0.0001), adjust it to 0.1; if the error is small (MSE < 0.00001), adjust it to 0.02.

[0063] Previous Fit Delay Residual Vector It is a vector formed by the difference between the delay value calculated by the model and the measured delay value during the previous fitting process, which provides a basis for residual correction.

[0064] In this embodiment, the frequency sensitivity index This is a constant reflecting the sensitivity of sampling frequency to noise, obtained through experimental calibration, and can improve the accuracy of noise fitting models. For example, based on different... Comparison of fitting effects for values ​​(0.5, 1, 1.5, 2), when The goodness of fit is maximized to 0.96 when the coefficient of performance is 1.2, therefore, the value is set to... =1.2.

[0065] Sampling frequency vector Power-extended vector It is each element in the sampling frequency vector A new vector formed by exponentiation is used to enhance the nonlinear characterization of the effect of frequency on noise.

[0066] The quadratic extension vector of the temperature vector It is a new vector formed by the square of each element in the temperature vector, used to enhance the nonlinear characterization of the effect of temperature on noise.

[0067] Noise reference vector It is a vector composed of the basic noise values ​​corresponding to each sampling frequency under the rated temperature operating conditions, which serves as the benchmark reference for noise fitting.

[0068] Noise residual correction factor It is a constant used to adjust the strength of noise residual correction, and... The calibration method is consistent to ensure the appropriateness of noise residual correction. For example, setting... =0.03, and dynamically adjusted according to the noise fitting error during the fitting process to make the noise fitting MSE≤0.0001.

[0069] The sign function is a function that returns the sign of the input value. It returns 1 when the input value is greater than 0, 0 when it is equal to 0, and -1 when it is less than 0. It is used for sign calibration of the noise residual correction vector.

[0070] The beneficial effects of the above technical solution are as follows: by constructing a vector-level frequency-temperature coupled fitting function and a nonlinear dimension coupled function, the combined influence of temperature and sampling frequency on delay and noise is accurately quantified, solving the problem of insufficient fitting accuracy of a single factor; the introduction of an adaptive residual correction vector improves the accuracy and robustness of the fitting model and can effectively compensate for model errors; the calibration of the frequency sensitivity index optimizes the noise fitting effect, ensuring that the noise model can accurately reflect the nonlinear influence of frequency; the final influence function can accurately predict the delay and noise levels under different operating conditions, providing a scientific and accurate basis for the subsequent determination of the temperature influence range and the selection of the sampling frequency.

[0071] This invention provides a method for determining reusable simulated Pyrofuse boards based on multi-parameter joint evaluation. It comprehensively analyzes the influence functions of all test signals to determine the temperature influence range, including: Based on the maximum allowable delay threshold and maximum noise threshold when the board is working, the effective temperature range corresponding to each group of test signals is selected. The effective temperature range is the temperature range in which the temperature-delay influence function calculation result of the corresponding group of test signals does not exceed the maximum delay threshold and the temperature-noise influence function calculation result does not exceed the maximum noise threshold. Based on the similarity between each group of test signals and the actual working signals of the board, the weight coefficients of each group of test signals are quantified, and the sum of all weight coefficients is 1. Calculate the weighted average of the lower limit values ​​of the effective temperature range corresponding to all test signals, and subtract the confidence correction value determined by the variance of the fitting residuals of each group of test signals to obtain the lower limit value of the comprehensive temperature influence range. Calculate the weighted average of the upper limits of the effective temperature ranges corresponding to all test signals, and add the confidence correction value to obtain the upper limit of the comprehensive temperature influence range; The interval formed by the lower and upper limits of the comprehensive temperature influence range is determined as the final temperature influence range.

