Resistor degradation prediction method and related equipment

Abstract

The invention relates to the technical field of industrial maintenance, and provides a resistor degradation prediction method and related equipment. The method comprises the following steps: acquiring resistance value data of a resistor at T moments; performing resistance prediction based on the resistance data of the resistor at all moments by using an improved variational mode decomposition algorithm to obtain a resistance prediction result of the resistor; and performing degradation analysis on the resistor according to the resistance prediction result to obtain a degradation prediction result of the resistor. According to the method, the data cost of degradation prediction of the resistor can be reduced, and the degradation prediction precision can be improved.

Description

Technical Field

[0001] This application relates to the field of industrial maintenance technology, and in particular to a method and related equipment for predicting resistor degradation. Background Technology

[0002] Resistors play an indispensable role in modern electronic devices, with their main functions including current limiting, voltage division, and signal conditioning. As electronic devices evolve towards higher efficiency and miniaturization, the requirements for resistor performance and stability are also increasing. The prediction and analysis of resistor performance has significant application value in many fields. For example, in industry, the performance of resistors in industrial equipment affects the operating status of the equipment, and the resistance value of resistors in industrial products is a crucial indicator of product quality. Therefore, understanding and predicting changes in resistor performance not only helps improve the reliability of electronic equipment but also effectively reduces maintenance costs and extends product lifespan.

[0003] Data-driven lifetime prediction methods have become mainstream in recent years. These methods involve establishing a historical database of resistor monitoring data throughout its operation using testing equipment, building a corresponding predictive model from this database, and then combining this database with actual resistor operation monitoring data to predict degradation. Currently, domestic resistor testing equipment mainly relies on imported instruments assembled together. While it can operate automatically, it has a limited number of resistors that can be tested at a time, the equipment is bulky, it cannot analyze long-term degradation, data costs are high, and degradation prediction accuracy is low. Summary of the Invention

[0004] This application provides a resistor degradation prediction method and related equipment, which can solve the problems of high data cost and low prediction accuracy in resistor degradation prediction.

[0005] In a first aspect, embodiments of this application provide a resistor degradation prediction method, which includes:

[0006] Obtain the resistance value of the resistor at T time points; the Tth time point is the current time.

[0007] An improved variational mode decomposition algorithm is used to predict the resistance value of a resistor based on its resistance data at all times, resulting in a resistance prediction result. This prediction result is used to describe the resistance value of the resistor at multiple future times after the current time.

[0008] Degradation analysis is performed on the resistor based on the resistance prediction results to obtain the degradation prediction results of the resistor; the degradation prediction results are used to describe the time it takes for the resistance value of the resistor to degrade to the resistance threshold.

[0009] Optionally, an improved variational mode decomposition algorithm is used to predict the resistance value based on the resistance data of the resistor at all times, resulting in a predicted resistance value, including:

[0010] An improved variational mode decomposition algorithm is used to perform variational mode decomposition on all resistance data of the resistor to obtain multiple mode components;

[0011] The resistance value of the resistor is predicted based on all modal components, and the resistance prediction result of the resistor is obtained.

[0012] Optionally, an improved variational mode decomposition algorithm is used to perform variational mode decomposition on all resistance data of the resistor, obtaining multiple mode components, including:

[0013] The variational mode decomposition algorithm is optimized by parameter optimization to obtain the optimized variational mode decomposition algorithm;

[0014] The optimized variational mode decomposition algorithm is used to decompose all resistance data of the resistor to obtain multiple mode components.

[0015] Optionally, the variational mode decomposition algorithm is optimized to obtain an optimized variational mode decomposition algorithm, including:

[0016] Acquire resistance data of multiple sample resistors at multiple historical moments and initialize multiple parameter combinations; the parameter combinations include the values ​​of multiple parameters in the variational mode decomposition algorithm;

[0017] For each parameter combination, the parameter combination is substituted into the variational mode decomposition algorithm, and the variational mode decomposition algorithm is used to perform variational mode decomposition on all resistance data of each sample resistor to obtain multiple initial mode components of each sample resistor. Resistance prediction is performed based on all initial mode components of each sample resistor to obtain the predicted resistance value. The mean absolute error corresponding to the parameter combination is calculated based on all predicted resistance values ​​and the actual resistance values ​​of all sample resistors.

[0018] The optimal variational mode decomposition algorithm is obtained by substituting the parameter combination that minimizes the mean absolute error into the variational mode decomposition algorithm.

