A self-compensating contact resistance online monitoring device and method in a high-temperature environment

Abstract

The present invention relates to a self-compensating contact resistance online monitoring device and method in a high-temperature environment. The device includes: a contact point for forming a contact resistance to be measured; a thermocouple for real-time monitoring of the temperature inside a high-temperature box; a constant current source system for applying a constant test current to the contact point and dynamically adjusting the output current value; a signal conditioning circuit for filtering out high-frequency interference in the voltage signal; a data acquisition card for high-speed acquisition of the voltage signal; and a processor for calculating the true value of the contact resistance after compensation. The beneficial effect of the present invention is that in the LSTM network based on Kalman filtering and physical constraints of the present invention, the KF dynamically fuses the observed value with the model prediction, outputs the denoised base value state in real time, and provides a clean input for the LSTM. Based on the denoising result of the KF, the LSTM models the nonlinear relationship between temperature and resistance, compensating for the linear assumption defect of the KF.

Description

Technical Field

[0001] The present invention relates to the field of contact resistance online monitoring, and more specifically, to a self-compensating contact resistance online monitoring device and method under high temperature environment. Background Art

[0002] Contact resistance is one of the key indicators of electrical connectors and is used to characterize the contact performance of electrical connectors. When monitoring the contact resistance of contact points in a high-temperature environment, due to the temperature-resistance coupling effect, the resistance change consists of two parts: one part is a reversible change, that is, the resistance change caused by temperature fluctuations, and the other part is an irreversible change, that is, the permanent increase in resistance caused by material oxidation and wear, resulting in contact performance degradation. Contact resistance monitoring needs to remove the resistance change caused by temperature fluctuations and retain the amount of contact resistance degradation. When the degradation value reaches the failure threshold, the contact point is judged to be failed. Conventional contact resistance monitoring is to wait for the contact point to cool to room temperature after the test equipment is stopped, and then use an ammeter and voltmeter to calculate according to Ohm's law after the resistance change caused by temperature fluctuations has naturally disappeared. However, this method cannot achieve online monitoring of contact resistance and cannot capture transient events at the contact point. In addition, the operation is cumbersome, which increases the test cost.

[0003] The Kalman filter (KF) is based on a linear state transition and observation model. However, thermal expansion hysteresis and material phase transitions can cause nonlinear resistance responses, making it difficult for the KF to accurately model them. Furthermore, the process noise covariance and observation noise covariance must be manually set, resulting in poor robustness to non-Gaussian and time-varying noise. The KF also lacks long-term modeling capabilities and cannot capture aging trends, making long-term predictions unreliable. When the LSTM is directly fed with noisy raw signals, high-frequency noise can cause the model to overfit transient fluctuations, increasing steady-state prediction errors. A purely data-driven LSTM may generate predictions that violate physical laws. Summary of the Invention

[0004] The purpose of the present invention is to address the deficiencies of the prior art and to provide a self-compensating contact resistance online monitoring device and method in a high temperature environment.

[0005] In a first aspect, a self-compensating contact resistance online monitoring device in a high temperature environment is provided, comprising:

[0006] The contact point is used to form the contact resistance to be measured; the contact point is arranged in a high temperature box and high temperature resistant wires are connected to both ends of the contact point;

[0007] Thermocouple, used to monitor the temperature inside the high-temperature box in real time; the output end of the thermocouple is connected to the processor;

[0008] A constant current source system is used to apply a constant test current to the contact point and dynamically adjust the output current value; the constant current source system is connected to the processor;

[0009] A signal conditioning circuit, used for filtering out high-frequency interference in the voltage signal; the signal conditioning circuit is connected to the contact point;

[0010] A data acquisition card is used for high-speed acquisition of voltage signals; the data acquisition card is connected to the signal conditioning circuit;

[0011] The processor is used to calculate the preliminary value of the contact resistance of the contact point and compensate for the contact resistance increment caused by temperature change to obtain the actual value of the contact resistance after compensation.