[0072] In this embodiment, the maximum latency threshold and the maximum noise threshold are the maximum allowable latency and noise values ​​when the board is operating normally. Exceeding these thresholds will affect the judgment accuracy and board performance, and are the core basis for selecting the effective temperature range. For example, based on the board's test accuracy requirements, the maximum latency threshold is 0.005ms, and the maximum noise threshold is 0.05ms. The threshold settings are determined based on the board's application scenario and performance indicators, through tolerance analysis of test errors, and by referring to the performance parameters of similar products in the industry to ensure the rationality and practicality of the threshold settings.

[0073] The effective temperature range is the temperature range within which the calculated temperature-delay / noise influence function for a set of test signals does not exceed the maximum threshold. It reflects the temperature boundary under which the board operates normally under that test signal. For example, the effective temperature range for a sine wave test signal (highly similar to the actual working signal) is T1_low = -8℃ and T1_high = 82℃; the effective temperature range for a square wave test signal is T2_low = -12℃ and T2_high = 88℃; and the effective temperature range for a triangular wave test signal is T3_low = -10℃ and T3_high = 85℃. The method for selecting the effective temperature range is as follows: iterate through the temperature values ​​(-40℃ to 125℃), substitute each temperature value into the temperature-delay influence function and the temperature-noise influence function to calculate the delay and noise values, and select the temperature range where both do not exceed the maximum threshold. This is the effective temperature range for the test signal.

[0074] Similarity refers to the degree of similarity between the test signal and the actual operating signal of the board in terms of amplitude, frequency, waveform, etc. The cosine similarity algorithm is used to calculate the coefficient between 0 and 1. The higher the similarity, the greater the reference value of the effective temperature range corresponding to the test signal for the actual operation of the board. For example, if the actual operating signal of the board is a sine wave (amplitude 1Vpp, frequency 1kHz), the similarity of the three sets of test signals are S1=0.95 (same waveform, same frequency, similar amplitude), S2=0.7 (square wave, different waveform), and S3=0.85 (triangular wave, different waveform but similar frequency).

[0075] The weighting coefficients are coefficients obtained by quantifying the similarity between each set of test signals and the actual working signals of the board. The sum of all weighting coefficients is 1, which is used to synthesize the effective temperature range of each test signal. For example, based on the above similarity S1=0.95, S2=0.7, S3=0.85, the weighting coefficients are calculated as follows: W1=0.95 / (0.95+0.7+0.85)=0.38, W2=0.7 / (0.95+0.7+0.85)=0.28, W3=0.85 / (0.95+0.7+0.85)=0.34, and the sum of the weighting coefficients is 1.

[0076] In this embodiment, the confidence correction value is calculated by the weighted square root of the variance of the fitting residuals of each group of test signals, with the weights being the weighting coefficients of each group of test signals, ensuring that the confidence correction value can reflect the comprehensive impact of the fitting error on the temperature range.

[0077] The lower limit of the overall temperature influence range is obtained by calculating the weighted average of the lower limits of the effective temperature ranges corresponding to all test signals, and then subtracting the confidence correction value. This value reflects the lowest temperature boundary for the board to operate normally. The lower limits of the effective temperature ranges for each test signal are T1_low, T2_low, and T3_low. The weighted average is: (W1×T1_low)+(W2×T2_low)+(W3×T3_low). Subtracting the confidence correction value yields the lower limit of the overall temperature influence range.

[0078] If the upper limits of the effective temperature ranges of each test signal are T1_high, T2_high, and T3_high, respectively, the upper limit of the comprehensive temperature influence range can be obtained by calculating the weighted average value and adding the confidence correction value.

[0079] The final temperature influence range is an interval consisting of the lower and upper limits of the comprehensive temperature influence range.

[0080] The beneficial effects of the above technical solution are as follows: by screening the effective temperature range based on the maximum threshold, determining the weight coefficient by combining similarity, and optimizing the boundary by introducing confidence correction values, the temperature influence range can be scientifically and accurately determined; this temperature influence range comprehensively considers the influence of different test signals, and can fully reflect the actual operating temperature boundary of the board, providing a reliable basis for subsequent calibration of qualified elements of the temperature matrix and selection of sampling frequency, ensuring that the board can work stably and accurately under different temperature conditions.

