A crimping device and a crimping method for a lead assembly

Abstract

The application discloses a crimping device and method of lead-out wire assembly, and relates to the technical field of electronic component crimping.The device comprises a crimping demand analysis module, a quality predictor construction module, an iterative optimization module and a crimping control module.The crimping demand analysis module is used for crimping demand analysis and determination of expected crimping demand index.The quality predictor construction module is used for historical crimping monitoring log sample training set collection and construction of a crimping quality predictor.The iterative optimization module is used for crimping parameter iterative optimization by using the crimping quality predictor and output of optimal crimping parameters.The crimping control module is used for crimping control of the lead-out wire assembly.The application solves the technical problems of the prior art, such as the inability of crimping parameter setting to adapt to complex working environments, the lack of historical data support and the iterative optimization mechanism leading to unstable crimping quality, and achieves the technical effects of realizing iterative optimization and precise control of crimping parameters, thereby improving crimping quality and parameter adaptability.

Description

Technical Field

[0001] This invention relates to the field of electronic component crimping technology, specifically to a crimping device and crimping method for lead wire assemblies. Background Technology

[0002] In the crimping process of lead wire assemblies, traditional crimping methods often fail to fully consider the expected working environment characteristics of the lead wire assembly (such as static and dynamic environmental characteristics such as temperature range, average humidity, and amplitude range), and are also difficult to combine the structural properties of terminals and conductors. At the same time, they lack a mechanism to build predictive models based on historical crimping monitoring log data and to iteratively optimize crimping parameters, resulting in unreasonable crimping parameter settings. Consequently, the crimped lead wire assemblies fail to meet actual requirements in terms of electrical performance and mechanical strength.

[0003] The existing technology suffers from technical problems such as the inability of crimping parameter settings to adapt to complex working environments, lack of historical data support and iterative optimization mechanisms, leading to unstable crimping quality. Summary of the Invention

[0004] This application provides a crimping device and crimping method for lead wire assemblies, which addresses the technical problem that the crimping parameter settings in the prior art cannot adapt to complex working environments, lacks historical data support and iterative optimization mechanisms, resulting in unstable crimping quality.

[0005] In view of the above problems, this application provides a crimping device and crimping method for lead wire assemblies.

[0006] A first aspect of this application provides a crimping device for a lead wire assembly, the crimping device comprising:

[0007] The module includes a crimping requirement analysis module to analyze crimping requirements based on the expected operating environment characteristics of the lead wire assembly and determine the expected crimping requirement indicators; a quality predictor construction module to construct a crimping quality predictor based on the terminal structure attributes and conductor structure attributes of the lead wire assembly, using a training set of samples collected from historical crimping monitoring logs as constraints; an iterative optimization module to perform iterative optimization of crimping parameters within the crimping parameter space using the crimping quality predictor, with the expected crimping requirement indicators as the target, and output the optimal crimping parameters; and a crimping control module to perform crimping control on the lead wire assembly based on the optimal crimping parameters.

[0008] A second aspect of this application provides a crimping method for a lead wire assembly, the crimping method comprising:

[0009] Based on the expected operating environment characteristics of the lead wire assembly, a crimping requirement analysis is performed to determine the expected crimping requirement indicators. Using the terminal structure attributes and conductor structure attributes of the lead wire assembly as constraints, a crimping quality predictor is constructed based on a training set of samples collected from historical crimping monitoring logs. With the expected crimping requirement indicators as the target, the crimping parameters are iteratively optimized within the crimping parameter space using the crimping quality predictor to output the optimal crimping parameters. Based on the optimal crimping parameters, crimping control is performed on the lead wire assembly.

[0010] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0011] The system comprises the following modules: a crimping requirement analysis module for analyzing crimping requirements and determining expected crimping requirements; a quality predictor construction module for building a crimping quality predictor based on a training set of samples collected from historical crimping monitoring logs, constrained by the terminal and conductor structural attributes of the lead-in assembly; an iterative optimization module for iteratively optimizing crimping parameters within the crimping parameter space using the crimping quality predictor, and outputting optimal crimping parameters; and a crimping control module for controlling the crimping of the lead-in assembly based on the optimal crimping parameters. This achieves the technical effect of iterative optimization and precise control of crimping parameters, thereby improving crimping quality and parameter adaptability. Attached Figure Description

[0012] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0013] Figure 1 A schematic diagram of the structure of a crimping device for a lead wire assembly provided in an embodiment of this application;

[0014] Figure 2 This is a schematic flowchart illustrating a crimping method for a lead wire assembly provided in an embodiment of this application.