[0012] Preferably, the processor comprises:

[0013] A calculation module, configured to calculate a preliminary contact resistance value based on the current value of the constant current source system and the voltage value of the data acquisition card;

[0014] The compensation module is based on Kalman filtering and long short-term memory network, combines the temperature data of the thermocouple, compensates for the resistance change caused by temperature, and outputs the true value of the contact resistance.

[0015] Preferably, it also includes: a feedback module; the input end of the feedback module is connected to the output end of the processor compensation module, and according to a preset correspondence table between the actual value gradient of the contact resistance and the measured current value, a feedback signal of the measured current value corresponding to the current actual value gradient of the contact resistance is output to the input end of the constant current source system.

[0016] Preferably, it also includes: a momentary disconnection monitoring module, the input end of the momentary disconnection monitoring module is connected to the output end of the processor compensation module, and judges whether the actual value of the contact resistance exceeds the momentary disconnection resistance threshold and the duration threshold based on the momentary disconnection resistance threshold and the duration threshold.

[0017] Preferably, the instantaneous interruption monitoring module is also connected to an alarm.

[0018] In a second aspect, a method for online monitoring of contact resistance with self-compensation in a high-temperature environment is provided, which is performed by any device described in the first aspect, comprising:

[0019] S1. Apply a constant current to the contact point through a constant current source system and collect the voltage value;

[0020] S2, the calculation module obtains the preliminary value of contact resistance based on the current and voltage values;

[0021] S3, the compensation module combines the temperature value of the thermocouple, uses Kalman filtering and a physically constrained LSTM network to compensate for the contact resistance increment caused by temperature changes and outputs the true value of the contact resistance;

[0022] S4, the feedback module outputs a feedback signal of the measured current value corresponding to the current contact resistance true value gradient to the constant current source system input terminal according to the preset correspondence table between the contact resistance true value gradient and the measured current value; the constant current source system input terminal dynamically adjusts the output current value according to the feedback signal;

[0023] S5. The instantaneous disconnection monitoring module determines whether the actual value of the contact resistance exceeds the instantaneous disconnection resistance threshold. If the actual value of the contact resistance exceeds the instantaneous disconnection resistance threshold and the duration exceeds the time threshold, an alarm is triggered.

[0024] Preferably, S3 includes:

[0025] S301, removing random noise of the temperature signal by weighted averaging, and retaining the low-frequency trend;

[0026] S302, using a central difference method to calculate the rate of change of the temperature estimate value and the preliminary contact resistance value after denoising;

[0027] S303, performing Kalman filter state estimation, separating base resistance and base temperature from noise observations through multivariable joint estimation;

[0028] S304, using a dynamic gating mechanism to distinguish whether the current state belongs to a transient phase or a steady-state phase, and generating a mixed input for the LSTM network;

[0029] S305. Predict the reversible resistance change caused by temperature through the physically constrained LSTM network and output the true value of the contact resistance.

[0030] Preferably, in S304, the LSTM network includes a short-term memory layer and a long-term memory layer.

[0031] Preferably, in S4, the input end of the constant current source system dynamically adjusts the output current value according to the feedback signal, including: if the current contact resistance increases, the feedback is given to the constant current source system to increase the current output; if the current contact resistance decreases, the feedback is given to the constant current source system to reduce the current output.

[0032] The beneficial effects of the present invention are:

[0033] 1. In this LSTM network based on Kalman filtering and physical constraints, the KF dynamically fuses observations with model predictions, outputting a denoised baseline state in real time and providing clean input to the LSTM. Based on the KF denoising results, the LSTM models the nonlinear relationship between temperature and resistance, compensating for the KF's linear assumption. Dynamic gating adaptively switches input between transient and steady-state phases. Oxidation kinetic equations are forcibly injected into the LSTM's long-term memory cells, ensuring that the degradation rate increases exponentially with increasing temperature. Even with insufficient training data, physical constraints prevent the model from generating inconsistent material science predictions. The KF noise covariance and LSTM weights are jointly optimized via gradient descent to adapt to different operating conditions.