[0019] Optionally, variational mode decomposition (VMD) is performed on all resistance data of each sample resistor using a variational mode decomposition algorithm to obtain multiple initial mode components for each sample resistor, including:

[0020] For each sample resistor, perform the following steps:

[0021] All resistance data of the sample resistor are decomposed into multiple first-mode components;

[0022] Construct the objective function;

[0023] With the objective function minimization as the goal, each first modal component is updated to obtain the second modal component corresponding to each first modal component;

[0024] Determine whether all second-mode components satisfy the iteration stopping condition;

[0025] If so, then each second mode component is used as the initial mode component;

[0026] Otherwise, treat each second modal component as a first modal component, and return the steps of updating each first modal component with the objective function minimization as the goal, to obtain the second modal component corresponding to each first modal component.

[0027] Optionally, the objective function is:

[0028]

[0029] in, Indicates the first The first mode component Indicates the first The center frequency of the first mode component Indicates the number of the first mode components. This indicates finding the partial derivative. Represents the Dirac function, Represents the imaginary unit. Indicates time, Indicates the first The first mode component at time... The signal This represents all resistance values ​​of the sample resistor.

[0030] Optionally, degradation analysis is performed on the resistor based on the resistance prediction results to obtain the resistor degradation prediction results, including:

[0031] The earliest future moment when the resistance value is less than or equal to the resistance threshold is determined from the resistance prediction results, and the earliest future moment is taken as the target future moment.

[0032] The time corresponding to the target's future moment is used as the degradation prediction result of the resistor.

[0033] Secondly, embodiments of this application provide a resistor degradation prediction device, comprising:

[0034] The acquisition module is used to acquire the resistance value data of the resistor at T time points; the Tth time point is the current time.

[0035] The prediction module is used to predict the resistance value of the resistor based on the resistance value data of the resistor at all times using an improved variational mode decomposition algorithm, and obtain the resistance prediction result of the resistor. The resistance prediction result is used to describe the resistance value of the resistor at multiple future times after the current time.

[0036] The degradation analysis module is used to perform degradation analysis on the resistor based on the resistance prediction results, and obtain the degradation prediction results of the resistor; the degradation prediction results are used to describe the time it takes for the resistance value of the resistor to degrade to the resistance threshold.

[0037] Thirdly, embodiments of this application provide a terminal device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the resistor degradation prediction method described above.

[0038] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the aforementioned resistor degradation prediction method.

[0039] The above-mentioned solution in this application has the following beneficial effects:

[0040] In the embodiments of this application, the resistance value data of the resistor at T time points is obtained, and then an improved variational mode decomposition algorithm is used to predict the resistance value based on the resistance value data of the resistor at all time points, thus obtaining the resistance prediction result of the resistor. Finally, degradation analysis of the resistor is performed based on the resistance prediction result to obtain the degradation prediction result of the resistor. Notably, obtaining the resistance value data of the resistor at multiple time points eliminates the need to monitor the data of the resistor throughout its entire operation process, reducing data costs. Predicting the resistance value at multiple future time points enables analysis of the resistance value change of the resistor over a future period, realizing the analysis of long-term resistor degradation. Degradation prediction is performed based on the resistance prediction result with long-term resistance value changes, effectively improving the accuracy of resistor degradation prediction.

[0041] Other beneficial effects of this application will be described in detail in the following detailed description section. Attached Figure Description

[0042] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0043] Figure 1 A flowchart of a resistor degradation prediction method provided in an embodiment of this application;

[0044] Figure 2 This is an architecture diagram of an automatic resistance testing system provided in an embodiment of this application;

[0045] Figure 3 This is an architecture diagram of a test subsystem provided in an embodiment of this application;

[0046] Figure 4 A circuit diagram for resistance value testing provided in one embodiment of this application;

[0047] Figure 5 A schematic diagram illustrating the prediction results of a random forest model provided in an embodiment of this application;

[0048] Figure 6 A schematic diagram illustrating the prediction results of VMD and random forest models provided in an embodiment of this application;

[0049] Figure 7 A schematic diagram illustrating the prediction results of the method of this application provided in an embodiment of this application;

[0050] Figure 8 A schematic diagram of the structure of a resistor degradation prediction device provided in an embodiment of this application;

[0051] Figure 9 This is a schematic diagram of the structure of a terminal device provided in an embodiment of this application. Detailed Implementation

[0052] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0053] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0054] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0055] As used in this application specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if detected [the described condition or event]" may be interpreted, depending on the context, as meaning "once determined," "in response to determination," "once detected [the described condition or event]," or "in response to detection [the described condition or event]."