[0015] Figure labeling: 10 for crimping demand analysis module, 20 for quality predictor construction module, 30 for iterative optimization module, and 40 for crimping control module. Detailed Implementation

[0016] This application provides a crimping device and crimping method for lead wire assemblies, which addresses the technical problem in the prior art where crimping parameter settings cannot adapt to complex working environments, and the lack of historical data support and iterative optimization mechanisms leads to unstable crimping quality.

[0017] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0018] Example 1, as Figure 1 As shown, this application provides a crimping device for a lead wire assembly, the crimping device comprising:

[0019] The crimping requirement analysis module 10 is used to perform crimping requirement analysis based on the expected working environment characteristics of the lead wire assembly and to determine the expected crimping requirement indicators.

[0020] Specifically, the expected static environmental characteristics (including temperature range, average temperature, humidity range, average humidity, air pressure range, and average air pressure), expected dynamic environmental characteristics (including amplitude range and average vibration frequency), and expected service life of the lead-out components are first obtained. Then, using the expected service life as a time limit constraint and the expected static and dynamic environmental characteristics as conditional constraints, historical operation and maintenance logs of the lead-out components are retrieved to obtain multiple sample crimping requirement indicators. Finally, anomalies are eliminated and the average value is calculated for the multiple sample crimping requirement indicators to obtain expected crimping requirement indicators including electrical performance indicators and mechanical strength indicators, providing a target basis for subsequent iterative optimization of crimping parameters.

[0021] The quality predictor construction module 20 is used to construct a crimping quality predictor based on a training set of samples collected from historical crimping monitoring logs, constrained by the terminal structure attribute characteristics and conductor structure attribute characteristics of the lead wire assembly.

[0022] Specifically, using the terminal structure (such as terminal material, shape, and size) and conductor structure (such as conductor diameter, material, and number of strands) of the lead wire assembly as constraints, sample crimping parameter sets (including crimping position and crimping force) are collected from historical crimping monitoring logs of similar lead wire assemblies. Historical crimping indices (such as electrical performance and mechanical strength) of lead wire assemblies under different sample crimping parameters are obtained to form a sample crimping index set. Then, the sample crimping parameter set and sample crimping index set are used to train the feedforward neural network until the network converges, thereby constructing a crimping quality predictor that can predict crimping quality based on crimping parameters, providing predictive support for subsequent iterative optimization of crimping parameters.

[0023] The iterative optimization module 30 is used to perform iterative optimization of the crimping parameters in the crimping parameter space using the crimping quality predictor, with the expected crimping requirement index as the target, and output the optimal crimping parameters.

[0024] Specifically, a crimping parameter space containing parameters such as crimping position, crimping force, crimping time, crimping depth, and crimping speed is obtained, and an initial crimping parameter generation unit randomly generates several initial crimping parameters within this space. These initial crimping parameters are then input into a crimping quality predictor, which outputs predicted crimping indices. The deviation of each predicted index is calculated based on the expected crimping requirements, and the initial crimping parameters are sorted in ascending order of deviation to construct a sequence. This sequence is considered the initial solution, and the first K best solutions and the last J worst solutions are selected. (J is N times K and N>10), K solution sets are obtained by clustering inferior solutions based on the excellent solutions. Within each solution set, the excellent solutions are used as the adjustment direction. The step size optimization coefficient is determined by calculating the mean deviation of inferior solutions and the deviation value of the deviation from the excellent solutions. Then, the initial optimization step size is adjusted to obtain the dynamic optimization step size. The inferior solutions are adjusted to obtain the updated solution set. If the deviation of inferior solutions in the updated solution set is less than or equal to that of the excellent solutions, the excellent solutions are replaced. After iterative optimization to a preset number of times, the excellent solution of the current solution set with the smallest sum of deviations is selected as the optimal pressing parameter.

[0025] The crimping control module 40 is used to control the crimping of the lead wire assembly according to the optimal crimping parameters.

[0026] Specifically, after obtaining the optimal crimping parameters (including crimping position, crimping force, crimping time, crimping depth, and crimping speed), these parameters are used as control commands for the crimping equipment. The actuators of the crimping equipment are precisely controlled to ensure that the crimping process is strictly performed according to the optimal crimping parameters. This achieves crimping control of the lead wire assembly, ensuring that the crimped lead wire assembly meets the expected crimping requirements for electrical performance and mechanical strength, and guaranteeing the reliable operation of the lead wire assembly under the expected working environment (including static and dynamic environments such as temperature, humidity, air pressure, and vibration).

[0027] In one possible implementation, the crimping requirement analysis module 10 further includes:

[0028] The environmental feature acquisition unit is used to acquire the expected static environmental features, expected dynamic environmental features, and expected usage duration of the lead wire assembly.