[0034] 2. The present invention eliminates the need to stop the test equipment and perform measurements after cooling during the entire measurement process, saving test time. The compensation module adaptively corrects the contact resistance based on the current temperature in the high-temperature chamber, improving the accuracy of online contact resistance monitoring. The feedback model adopted by the present invention enables the constant current source to adaptively adjust the measurement current based on contact resistance fluctuations. Furthermore, the transient state of the contact resistance is monitored in real time during the measurement process, avoiding the disadvantage of being unable to capture transient events in static measurements.

[0035] 3. Contact resistance is a key indicator for evaluating the lifespan and reliability of electrical connectors. Static, timed measurement methods cannot capture contact resistance fluctuations in real time, ignore the impact of transient disconnections on connector lifespan and reliability, and are time-consuming and labor-intensive. The self-compensating online contact resistance monitoring device and method of the present invention can significantly reduce testing costs, obtain more test data, and improve the efficiency of electrical connector lifespan assessment. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 This is a schematic diagram of the structure of the self-compensating contact resistance online monitoring device under high temperature environment provided by this application;

[0037] Figure 2 Flowchart of the method for online monitoring of contact resistance under high temperature environment with self-compensation provided by this application;

[0038] Figure 3 This is a flow chart for outputting the true value of contact resistance;

[0039] Explanation of the accompanying reference numerals: 1 constant current source system, 2 signal conditioning circuit, 3 data acquisition card, 4 contact point, 5 thermocouple, 6 processor, 61 calculation module, 62 compensation module, 7 feedback module, 8 instantaneous interruption monitoring module, 9 alarm, 10 high temperature box. DETAILED DESCRIPTION

[0040] The present invention will be further described below with reference to the following examples. The following examples are provided only to facilitate understanding of the present invention. It should be noted that, without departing from the principles of the present invention, it is possible for a person skilled in the art to make various modifications to the present invention, and such improvements and modifications fall within the scope of the claims of the present invention.

[0041] Example 1:

[0042] To solve the problems of the prior art, Example 1 of the present application provides a self-compensating contact resistance online monitoring device in a high-temperature environment, comprising:

[0043] The contact point 4 is used to form the contact resistance to be measured; the contact point 4 is arranged in the high temperature box 10 and high temperature resistant (temperature resistant above 500° C.) wires are welded at both ends of the contact point 4 .

[0044] Thermocouple 5 is used to monitor the temperature inside the high-temperature box 10 in real time. The output end of thermocouple 5 is connected to processor 6. For example, thermocouple 5 is arranged near contact point 4 in the high-temperature box 10. For example, the distance between thermocouple 5 and contact point 4 does not exceed 10 mm.

[0045] A constant current source system 1 is used to apply a constant test current to the series circuit of the contact point 4-constant current source system 1-processor 6; the constant current source system 1 is arranged outside the high temperature box 10 and the output end of the constant current source system 1 is connected to the current application wire of the contact point 4 and the input end of the processor 6;

[0046] The signal conditioning circuit 2 is used to filter out high-frequency interference in the voltage signal; the signal conditioning circuit 2 is arranged outside the high-temperature box 10 and is connected to the voltage detection wire of the contact point 4, and the output end is connected to the data acquisition card 3;

[0047] The data acquisition card 3 is used to acquire voltage signals at high speed. The data acquisition card 3 is arranged after the signal conditioning circuit 2, and the input end is connected to the signal conditioning circuit 2. Specifically, the sampling rate of the data acquisition card 3 is at least 250 kHz, which can capture voltage values ​​at high speed and is used to capture 10μs-level contact point transient events.

[0048] The processor 6 is configured to calculate a preliminary contact resistance value at the contact point 4 and compensate for the incremental contact resistance caused by temperature changes to obtain a true contact resistance value after compensation. The input end of the processor 6 is connected to the output end of the thermocouple 5, the output end of the constant current source system 1, the current application wire of the contact point 4, and the output end of the data acquisition card 3, wherein the input end of the processor 6, the output end of the constant current source system 1, and the current application wire of the contact point 4 form a series circuit; the output end of the processor 6 is connected to the input end of the constant current source system 1, an alarm 9, a display device, and / or a storage device.