[0056] Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0057] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0058] To address the issues of high data costs and low accuracy in existing resistor degradation prediction methods, this application provides a resistor degradation prediction method. This method acquires the resistor's resistance data at T time points, then uses an improved variational mode decomposition algorithm to predict the resistance based on the resistance data at all times, obtaining the resistor's predicted resistance result. Finally, degradation analysis is performed on the resistor based on the predicted resistance result to obtain the resistor degradation prediction result. Notably, acquiring resistance data at multiple time points eliminates the need to monitor the resistor's entire operating process, reducing data costs. Predicting resistance values ​​at multiple future time points allows for analysis of resistance changes over a future period, enabling long-term resistor degradation analysis. By performing degradation prediction based on resistance prediction results with long-term resistance changes, the accuracy of resistor degradation prediction is effectively improved.

[0059] The resistor degradation prediction method provided in this application will be illustrated below.

[0060] like Figure 1 As shown, the resistor degradation prediction method provided in this application includes the following steps:

[0061] Step 11: Obtain the resistance data of the resistor at T time points.

[0062] The resistors mentioned above are resistors used in industrial equipment or products, and the T-th time point represents the current time. The resistance data mentioned above includes information such as the resistor model, batch number, resistance value, and environmental stress.

[0063] In some embodiments of this application, an automated resistance testing system can be built using equipment such as a tester, and the resistance value data of the resistor can be obtained using the automated resistance testing system.

[0064] For example, the above-mentioned automatic resistance testing system is as follows: Figure 2 As shown, it includes a big data platform, a monitoring subsystem (with multiple monitoring units), and a testing subsystem. The testing subsystem includes multiple testers that test the resistance data of the test fixture's industrial control computer under different temperature environments. The industrial control computer is used to control the testers. The different temperature environments include high temperature, normal temperature, and low temperature environments. Each monitoring unit monitors the data output by the testers under a set of high temperature, normal temperature, and low temperature environments, and uploads the obtained resistance data to the big data platform.

[0065] The above-mentioned test subsystems are as follows Figure 3 As shown, the system includes: test software (running on an industrial computer, enabling human-machine interaction, timed / manual data acquisition mode startup, test data storage, and display of data, status, and alarm information); a tester (the uninterruptible power supply (UPS) shown in the diagram, including test boards, UPS, Ethernet switch, and power module; one set of testers consists of 3 test boards + UPS; the test equipment has high and low electrical stress outputs, two models, realizing electrical stress testing and resistance value acquisition testing functions); and experimental equipment (the ambient temperature device shown in the diagram, including test boards; one set of experimental equipment consists of 3 test boards, representing the object under test running in the experimental environment). The test boards, composed of communication units, AD units, DA units, and IO units, connect one-to-one with the test boards for electrical stress testing and resistance value acquisition testing. The Ethernet switch is used for command and data interaction between the various test boards and the test software. The power module provides power to the tester. Each test board consists of 100 resistors (5 batches). 2 per batch Ten types (models) were tested in environmental equipment.

[0066] In practical applications of the aforementioned automatic resistance testing system, the resistors under test are soldered onto test boards and connected to the testing instrument via test fixtures. Test boards composed of identical resistors are divided into four groups and connected to the testing instrument. An aging test system is connected, and the test fixtures are placed in ambient temperature chambers (25℃, -40℃, and 60℃). Test software runs on an industrial computer, automatically applying corresponding aging electrical stresses (e.g., 3.3V, 5V) based on the resistance's withstand capacity. The system automatically switches to measurement mode every four hours, triggering the testing instrument to collect resistance data. The test software sends commands via the Modbus TCP protocol to read measurement data from the testing instrument's registers. Data is stored in real-time in a database, supporting queries by model, batch, time, and other criteria. Test data is extracted from the database and stored as a CSV file, containing timestamps, environmental stress, resistor model, batch, measured value, and correction value.