[0081] This invention provides a method for determining a reusable simulated pyrofuse board based on multiple parameters. Before the relay of the CPU driver output module closes, the method includes: An auxiliary circuit for relay sensitivity testing is constructed. The auxiliary circuit includes an indicator light circuit, a multi-dimensional signal sampling circuit, a drive closed-loop adjustment unit, and a freewheeling diode performance monitoring unit. The auxiliary circuit is connected in series with the relay and the CPU output drive circuit. The multi-dimensional signal sampling circuit is used to collect the actual current signal of the relay coil, the contact action feedback signal, and the coil magnetic field strength signal. The freewheeling diode performance monitoring unit is used to collect the reverse current signal of the freewheeling diode connected in parallel across the coil. Based on the multi-dimensional signal sampling circuit and the freewheeling diode performance monitoring unit, the signal data corresponding to each pre-signal is collected. Combined with the status signal of the indicator light circuit, the critical drive parameter set corresponding to the relay contact trigger action is recorded. The critical drive parameter set includes the critical drive current, the critical signal slope, the critical magnetic field strength, and the critical remaining energy value calculated based on the reverse current of the freewheeling diode. Based on the set of critical driving parameters and the rated operating parameters of the relay, the sensitivity adaptation coefficient is calculated; Determine whether the sensitivity adaptation coefficient is within the preset qualified range. If not, adaptively adjust the gain of the CPU output drive signal and the output level of the relay drive chip. If the sensitivity matching coefficient is still not within the acceptable range after adjustment, replace the relay and the freewheeling diode with matching parameters, and retest. If so, in the scenario of simulating the repeated use of the pyrofuse board, after performing a preset number of closing-opening actions, the decay rate of the reverse current of the freewheeling diode and the stability of the coil magnetic field strength are monitored simultaneously. If the sensitivity adaptation coefficient is still within the acceptable range after performing a preset number of closing-opening actions, and the stability is greater than the preset value, it is determined that the CPU is allowed to drive the relay to close. Otherwise, replace the relay.

[0082] Preferably, the sensitivity adaptation coefficient is the normalized deviation weighted sum of the critical drive current, critical signal slope, critical residual energy value and corresponding rated parameters, multiplied by the correlation coefficient between coil magnetic field strength and contact action, and the weight of the deviation weighted sum is jointly determined by the reusability requirements of the pyrofuse board and the reverse current decay characteristics of the freewheeling diode.

[0083] In this embodiment, the auxiliary circuit is connected in series with the relay and the CPU output drive circuit, enabling comprehensive acquisition of relay operating status parameters. For example, the indicator light circuit uses an LED indicator in series with a 1kΩ current-limiting resistor to visually display the on / off status of the relay contacts; the multi-dimensional signal sampling circuit uses an ACS712 current sensor to collect the coil current, an EE-SX672 photoelectric sensor to collect the contact action feedback signal, and an A1324 Hall sensor to collect the coil magnetic field strength; the drive closed-loop adjustment unit uses an LM324 operational amplifier to construct the feedback adjustment circuit; and the freewheeling diode performance monitoring unit uses an ACS712 current sensor to collect the reverse current of the freewheeling diode.

[0084] The indicator light circuit is a component of the auxiliary circuit for relay sensitivity testing. It is used to visually display the on / off state of the relay contacts, providing visual feedback for sensitivity testing. When the relay contacts are closed, the LED indicator light illuminates; when the contacts are open, the LED indicator light turns off.