[0049] The processor 6 is also used to obtain the ambient temperature value of the contact point 4 and the preliminary value of the contact resistance through the thermocouple 5, suppress the input noise by Kalman filtering, output the denoised base value state, balance the transient and steady states by dynamic gating, separate the transient effects and long-term degradation trends through the dual-scale memory structure, and embed the oxidation kinetic equation network to ensure that the long-term degradation conforms to the physical laws and output the true value of the contact resistance.

[0050] The processor includes:

[0051] The calculation module 61 is used to calculate a preliminary contact resistance value according to the current value of the constant current source system 1 and the voltage value of the data acquisition card 3 .

[0052] Specifically, the first input end and the second input end of the calculation module 61 form a series circuit with the output end of the constant current source system 1 and the current application wire of the contact point 4, and the third input end is connected to the output end of the data acquisition card 3, and is used to calculate the preliminary value of the contact resistance in real time online based on the current value of the constant current source system 1 and the voltage value of the data acquisition card 3.

[0053] The compensation module 62 compensates for the resistance change caused by temperature based on Kalman filtering and long short-term memory network in combination with the temperature data of the thermocouple 5 and outputs the true value of the contact resistance.

[0054] Specifically, the first input end of the compensation module 62 is connected to the output end of the calculation module 61, and the second input end is connected to the output end of the thermocouple 5. According to the preliminary value of the contact resistance of the calculation module 61 and the ambient temperature value of the contact point 4 obtained by the thermocouple 5, the contact resistance increment caused by the temperature change is compensated by the LSTM network based on Kalman filtering and physical constraints to obtain the true value of the contact resistance after compensation. The output end is externally connected to a display device and / or storage device for displaying and / or storing the true value of the contact resistance.

[0055] Example 2:

[0056] Based on Example 1, Example 2 of the present application provides a more specific self-compensating contact resistance online monitoring device in a high-temperature environment, including: a contact point 4, a thermocouple 5, a constant current source system 1, a data acquisition card 3, and a processor 6.

[0057] At least one thermocouple 5 is provided, and the data acquisition card 3 does not directly collect the voltage value of the contact point 4. Instead, the signal conditioning circuit 2 filters out high-frequency interference before inputting the voltage value into the data acquisition card 3. In a specific implementation, the contact point 4 and thermocouple 5 can be placed in a constant temperature test chamber, a thermal cycle test chamber, or a temperature-vibration combined test chamber. The thermocouple 5 can be a commonly used commercially available thermocouple.

[0058] Contact point 4 is connected in series with constant current source system 1 and calculation module 61 in processor 6 via high-temperature-resistant wires, with the measurement current supplied by constant current source system 1. Contact point 4 is also connected to signal conditioning circuit 2 via high-temperature-resistant wires. The voltage measured at contact point 4 by signal conditioning circuit 2 undergoes high-frequency filtering and is then transmitted to data acquisition card 3. Multiple contact points can be connected in series, with the current supplied uniformly by constant current source system 1. Data acquisition card 3 typically employs a multi-channel 16-bit data acquisition card.

[0059] The calculation module 61 collects the voltage value collected by the acquisition card 2 and the current value supplied by the constant current source system 1 to calculate the preliminary value of the contact resistance. However, the preliminary value of the contact resistance contains interference caused by high temperature.

[0060] The compensation module 62 compensates for the contact resistance increment caused by temperature change based on Kalman filtering and a physical-constrained LSTM network to obtain a true value of the contact resistance after compensation.

[0061] During specific implementation, the processor adopts a multi-core processor, such as the ARM Cortex-A series.

[0062] The processor 6 is connected to a display device and / or a storage device to display and / or store the actual value of the contact resistance in real time.

[0063] During specific implementation, different display devices or storage devices may be selected according to actual needs, such as a liquid crystal display or an SD memory card.