[0067] The circuit diagram for resistance value testing is as follows: Figure 4 As shown, the device includes a tester and a test board. The tester includes an electrical stress resistor U2, a precision voltage source U3, 10 resistor switches k1-k10, a self-test resistor R1, a first switch k11, a second switch k12, a third switch k13, a fourth switch k14, a 24-bit AD chip, a first resistor R3, a second resistor R4, a third resistor R5, and an operational amplifier. The test board includes 10 resistors Rx to be tested. The positive terminal of the electrical stress resistor is connected to the first terminal of the second switch, and the negative terminal of the electrical stress resistor is connected to the negative terminal of the precision voltage source. The second terminal of the second switch is connected to the second terminal of the third switch, the first terminal of the first switch, and the first terminal of each of the 10 resistor switches. One end is connected to the first end of the third switch, which is connected to the positive terminal of the precision voltage source. The second end of the first switch is connected to the first end of the self-built resistor. The second end of each resistor switch is connected to the first end of a corresponding resistor. The second end of the self-test resistor is connected to the second end of each resistor, the first end of the fourth switch, the first end of the first resistor, and the positive input terminal of the operational amplifier. The second end of the fourth switch is connected to the second end of the first resistor and the first end of the third resistor. The second end of the third resistor is connected to the first end of the second resistor, the output terminal of the operational amplifier, and the input terminal of the AD chip. The second end of the second resistor is connected to the inverting input terminal of the operational amplifier. Figure 4 In this diagram, V1 represents the voltage at the positive input terminal of the operational amplifier, and V2 represents the voltage at the output terminal of the operational amplifier.

[0068] It should be noted that after obtaining the resistance data, it is necessary to perform preprocessing such as missing value handling, outlier handling, and normalization.

[0069] Missing value handling: Two methods are used to handle missing values. If there are few missing values, they are directly deleted. If there are many missing values, Lagrange interpolation is used to fill in the missing values. By performing high-order nonlinear fitting on the index, better results can be obtained.

[0070] Outlier Handling: Since the resistance data is a non-stationary, non-normally distributed time series, and exhibits a clear trend, Z-score is used to identify outliers. After detecting outliers, they are removed, and a moving window method is used to set an appropriate window size and threshold to fill the gap.

[0071] Standardization adjusts data to a fixed range, typically [0,1], facilitating data processing. In machine learning, normalization accelerates model convergence, especially when processing water quality data, where different indicators have significant differences in dimensions and values, requiring normalization to ensure they are within the same order of magnitude. The Min-Max method preserves the relationships existing in the original data while eliminating the influence of data value range and dimensions.

[0072] Specifically, since missing values ​​may exist in the data, affecting the accuracy of subsequent predictions, it is necessary to handle these missing values. Lagrange interpolation, a type of polynomial interpolation, is computationally simple and capable of fitting high-order data, showing good results in imputing missing values. Therefore, Lagrange interpolation is used to impute missing values ​​in resistance data. By performing high-order nonlinear fitting on the index, better results are obtained. The Lagrange interpolation expression is as follows:

[0073]

[0074] in For the first and the Data.

[0075] Meanwhile, since the collected raw aging data contains a large amount of noise and spikes, and these outliers can interfere with the algorithm training process, it is necessary to clean the raw data. Because the resistance data is a non-stationary, non-normally distributed time series, and exhibits a clear trend, a moving window outlier handling method is chosen, and Z-score is used in each window to identify outliers. Z-score is a statistical outlier detection method that detects outliers in resistance by calculating the standard deviation distance between each data point and the mean.

[0076]

[0077] in, It is a single raw data value. It is the overall mean. It is the overall standard deviation.

[0078] Specifically, normalization adjusts data to a fixed range, typically [0,1], to facilitate data processing. In machine learning, normalization can accelerate model convergence, especially when processing water quality data, where different indicators have significant differences in dimensions and values, requiring normalization to ensure they are within the same order of magnitude. The Min-Max method preserves the relationships existing in the original data while eliminating the influence of data value range and dimensions.

[0079] The formula for standardization using the Min-Max method is: ,in, and These are the maximum and minimum values ​​of all samples, respectively. It is the normalized value.

[0080] Step 12: Using the improved variational mode decomposition algorithm, resistance prediction is performed based on the resistance data of the resistor at all times, and the resistance prediction result of the resistor is obtained.

[0081] The above resistance prediction results are used to describe the resistance value of the resistor at multiple future moments after the current moment.