[0085] The multi-dimensional signal sampling circuit is used to acquire the actual current signal of the relay coil, the contact action feedback signal, and the coil magnetic field strength signal, comprehensively reflecting the relay's operating status. For example, the current sensor ACS712 has a measurement range of 0~5A and is used to acquire the actual coil current; the photoelectric sensor EE-SX672 has a response time ≤1ms and is used to detect whether the contacts have actuated; the Hall sensor A1324 has a measurement range of 0~20mT and is used to acquire the coil magnetic field strength. The current sensor is connected in series in the coil circuit, the photoelectric sensor is installed near the relay contacts, and the Hall sensor is attached to the coil surface. The output signals of each sensor are filtered and amplified before being input to the CPU's ADC input channel. The sampling rate is set to 10kSPS to ensure the accuracy and real-time performance of the signal acquisition.

[0086] The drive closed-loop adjustment unit is a circuit used to adaptively adjust the CPU output drive signal. It can adjust the drive parameters based on the relay sensitivity test results to achieve sensitivity adaptation. For example, if an operational amplifier LM324 is used to construct a proportional-integral (PI) adjustment circuit, the input signal is the deviation of the sensitivity adaptation coefficient from the acceptable range, and the output signal is used to adjust the gain of the CPU output drive signal. The drive closed-loop adjustment unit is connected to the CPU's GPIO output interface and the relay driver chip, and dynamically adjusts the drive signal gain through a PI adjustment algorithm to ensure that the relay sensitivity adaptation coefficient remains within the acceptable range.

[0087] The freewheeling diode performance monitoring unit is a circuit used to acquire the reverse current signal of the freewheeling diode connected in parallel across the coil, reflecting the performance status of the freewheeling diode. For example, if an ACS712 current sensor is used to acquire the reverse current, with a measurement range of 0~1A, the freewheeling diode is connected in parallel across the coil, and the current sensor is connected in series in the freewheeling diode circuit. The acquired reverse current signal is input to the CPU's ADC input channel to calculate the critical remaining energy value and monitor the reverse current decay rate.

[0088] In this embodiment, the critical drive current refers to the minimum drive current of the coil when the relay contacts are just triggered to close, and it is a key parameter reflecting the sensitivity of the relay. For example, if the coil drive current is gradually increased during testing, and the relay contacts close for the first time and the indicator light illuminates when the current reaches 30mA, then the critical drive current is 30mA. The CPU outputs a PWM signal to control the output current of the drive chip, gradually increasing the PWM duty cycle. Simultaneously, a current sensor monitors the coil current in real time, and combined with the indicator light status, records the current value at the moment the contacts close; this value is the critical drive current.

[0089] The critical signal slope refers to the rising slope of the critical drive current, reflecting the impact of the rate of change of the drive signal on the relay operation. The CPU's timer records the time it takes for the critical drive current to rise from 0 to the critical value. Combined with the critical drive current value, the critical signal slope is calculated. For example, if the time for the critical drive current to rise from 0mA to 30mA is 6ms, then the critical signal slope is 30mA / 6ms = 5mA / ms.

[0090] The critical magnetic field strength refers to the minimum magnetic field strength generated by the coil when the relay contacts are just triggered to close, and it is related to the critical drive current. While testing the critical drive current, the coil magnetic field strength is monitored in real time using a Hall sensor. The magnetic field strength value at the instant the contacts close is recorded as the critical magnetic field strength. For example, if the coil magnetic field strength collected by the Hall sensor is 8 mT when the coil current reaches the critical drive current of 30 mA, then the critical magnetic field strength is 8 mT.

[0091] In this embodiment, real-time data of the reverse current is collected by the freewheeling diode performance monitoring unit, and an integration algorithm is written in the CPU to calculate the critical remaining energy value. The decay curve of the reverse current of the freewheeling diode is as follows: The remaining energy is obtained through integration. .

[0092] In this embodiment, the preset qualified range is determined based on the rated parameters of the relay, the test accuracy requirements of the board, and the actual application scenario. It is stored in the CPU's Flash and serves as the basis for sensitivity adaptation judgment. For example, according to the performance requirements of the board, the preset qualified range is set to 0.9~1.1.