[0064] In addition, it also includes: a feedback module 7; the input end of the feedback module 7 is connected to the output end of the compensation module 62 of the processor 6, and according to the preset contact resistance true value gradient and measured current value correspondence table, the feedback signal of the measured current value corresponding to the current contact resistance true value gradient is output to the input end of the constant current source system 1. The input end of the constant current source system 1 is connected to the feedback module 7, and receives the feedback signal to dynamically adjust the output current value. If the current contact resistance increases, it is fed back to the constant current source system 1 to increase the current output. If the current contact resistance decreases, it is fed back to the constant current source system 1 to reduce the current output. For example, when the contact point is in a high-resistance state, the current is increased to 100mA, and when it is in a low-resistance state, it is reduced to 10mA.

[0065] The system further includes a momentary disconnection monitoring module 8, the input of which is connected to the output of the compensation module 62 of the processor 6. The module determines, based on a momentary disconnection resistance threshold and a duration threshold, whether the actual contact resistance exceeds the momentary disconnection resistance threshold and the duration exceeds the time threshold. The momentary disconnection monitoring module 8 is also connected to an alarm 9. If the actual contact resistance exceeds the momentary disconnection resistance threshold and the duration exceeds the time threshold, an alarm is issued indicating a momentary disconnection event at the contact point.

[0066] In specific implementation, the instantaneous resistance threshold and duration threshold can be set according to actual engineering requirements, for example, setting the contact resistance , duration is the threshold for the momentary interruption event.

[0067] The entire measurement process only requires that the current application wire and voltage detection wire of the contact point 4 and the detection wire of the thermocouple 5 be led out through the wire hole of the high-temperature box 10 and connected to the online monitoring device, so as to realize the online test of the contact resistance in high temperature or temperature cycle or temperature vibration state. The measurement result can filter out the influence of interference signals and temperature drift, improve the accuracy of the measurement result, and can also realize the alarm of the instantaneous disconnection event of the contact point 4, reduce the monitoring cost, and the monitoring process is also very simple.

[0068] It should be noted that the parts in this embodiment that are the same or similar to those in Example 1 can be referenced to each other and will not be described in detail in this application.

[0069] Example 3:

[0070] Based on Example 2, Example 3 of the present application provides a self-compensating contact resistance online monitoring method under high temperature environment, such as Figure 2 Shown, including:

[0071] S1. Apply a constant current to the contact point through a constant current source system and collect the voltage value.

[0072] S2. The calculation module obtains a preliminary value of the contact resistance based on the current and voltage values.

[0073] S3. The compensation module combines the temperature value of the thermocouple and uses the Kalman filter and the physically constrained LSTM network to compensate for the contact resistance increment caused by temperature change and output the true value of the contact resistance.

[0074] like Figure 3 As shown, S3 includes:

[0075] S301 , removing random noise of the temperature signal by weighted averaging, and retaining the low-frequency trend.

[0076] Specifically, the ambient temperature value of the contact point 4 obtained by the thermocouple 5 is Including environmental noise and electromagnetic interference, the preliminary value of contact resistance obtained by the calculation module 61 It includes the reversible resistance change caused by temperature and the irreversible change caused by degradation, and is expressed as:

[0077]

[0078] in, Is the resistance change caused only by irreversible degradation, which is a long-term trend. It is a reversible change caused by temperature fluctuations and is a short-term fluctuation. The random noise of the temperature signal is removed by weighted averaging, retaining the low-frequency trend, which is expressed as:

[0079]

[0080] Where, is the sliding window weight coefficient, satisfying , used to suppress high-frequency noise; n is the window radius, which controls the smoothing intensity.

[0081] S302: Calculate the estimated temperature value after denoising using the central difference method and preliminary contact resistance values The rate of change captures the dynamic change characteristics of the signal and provides a basis for transient detection, which is expressed as:

[0082]

[0083] Where, is the time step used for computing the first derivative using central differencing.

[0084] S303, perform Kalman filter state estimation, and separate the base resistance and base temperature from the noise observation through multivariate joint estimation. Suppress noise and output preliminary denoising results.