[0082] In some embodiments of this application, the steps described above for using an improved variational mode decomposition algorithm to predict the resistance of a resistor based on its resistance data at all times, and obtaining the predicted resistance result of the resistor, include:

[0083] The first step is to use an improved variational mode decomposition algorithm to perform variational mode decomposition on all resistance data of the resistor to obtain multiple mode components.

[0084] All the resistance values ​​of the above resistors are the resistance values ​​of the resistors obtained in step 11 at T time points.

[0085] Specifically, the variational mode decomposition algorithm is optimized to obtain an optimized variational mode decomposition algorithm. Using the optimized variational mode decomposition algorithm, all resistance data of the resistor are decomposed to obtain multiple mode components (that is, all resistance data of the resistor are used as input data of the optimized variational mode decomposition algorithm, and multiple mode components are obtained through the calculation of the optimized variational mode decomposition algorithm. The specific decomposition process of the variational mode decomposition algorithm is the same as the process of performing variational mode decomposition on all resistance data of the sample resistor to obtain multiple initial mode components of the sample resistor in the following text. In this step, all resistance data of the resistor are input to obtain multiple mode components of the resistor).

[0086] It should be noted that the specific steps for optimizing the variational mode decomposition algorithm to obtain the optimized algorithm are as follows: acquiring resistance data of multiple sample resistors at multiple historical moments and initializing multiple parameter combinations; the parameter combinations include the values ​​of multiple parameters in the variational mode decomposition algorithm (such as the number of modal components). and penalty factor For each parameter combination, the parameter combination is substituted into the variational mode decomposition algorithm, and the variational mode decomposition algorithm is used to perform variational mode decomposition on all resistance data of each sample resistor to obtain multiple initial mode components of each sample resistor. Resistance prediction is performed based on all initial mode components of each sample resistor to obtain the predicted resistance value. The mean absolute error corresponding to the parameter combination is calculated based on all predicted resistance values ​​and the actual resistance values ​​of all sample resistors. The parameter combination with the smallest mean absolute error is substituted into the variational mode decomposition algorithm to obtain the optimized variational mode decomposition algorithm.

[0087] In some embodiments of this application, the resistance value data of the sample resistors can be obtained using the automated resistance testing system in step 11. Then, the resistance prediction value can be obtained by using the random forest algorithm based on all initial mode components of each sample resistor. The above-mentioned step of using the variational mode decomposition algorithm to perform variational mode decomposition on all resistance value data of each sample resistor to obtain multiple initial mode components of each sample resistor includes:

[0088] For each sample resistor, perform the following steps:

[0089] First, all resistance data of the sample resistors are decomposed into multiple first-mode components.

[0090] Then, construct the objective function.

[0091] Specifically, the objective function is:

[0092]

[0093] in, Indicates the first The first mode component Indicates the first The center frequency of the first mode component Indicates the number of the first mode components. This indicates finding the partial derivative. Represents the Dirac function, Represents the imaginary unit. Indicates time, Indicates the first The first mode component at time... The signal This represents all resistance values ​​of the sample resistor.

[0094] Then, with the objective function minimization as the goal, each first modal component is updated to obtain the second modal component corresponding to each first modal component.

[0095] For example, the alternating direction multiplier method is used to update the first modal component. First, a constrained Lagrange equation is constructed and converted into a frequency domain representation using Fourier transform. Under the constraint of the constrained Lagrange equation, the alternating direction multiplier method is used to update the first modal component and the center frequency. The expression of the constrained Lagrange equation is as follows:

[0096]

[0097] in, For Lagrange units, Indicates the penalty factor. It represents the Lagrange multiplier.

[0098] The expression for the alternating direction multiplier method is:

[0099]

[0100]

[0101]

[0102] in, Indicates the updated result of the first... The second mode component, Indicates the center frequency. Indicates the first The first mode component This represents the updated Lagrange multipliers. This represents the Lagrange multipliers before the update. Indicates the updated result of the first... The center frequency of the second modal component.

[0103] Finally, determine whether all second-mode components satisfy the iteration stopping condition.

[0104] Specifically, the iteration conditions are:

[0105]

[0106] in, This indicates the preset value.

[0107] If so, then each second mode component is used as the initial mode component.

[0108] Otherwise, treat each second modal component as a first modal component, and return the steps of updating each first modal component with the objective function minimization as the goal, to obtain the second modal component corresponding to each first modal component.