[0093] Adaptive adjustment involves the CPU controlling the output level of the driver chip via the GPIO interface and adjusting the drive signal gain through the PWM signal. After each adjustment, the critical drive parameters are re-acquired, and the sensitivity adaptation coefficient is calculated until the coefficient enters the qualified range. For example, if the sensitivity adaptation coefficient is 0.85 (below the lower limit of the qualified range), the gain of the CPU output drive signal is increased (from 1x to 1.2x), and the output level of the driver chip ULN2003 is adjusted from low to medium. The sensitivity adaptation coefficient is then retested.

[0094] The reverse current decay rate of a freewheeling diode refers to the degree of decay of the reverse current of the freewheeling diode after a preset number of closed-open cycles in a simulated Pyrofuse board reuse scenario, reflecting the performance stability of the freewheeling diode. For example, if the initial peak reverse current is 1A, and after 1000 closed-open cycles the peak reverse current is 0.95A, then the decay rate is (1-0.95) / 1×100%=5%.

[0095] In this embodiment, after performing a preset number of closing-opening actions, the critical magnetic field strength is retested, and the rate of change from the initial value is calculated. If the rate of change is ≤5%, the stability is deemed acceptable. For example, if the initial critical magnetic field strength is 8mT, and after performing 1000 closing-opening actions, the critical magnetic field strength is 8.1mT, then the rate of change is (8.1-8) / 8×100%=1.25%, indicating good stability.

[0096] The beneficial effects of the above technical solution are as follows: by building an auxiliary circuit for relay sensitivity testing, the operating status parameters of the relay are comprehensively collected, enabling accurate testing and adaptation of relay sensitivity; by determining and adaptively adjusting the sensitivity adaptation coefficient, the compatibility between the relay and the CPU drive signal is ensured; by simulating stability testing in repeated use scenarios, the reliable performance of the relay under high-frequency use is guaranteed; the entire process is progressive, effectively avoiding fuse simulation errors caused by relay sensitivity mismatch or performance instability, ensuring the reliability of the board for repeated use and the accuracy of fuse simulation.

[0097] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for determining reusable simulated Pyrofuse boards based on multi-parameter joint determination, characterized in that, include: Step 1: Configure test parameters based on the communication module. The test parameters include the current threshold ITH, time threshold Ts, energy threshold Eth, and the target resistance value of the variable resistor module. Step 2: The high-speed ADC sampling module measures the voltage across the low-temperature drift SHUNT resistor connected in series in the controlled loop, and calculates the loop current signal in real time based on Ohm's law. This signal is then processed by a 24-bit... After the chip performs digital processing, the data is transmitted to the CPU via the SPI interface; Step 3: The CPU processes the digitized current signal in real time and uses multi-parameter joint judgment logic to determine whether the fuse condition is met. The multi-parameter joint judgment logic is: parallel execution of energy accumulation integral judgment and overcurrent duration-threshold current joint judgment. If any judgment condition is met, the fuse is triggered. Step 4: If the fuse condition is met, the CPU drives the relay of the output module to disconnect, making the output module present a high impedance state, simulating the Pyrofuse fuse action; Step 5: After the test is completed, a reset operation is performed through the reset module. The relay of the output module closes and restores the low-resistance state, enabling the board to be reused.

2. The method for determining reusable simulated Pyrofuse boards based on multi-parameter joint determination according to claim 1, characterized in that, The energy accumulation integral determination includes: Using the trapezoidal integral method according to the formula E = Calculate the cumulative energy E, where, The equivalent resistance of the melt is denoted as I(t); the real-time current is denoted as t; and t is a time variable, ranging from t1 to t2. When both conditions are met and When the circuit breaker condition is met, it is determined that the circuit breaker condition has been met. The joint determination of overcurrent duration and threshold current includes: Real-time statistics of overcurrent duration T, when simultaneously satisfying and When the condition is met, it is determined that the circuit breaker condition is met.