[0085] Specifically, define the state vector:

[0086]

[0087] Where, is the base resistance, which indicates the actual resistance after removing the temperature effect and only reflects the degradation state; is the base temperature, which represents the stable temperature estimate after removing noise; , is the rate of change of the base value state.

[0088] Define the state transition model:

[0089]

[0090] Where, is the state transfer matrix, which models the dynamic evolution of state variables, preferably using a first-order motion model; is the process noise, and the covariance matrix Q represents the uncertainty of the model prediction.

[0091] Define the observation model:

[0092]

[0093] Where, is the observation matrix, which defines the linear mapping relationship between state variables and observation signals; is the observation noise, and the covariance matrix R represents the temperature measurement error.

[0094] Define the update steps:

[0095]

[0096] Where, is the Kalman gain, which balances the weights of the predicted value and the observed value, through the covariance matrix Dynamic calculation.

[0097] S304: Differentiate whether the current state belongs to the transient phase or the steady-state phase through a dynamic gating mechanism, and generate a mixed input for the LSTM network.

[0098] In S304, the gating coefficient is calculated and expressed as:

[0099]

[0100] Where, Sigmoid function, output range (0,1); is the slope coefficient, which controls the steepness of the gate switching; is the threshold value. When the sum of the signal change rates exceeds It is judged as transient stage.

[0101] Generate mixed input to the LSTM network, expressed as:

[0102]

[0103] When the current state is transient , LSTM network gives priority to using KF denoising results to suppress noise interference; when the current stage is dynamic ,LSTM network uses the signal which retains the original details and avoids the phase delay introduced by filtering.

[0104] In S304, the LSTM network contains a short-term memory layer and a long-term memory layer. The short-term memory layer captures second-level fluctuations, such as transient resistance changes caused by sudden temperature changes, which can be expressed as:

[0105]

[0106] The long-term memory layer is:

[0107] ,

[0108] Where, A physically constrained forget gate modulated by temperature; It is the input gate that controls the ratio of new information written.

[0109] Physical constraints The specific design is:

[0110]

[0111] Where, is the activation energy of the oxidation reaction; R is the gas constant; is the base temperature of the KF estimate. As the temperature increases, the forget gate coefficient approaches 1, retaining more historical memory and simulating the cumulative effect of oxidation.

[0112] The regularization loss term is calculated to force the long-term memory update rate to conform to the Arrhenius equation to ensure the physical plausibility of the degradation prediction, which is expressed as:

[0113]

[0114] Where k is the oxidation reaction rate constant.

[0115] S305. Predict the reversible resistance change caused by temperature through the physically constrained LSTM network and output the true value of the contact resistance.

[0116] Specifically, the final output result is generated in S305, including the LSTM prediction of the reversible resistance change caused by temperature. , expressed as:

[0117]

[0118] Where, is a weight matrix that integrates short-term and long-term temperature features and quantifies their dynamic contributions to the resistance change; is a bias term that provides a temperature-independent resistance reference offset used to correct the baseline value of the model output; It is a long-term characteristic, indicating a historical cumulative effect or a slow-changing trend.

[0119] Subtract the temperature effect from the original resistance to get a base resistance that reflects only degradation That is the true value of the contact resistance, expressed as:

[0120]

[0121] S4. The feedback module outputs a feedback signal of the measured current value corresponding to the current contact resistance true value gradient to the constant current source system input end according to the preset contact resistance true value gradient and measured current value correspondence table; the constant current source system input end dynamically adjusts the output current value according to the feedback signal.

[0122] In S4, the input end of the constant current source system dynamically adjusts the output current value according to the feedback signal, including: if the current contact resistance increases, the feedback is fed back to the constant current source system to increase the current output; if the current contact resistance decreases, the feedback is fed back to the constant current source system to reduce the current output.

[0123] S5. The instantaneous disconnection monitoring module determines whether the actual value of the contact resistance exceeds the instantaneous disconnection resistance threshold. If the actual value of the contact resistance exceeds the instantaneous disconnection resistance threshold and the duration exceeds the time threshold, an alarm is triggered.