[0109] For example, the Optuna optimization algorithm can also be used to optimize the parameters of the variational mode decomposition algorithm to obtain an optimized variational mode decomposition algorithm. The process is as follows:

[0110] First, construct the search space for the parameters to be optimized and define the number of modal components. and penalty factor As the variable to be optimized; secondly, design the objective function objective(trial), which dynamically obtains the current parameter combination through the trial object. In each iteration: ① Based on the current and The process involves: 1) Using Variational Mode Decomposition (VMD) to decompose the preprocessed resistance value sequence into K modal components (IMFs); 2) Constructing a time-series feature matrix based on the IMFs (containing the IMFs within the current and historical windows along with the original signal values); 3) Dividing the training and validation sets into a 7:3 ratio and normalizing the features using MinMaxScaler (fitting scaling parameters only based on training set statistics to avoid data leakage); 4) Training a random forest regression model (fixed hyperparameters: 150 trees, maximum depth 8) and calculating the mean absolute error (MAE) on the validation set as the objective function return value; if the parameter combination causes VMD decomposition to fail, returning the maximum value (np.inf) to trigger the elimination mechanism. Then, an Optuna study instance is initialized, with the optimization direction set to minimize MAE, and a median pruning tool is configured to terminate poorly performing experimental branches early. By calling study.optimize(objective, n_trials=30) to perform automated parameter search, the performance of different parameter combinations is dynamically evaluated in 30 iterations of trials, and finally the optimal parameter combination that minimizes the MAE of the validation set is extracted from the trial records.

[0111] The second step is to predict the resistance value of the resistor based on all modal components, and obtain the resistance prediction result of the resistor.

[0112] For example, a random forest model can be used to predict the resistance value of a resistor. That is, all modal components are input into the random forest model to obtain the predicted resistance result. The random forest model described above is a random forest model trained using training data (such as the predicted and actual resistance values ​​of multiple sample resistors). During the training process, an error evaluation metric is used to analyze the error, expressed as:

[0113]

[0114] in, Indicates the number of sample resistors. Indicates the first The actual resistance value of each sample resistor. Indicates the first The predicted resistance value of each sample resistor.

[0115] Step 13: Perform degradation analysis on the resistor based on the resistance prediction results to obtain the degradation prediction results of the resistor.

[0116] The above degradation prediction results are used to describe the time it takes for the resistance value of a resistor to degrade to the resistance threshold.

[0117] Specifically, the earliest future time when the resistance value is less than or equal to the resistance threshold is determined from the resistance prediction results, and the earliest future time is taken as the target future time; the time corresponding to the target future time is taken as the degradation prediction result of the resistor.

[0118] For example, the resistance prediction result includes resistance values ​​at four future times: 200Ω, 170Ω, 160Ω, and 140Ω. The resistance threshold is a preset value, set to 165Ω in this example. The future time corresponding to 160Ω is the target future time. It is assumed that the resistance value of the resistor at the target future time will degrade to less than the resistance threshold, and the resistor needs to be maintained before that time.

[0119] It is worth mentioning that obtaining the resistance value data of a resistor at multiple moments eliminates the need to monitor the data of the resistor throughout its entire operation, reducing data costs. Predicting the resistance value at multiple future moments allows for analysis of the resistance value change over a period of time, enabling analysis of the long-term degradation of the resistor. Based on the prediction results of resistors with long-term resistance changes, degradation prediction is performed, effectively improving the accuracy of resistor degradation prediction.

[0120] The method of this application will be illustrated below with a specific example.

[0121] The prediction results obtained using the random forest model are as follows Figure 5As shown in the figure, the horizontal axis represents time, and the vertical axis represents resistance, with the unit being ohms (Ω). The solid line represents the actual resistance value, and the dashed line represents the predicted resistance value. The mean absolute error (MAE) is 0.0012, the root mean square error (RMSE) is 0.0019, and the mean absolute percentage error (MAPE) is 0.0033%.

[0122] The prediction results obtained using VMD and random forest models are as follows Figure 6 As shown in the figure, the horizontal axis represents time, and the vertical axis represents resistance, with the unit being ohms (Ω). The solid line represents the actual resistance value, and the dashed line represents the predicted resistance value. The mean absolute error is 0.0011, the root mean square error is 0.0016, and the mean absolute percentage error is 0.0030%.