3. The method for determining reusable simulated Pyrofuse boards based on multi-parameter joint determination according to claim 1, characterized in that, The pyrofuse board includes a high-speed ADC sampling module, a variable resistor module, a CPU, an output module, a reset module, a communication module, and a power supply module. The high-speed ADC sampling module is connected to the CPU via an SPI interface. The variable resistor module, output module, and reset module are connected to the CPU via GPIO interfaces. The communication module is connected to the CPU via a CAN / EtherCAT interface. The power supply module provides stable 5V and 12V power to each module. The output module uses an electromagnetic relay connected in series in the controlled circuit to switch between high and low impedance states.

4. The method for determining reusable simulated Pyrofuse boards based on multi-parameter joint determination according to claim 1, characterized in that, Through 24 people Before the chip undergoes digital processing, it includes: A test circuit including the chip is constructed, and different test signals are sent to the test circuit. The output signals of the chip under different sampling frequencies of different test signals are captured respectively. A signal-time matrix based on each test signal is constructed. In the signal matrix, each element in the first row is a reference pair consisting of the test signal value and the standard test time, and each element in the remaining rows is an actual pair consisting of the output signal value and the signal test time. The maximum temperature value of the chip under different sampling frequencies of the same test signal is measured to construct a temperature vector and obtain a temperature matrix; The signal-time matrix is ​​analyzed to obtain a first eigenvector, and the first eigenvector is compared with the first row of the signal matrix to obtain a first difference vector. At the same time, the matrix after removing the first row vector of the signal-time matrix is ​​used to obtain a second eigenvector, and the second eigenvector is compared with the first row of the signal matrix to obtain a second difference vector. Based on the first difference vector and the second difference vector, the first reference delay and the first reference noise of the corresponding test signal are determined. At the same time, the third difference vector of the second feature vector and each remaining row of the corresponding signal matrix (excluding the first row) are obtained and compared with the first difference vector to obtain the corresponding second actual delay and the second actual noise. The delay vector is formed by the first reference delay under the same test signal and all the second actual delays. At the same time, the noise vector is formed by the first reference noise under the same test signal and all the second actual noises. The delay vector noise vector Temperature vector A fitting analysis was performed to determine the effect function of temperature on delay and noise, where, This represents the delay value at the i-th sampling frequency; This represents the noise value at the i-th sampling frequency. Let be the maximum temperature value at the i-th sampling frequency; n is the number of sampling frequencies. A comprehensive analysis of the influence functions of all test signals is performed to determine the temperature influence range. The qualified elements of the temperature matrix are then calibrated according to the temperature influence range to obtain the recommendation coefficient for each sampling frequency. The sampling frequency corresponding to the highest coefficient among all recommendation coefficients is selected as the subsequent operating frequency of the chip.

5. The method for determining reusable simulated Pyrofuse boards based on multi-parameter joint determination according to claim 4, characterized in that, Determine the effect function of temperature on delay and noise, including: Construct sampling frequency vector ,in, Let i be the i-th sampling frequency; Constructing a vector-level frequency-temperature coupled fitting function ,in, This is the frequency-temperature coupling weight matrix, with elements... This represents the weight of the effect of temperature on the delay corresponding to the i-th sampling frequency at the j-th sampling frequency; For Hadamah accumulation; This is the delay reference vector, corresponding to the basic delay value of each sampling frequency under rated temperature conditions; It is the adaptive residual correction vector, and ,and This is the correction factor for delayed residuals; It is an n-order identity matrix; This is the delay residual vector obtained from the previous fitting; Constructing nonlinear dimensional coupling functions ,in, This is the frequency-temperature coupling weight matrix corresponding to the noise; Frequency sensitivity index; , These are respectively the sampling frequency vectors The power-extended vector, the quadratic extension vector of the temperature vector; This is the noise reference vector, corresponding to the basic noise value of each sampling frequency under rated temperature conditions; This is the noise adaptive residual correction vector, and ,and This is the noise residual correction factor; It is a symbolic function; It is an L1 norm; This is the noise residual vector obtained from the previous fitting.