[0124] It should be noted that the method provided in this embodiment is the method corresponding to the device provided in Example 2. Therefore, the parts in this embodiment that are the same or similar to those in Example 2 can be referenced to each other and will not be repeated in this application.

[0125] Example 4:

[0126] Based on Example 3, Example 4 of the present application provides a more specific method for online monitoring of contact resistance with self-compensation in a high-temperature environment, including:

[0127] 1. Input signal and preprocessing (t=10s)

[0128] 1) The original signal is as follows

[0129] , including ±2℃ noise; , including ±5mΩ measurement noise.

[0130] 2) Through sliding average filtering For preprocessing, the window weight is selected as , get the temperature estimate after removing high-frequency noise

[0131]

[0132] 3) Take , calculate the rate of change using the central difference method

[0133]

[0134] 2. Kalman filter (KF) state estimation

[0135] 1) KF initialization (t=9s state)

[0136]

[0137] Covariance matrix The diagonal elements of represent the uncertainty of the state variables, and The variance of is 0.1, the variance of the rate of change is 0.01, and the covariance matrix Expressed as:

[0138]

[0139] 2) Prediction step (t = 10s), assuming that the base resistance and temperature change linearly with time, such as the first-order motion model, the state transition equation is:

[0140]

[0141] The state transfer matrix F is:

[0142]

[0143] Update the forecast covariance:

[0144]

[0145] Process noise , which represents the uncertainty of the model prediction.

[0146] 3) Update step (t=10s), first get the observation vector:

[0147]

[0148] Then calculate the observation residuals:

[0149]

[0150] The second row of the observation matrix H , where α=0.004 is the resistance temperature coefficient, and the observation matrix H is

[0151]

[0152] 3. Dynamic gating

[0153] Calculate the gating coefficient:

[0154]

[0155] , ,when When , the gating coefficient approaches 1. The current stage is transient, and the KF result is used first to suppress noise.

[0156] Generating mixed input , to avoid the original noise interfering with the LSTM network.

[0157]

[0158] 4. LSTM Networks and Physical Constraints

[0159] The short-term memory layer models second-level fluctuations and inputs mixed features , output 32-dimensional hidden state , capturing the transient response of resistance caused by sudden temperature changes.

[0160] The long-term memory layer models hourly aging. First, the forget gate is modulated as follows:

[0161]

[0162] At this time, the temperature , high temperature accelerates oxidation, the forget gate retains more historical memory, and simulates degradation accumulation.

[0163] Update cell status , is the new information candidate value at the current moment, expressed as:

[0164]

[0165] Verify physical constraints, expressed as:

[0166]

[0167] When the long-term memory update rate deviates from the theoretical value, The gradient forces LSTM to adjust weights so that the long-term memory update rate Approaching the theoretical oxidation rate.

[0168] Decoupling the temperature effect to obtain the temperature-induced resistance change predicted by LSTM, is the weight matrix, is the bias term, obtained by fitting the training data:

[0169]

[0170] Finally, the base resistance, that is, the true value of the contact resistance, is calculated as:

[0171]

[0172] With KF In contrast, LSTM is further modified through nonlinear modeling.

[0173] All parameters (Q, R of KF, LSTM weights, gating coefficients) are jointly optimized through end-to-end training. The above embodiment is just one of the cases.

[0174] It should be noted that the parts in this embodiment that are the same or similar to those in Example 3 can be referenced to each other and will not be repeated in this application.