[0123] The prediction results obtained using the method of this application are as follows: Figure 7 As shown in the figure, the horizontal axis represents time, and the vertical axis represents the resistance value in ohms (Ω). The solid line represents the actual resistance value, and the dashed line represents the predicted resistance value obtained using the method of this application. The vertical dashed line is the prediction start point (n_steps=5), indicating the start time for predicting the resistance value. The mean absolute error is 0.0010, the root mean square error is 0.0014, and the mean absolute percentage error is 0.0028%.

[0124] As can be seen, the method in this application significantly improves prediction accuracy and reduces prediction error. Compared with the single random forest model, MAE is reduced by 16.7%, RMSE by 26.3%, and MAPE by 15.2%.

[0125] The resistor degradation prediction device provided in this application is described below by way of example.

[0126] like Figure 8 As shown, this application embodiment provides a resistor degradation prediction device, the resistor degradation prediction device 800 including:

[0127] The acquisition module 801 is used to acquire the resistance value data of the resistor at T time points; the Tth time point is the current time.

[0128] The prediction module 802 is used to predict the resistance value of the resistor based on the resistance value data of the resistor at all times using an improved variational mode decomposition algorithm, and to obtain the resistance prediction result of the resistor; the resistance prediction result is used to describe the resistance value of the resistor at multiple future times after the current time.

[0129] The degradation analysis module 803 is used to perform degradation analysis on the resistor based on the resistance prediction results, and obtain the degradation prediction results of the resistor; the degradation prediction results are used to describe the time when the resistance value of the resistor degrades to the resistance threshold.

[0130] It should be noted that the information interaction and execution process between the above-mentioned devices / units are based on the same concept as the method embodiments of this application. For details on their specific functions and technical effects, please refer to the method embodiments section, and they will not be repeated here.

[0131] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0132] like Figure 9 As shown, an embodiment of this application provides a terminal device, wherein the terminal device D10 of this embodiment includes: at least one processor D100 ( Figure 9 The diagram shows only one processor, a memory D101, and a computer program D102 stored in the memory D101 and executable on the at least one processor D100, wherein the processor D100 executes the computer program D102 to implement the steps in any of the above method embodiments.

[0133] Specifically, when the processor D100 executes the computer program D102, it acquires the resistance value data of the resistor at T time points, then uses an improved variational mode decomposition algorithm to predict the resistance value based on the resistance value data of the resistor at all time points, obtaining the resistance prediction result of the resistor. Finally, it performs degradation analysis on the resistor based on the resistance prediction result, obtaining the degradation prediction result of the resistor. Acquiring the resistance value data of the resistor at multiple time points eliminates the need to monitor the data of the resistor throughout its entire operation, reducing data costs. Predicting the resistance value at multiple future time points allows for analysis of the resistance value changes over a future period, enabling analysis of the long-term degradation of the resistor. The degradation prediction based on the resistance prediction result with long-term resistance value changes effectively improves the accuracy of resistor degradation prediction.

[0134] The processor D100 can be a central processing unit (CPU), or it can be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor.

[0135] In some embodiments, the memory D101 may be an internal storage unit of the terminal device D10, such as a hard disk or memory of the terminal device D10. In other embodiments, the memory D101 may be an external storage device of the terminal device D10, such as a plug-in hard disk, smart media card (SMC), secure digital card (SD), flash card, etc., equipped on the terminal device D10. Furthermore, the memory D101 may include both internal and external storage units of the terminal device D10. The memory D101 is used to store the operating system, applications, bootloader, data, and other programs, such as the program code of the computer program. The memory D101 can also be used to temporarily store data that has been output or will be output.

[0136] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps described in the various method embodiments above.

[0137] This application provides a computer program product that, when run on a terminal device, enables the terminal device to implement the steps described in the various method embodiments above.

[0138] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of this application can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include at least: any entity or device capable of carrying the computer program code to the resistor degradation prediction method apparatus / terminal device, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electrical carrier signals or telecommunication signals.

[0139] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0140] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0141] The above description represents the preferred embodiments of this application. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles described in this application, and these improvements and modifications should also be considered within the scope of protection of this invention.

Claims

1. A method for predicting resistor degradation, characterized in that, include: Obtain the resistance value of the resistor at T time points; The Tth moment is the current moment; An improved variational mode decomposition algorithm is used to predict the resistance value of the resistor based on the resistance value data of the resistor at all times, and the resistance prediction result of the resistor is obtained; the resistance prediction result is used to describe the resistance value of the resistor at multiple future times after the current time. The resistor is subjected to degradation analysis based on the resistance prediction results to obtain the degradation prediction results of the resistor; the degradation prediction results are used to describe the time when the resistance value of the resistor degrades to the resistance threshold.