6. The method for determining reusable simulated Pyrofuse boards based on multi-parameter joint determination according to claim 4, characterized in that, A comprehensive analysis of the influence functions of all test signals is performed to determine the range of temperature influence, including: Based on the maximum allowable delay threshold and maximum noise threshold when the board is working, the effective temperature range corresponding to each group of test signals is selected. The effective temperature range is the temperature range in which the temperature-delay influence function calculation result of the corresponding group of test signals does not exceed the maximum delay threshold and the temperature-noise influence function calculation result does not exceed the maximum noise threshold. Based on the similarity between each group of test signals and the actual working signals of the board, the weight coefficients of each group of test signals are quantified, and the sum of all weight coefficients is 1. Calculate the weighted average of the lower limit values ​​of the effective temperature range corresponding to all test signals, and subtract the confidence correction value determined by the variance of the fitting residuals of each group of test signals to obtain the lower limit value of the comprehensive temperature influence range. Calculate the weighted average of the upper limits of the effective temperature ranges corresponding to all test signals, and add the confidence correction value to obtain the upper limit of the comprehensive temperature influence range; The interval formed by the lower and upper limits of the comprehensive temperature influence range is determined as the final temperature influence range.

7. The method for determining reusable simulated Pyrofuse boards based on multi-parameter joint determination according to claim 4, characterized in that, Before the relay of the CPU driver output module closes, the following steps are included: An auxiliary circuit for relay sensitivity testing is constructed. The auxiliary circuit includes an indicator light circuit, a multi-dimensional signal sampling circuit, a drive closed-loop adjustment unit, and a freewheeling diode performance monitoring unit. The auxiliary circuit is connected in series with the relay and the CPU output drive circuit. The multi-dimensional signal sampling circuit is used to collect the actual current signal of the relay coil, the contact action feedback signal, and the coil magnetic field strength signal. The freewheeling diode performance monitoring unit is used to collect the reverse current signal of the freewheeling diode connected in parallel across the coil. Based on the multi-dimensional signal sampling circuit and the freewheeling diode performance monitoring unit, the signal data corresponding to each pre-signal is collected. Combined with the status signal of the indicator light circuit, the critical drive parameter set corresponding to the relay contact trigger action is recorded. The critical drive parameter set includes the critical drive current, the critical signal slope, the critical magnetic field strength, and the critical remaining energy value calculated based on the reverse current of the freewheeling diode. Based on the set of critical driving parameters and the rated operating parameters of the relay, the sensitivity adaptation coefficient is calculated; Determine whether the sensitivity adaptation coefficient is within the preset qualified range. If not, adaptively adjust the gain of the CPU output drive signal and the output level of the relay drive chip. If the sensitivity matching coefficient is still not within the acceptable range after adjustment, replace the relay and the freewheeling diode with matching parameters, and retest. If so, in the scenario of simulating the repeated use of the pyrofuse board, after performing a preset number of closing-opening actions, the decay rate of the reverse current of the freewheeling diode and the stability of the coil magnetic field strength are monitored simultaneously. If the sensitivity adaptation coefficient is still within the acceptable range after performing a preset number of closing-opening actions, and the stability is greater than the preset value, it is determined that the CPU is allowed to drive the relay to close. Otherwise, replace the relay.

8. The method for determining reusable simulated Pyrofuse boards based on multi-parameter joint determination according to claim 7, characterized in that, The sensitivity adaptation coefficient is the normalized weighted sum of the critical drive current, critical signal slope, critical residual energy value and the corresponding rated parameters, multiplied by the correlation coefficient between the coil magnetic field strength and the contact action. The weight of the deviation weighted sum is determined by the reusability requirements of the pyrofuse board and the reverse current decay characteristics of the freewheeling diode.

Images