Claims

1. A self-compensating contact resistance online monitoring method in a high temperature environment, characterized in that: The monitoring is performed by a self-compensating contact resistance online monitoring device in a high temperature environment. The self-compensating contact resistance online monitoring device in a high temperature environment includes: The contact point is used to form the contact resistance to be measured; the contact point is arranged in a high temperature box and high temperature resistant wires are connected to both ends of the contact point; A thermocouple is used to monitor the temperature inside the high-temperature box in real time; the output end of the thermocouple is connected to the processor; and the distance between the thermocouple and the contact point does not exceed 10 mm; A constant current source system is used to apply a constant test current to the contact point and dynamically adjust the output current value; the constant current source system is connected to the processor; A signal conditioning circuit, used for filtering out high-frequency interference in the voltage signal; the signal conditioning circuit is connected to the contact point; A data acquisition card is used for high-speed acquisition of voltage signals; the data acquisition card is connected to the signal conditioning circuit; a processor for calculating a preliminary value of the contact resistance of the contact point and compensating for an increase in the contact resistance caused by temperature change to obtain a true value of the contact resistance after compensation; Feedback module; the feedback module input end is connected to the processor compensation module output end, and according to the preset contact resistance true value gradient and measured current value correspondence table, outputs the feedback signal of the measured current value corresponding to the current contact resistance true value gradient to the constant current source system input end; A transient interruption monitoring module, wherein the input end of the transient interruption monitoring module is connected to the output end of the processor compensation module, and determines whether the actual value of the contact resistance exceeds the transient interruption resistance threshold and the duration exceeds the time threshold according to the transient interruption resistance threshold and the duration threshold; The processor includes: A calculation module, configured to calculate a preliminary contact resistance value based on the current value of the constant current source system and the voltage value of the data acquisition card; The compensation module, based on Kalman filtering and long short-term memory network, combines the temperature data of the thermocouple to compensate for the resistance change caused by temperature and output the true value of the contact resistance; the instantaneous disconnection monitoring module is also connected to the alarm; The self-compensating contact resistance online monitoring method under high temperature environment includes: S1. Apply a constant current to the contact point through a constant current source system and collect the voltage value; S2, the calculation module obtains the preliminary value of contact resistance based on the current and voltage values; S3, the compensation module combines the temperature value of the thermocouple, uses Kalman filtering and a physically constrained LSTM network to compensate for the contact resistance increment caused by temperature changes and outputs the true value of the contact resistance; S3 includes: S301, removing random noise of the temperature signal by weighted averaging, and retaining the low-frequency trend; S302, using a central difference method to calculate the rate of change of the denoised temperature estimate and the preliminary contact resistance value; S303, performing Kalman filter state estimation, separating base resistance and base temperature from noise observations through multivariable joint estimation; S304, using a dynamic gating mechanism to distinguish whether the current state belongs to a transient phase or a steady-state phase, and generating a mixed input for the LSTM network; When the current state is transient , the LSTM network gives priority to using the KF denoising result to suppress noise interference; when the current state is steady ,LSTM network uses the signal that retains the original details and avoids the phase delay introduced by filtering; The LSTM network contains a short-term memory layer and a long-term memory layer; the long-term memory layer is: ， Where, A physically constrained forget gate modulated by temperature; It is the input gate, which controls the ratio of new information written; The specific design is: Where, is the activation energy of the oxidation reaction; R is the gas constant; is the base temperature estimated by KF; The regularization loss term is calculated to force the long-term memory update rate to conform to the Arrhenius equation to ensure the physical rationality of the degradation prediction, which is expressed as: Where, k is the oxidation reaction rate constant; S305, predicting the reversible resistance change caused by temperature through the physically constrained LSTM network and outputting the true value of the contact resistance; S4, the feedback module outputs a feedback signal of the measured current value corresponding to the current contact resistance true value gradient to the constant current source system input terminal according to the preset correspondence table between the contact resistance true value gradient and the measured current value; the constant current source system input terminal dynamically adjusts the output current value according to the feedback signal; S5. The instantaneous disconnection monitoring module determines whether the actual value of the contact resistance exceeds the instantaneous disconnection resistance threshold. If the actual value of the contact resistance exceeds the instantaneous disconnection resistance threshold and the duration exceeds the time threshold, an alarm is triggered.

2. The method for online monitoring of contact resistance with self-compensation in a high temperature environment according to claim 1, characterized in that: In S4, the input end of the constant current source system dynamically adjusts the output current value according to the feedback signal, including: if the current contact resistance increases, the feedback is fed back to the constant current source system to increase the current output; if the current contact resistance decreases, the feedback is fed back to the constant current source system to reduce the current output.

Images