2. The resistor degradation prediction method according to claim 1, characterized in that, The improved variational mode decomposition algorithm is used to predict the resistance value of the resistor based on the resistance data of the resistor at all times, and the predicted resistance result of the resistor is obtained, including: An improved variational mode decomposition algorithm is used to perform variational mode decomposition on all resistance data of the resistor to obtain multiple mode components; The resistance value of the resistor is predicted based on all modal components, and the resistance prediction result of the resistor is obtained.

3. The resistor degradation prediction method according to claim 2, characterized in that, The improved variational mode decomposition algorithm is used to perform variational mode decomposition on all resistance data of the resistor, obtaining multiple mode components, including: The variational mode decomposition algorithm is optimized by parameters to obtain the optimized variational mode decomposition algorithm; The optimized variational mode decomposition algorithm is used to decompose all resistance data of the resistor to obtain multiple mode components.

4. The resistor degradation prediction method according to claim 3, characterized in that, The step of optimizing the variational mode decomposition algorithm to obtain the optimized variational mode decomposition algorithm includes: Obtain resistance data of multiple sample resistors at multiple historical moments and initialize multiple parameter combinations; the parameter combinations include the values ​​of multiple parameters in the variational mode decomposition algorithm; For each of the parameter combinations, the parameter combination is substituted into the variational mode decomposition algorithm, and the variational mode decomposition algorithm is used to perform variational mode decomposition on all resistance data of each sample resistor to obtain multiple initial mode components of each sample resistor. Resistance prediction is performed based on all initial mode components of each sample resistor to obtain the predicted resistance value. The mean absolute error corresponding to the parameter combination is calculated based on all predicted resistance values ​​and the actual resistance values ​​of all sample resistors. Substituting the parameter combination that minimizes the mean absolute error into the variational mode decomposition algorithm yields the optimized variational mode decomposition algorithm.

5. The resistor degradation prediction method according to claim 4, characterized in that, The variational mode decomposition algorithm is used to perform variational mode decomposition on all resistance data of each sample resistor to obtain multiple initial mode components for each sample resistor, including: For each of the sample resistors, the following steps are performed: All resistance data of the sample resistor are decomposed into multiple first mode components; Construct the objective function; With the objective function minimization as the goal, each first modal component is updated to obtain the second modal component corresponding to each first modal component; Determine whether all second-mode components satisfy the iteration stopping condition; If so, then each second mode component is used as the initial mode component; Otherwise, treat each second modal component as a first modal component, and return to the step of updating each first modal component with the objective function as the goal, to obtain the second modal component corresponding to each first modal component.

6. The resistor degradation prediction method according to claim 5, characterized in that, The objective function is: ； in, Indicates the first The first mode component Indicates the first The center frequency of the first mode component Indicates the number of the first mode components. This indicates finding the partial derivative. Represents the Dirac function, Represents the imaginary unit. Indicates time, Indicates the first The first mode component at time... The signal This represents all resistance values ​​of the sample resistor.

7. The resistor degradation prediction method according to claim 1, characterized in that, The step of performing degradation analysis on the resistor based on the resistance prediction result to obtain the degradation prediction result of the resistor includes: The earliest future time when the resistance value is less than or equal to the resistance threshold is determined from the resistance prediction results, and the earliest future time is taken as the target future time. The time corresponding to the target future moment is used as the degradation prediction result of the resistor.

8. A resistor degradation prediction device, characterized in that, include: The acquisition module is used to acquire the resistance value data of the resistor at T time points; The Tth moment is the current moment; The prediction module is used to predict the resistance value of the resistor based on the resistance value data of the resistor at all times using an improved variational mode decomposition algorithm, and to obtain the resistance prediction result of the resistor; the resistance prediction result is used to describe the resistance value of the resistor at multiple future times after the current time. The degradation analysis module is used to perform degradation analysis on the resistor based on the resistance prediction result, and obtain the degradation prediction result of the resistor; the degradation prediction result is used to describe the time when the resistance value of the resistor degrades to the resistance threshold.

9. A terminal device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the resistor degradation prediction method as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the resistor degradation prediction method as described in any one of claims 1 to 7.

Images