[0029] The crimping requirement indicator acquisition unit is used to retrieve the operation and maintenance logs of historical lead wire components and obtain multiple sample crimping requirement indicators, with the expected usage duration as the year limit constraint and the expected static environment characteristics and expected dynamic environment characteristics as the condition constraints.

[0030] The expected crimping demand index acquisition unit is used to perform anomaly removal and mean calculation on the multiple sample crimping demand indices to obtain the expected crimping demand index.

[0031] Specifically, the environmental characteristic acquisition unit is used to acquire the expected static environmental characteristics, expected dynamic environmental characteristics, and expected service life of the lead wire assembly. The expected static environmental characteristics include at least the temperature range, average temperature, humidity range, average humidity, air pressure range, and average air pressure. The expected dynamic environmental characteristics include at least the amplitude range and average vibration frequency. The expected service life is defined in years. By acquiring these characteristics and durations, basic data support is provided for subsequent crimping requirement analysis.

[0032] The crimping requirement indicator acquisition unit uses the expected usage time as a time-dimension constraint, and combines the expected static environmental characteristics (such as temperature range, average humidity, etc.) and expected dynamic environmental characteristics (such as amplitude range, average vibration frequency, etc.) to filter records from historical operation and maintenance logs that match the expected working environment of the current lead wire component, and then extracts multiple sample crimping requirement indicators. These indicators include at least electrical performance indicators and mechanical strength indicators, providing data support for subsequent determination of expected crimping requirement indicators through average calculation.

[0033] The expected crimping requirement indicator acquisition unit first performs outlier detection on multiple sample crimping requirement indicators (including electrical performance indicators, mechanical strength indicators, etc.) retrieved from historical operation and maintenance logs. Data that deviates significantly from the normal range is removed using statistical methods or preset thresholds to eliminate the influence of abnormal operating conditions or recording errors. Then, the arithmetic mean or weighted mean of the remaining valid sample crimping requirement indicators is calculated to finally obtain the expected crimping requirement indicators that meet the expected working environment characteristics (static environment, dynamic environment, and usage time constraints) of the lead wire component, providing a quantitative target for subsequent iterative optimization of crimping parameters.

[0034] In one possible implementation, the environmental feature acquisition unit further includes:

[0035] The expected static environmental characteristics include at least a temperature range, a temperature average, a humidity range, a humidity average, an air pressure range, and an air pressure average; the expected dynamic environmental characteristics include at least an amplitude range and a vibration frequency average.

[0036] Specifically, the expected static environmental characteristics include at least a temperature range, average temperature, humidity range, average humidity, air pressure range, and average air pressure; the expected dynamic environmental characteristics include at least an amplitude range and an average vibration frequency. The expected static environmental characteristics characterize the relatively stable parameters of the environment in which the lead-in assembly is located, such as the temperature fluctuation range and average temperature value within a certain range, the humidity variation range and average humidity value, and the air pressure fluctuation range and average air pressure value. The expected dynamic environmental characteristics describe parameters in the environment that change periodically, including the amplitude range of object vibration and the average vibration frequency. These characteristic parameters together constitute the key elements of the expected working environment of the lead-in assembly, providing specific environmental constraints for subsequent crimping requirement analysis.

[0037] In one possible implementation, the expected crimping demand indicator acquisition unit further includes:

[0038] The crimping specifications should include at least the electrical performance and mechanical strength specifications of the lead wire assembly.

[0039] Specifically, crimping parameters include at least the electrical performance parameters and mechanical strength parameters of the lead assembly. The electrical performance parameters are used to measure the conductivity and insulation performance of the lead assembly after crimping, such as contact resistance and insulation resistance, to ensure that the current can be transmitted stably and without the risk of leakage. The mechanical strength parameters are used to assess the firmness of the connection at the crimping point, such as tensile strength and shear strength, to ensure that the lead assembly is not easily detached or damaged when subjected to external forces. These two types of parameters together constitute the core parameters for measuring crimping quality.

[0040] In one possible implementation, the quality predictor building module 20 further includes:

[0041] The sample crimping index set acquisition unit is used to collect sample crimping parameter sets based on the historical crimping monitoring logs of similar lead wire assemblies, constrained by the terminal structure attribute characteristics and conductor structure attribute characteristics of the lead wire assembly, and to obtain the historical crimping indexes of the lead wire assembly under different sample crimping parameters, thus obtaining the sample crimping index set.

[0042] The crimping quality predictor acquisition unit is used to train a feedforward neural network until convergence using the sample crimping parameter set and the sample crimping index set to obtain a crimping quality predictor.

[0043] Specifically, the sample crimping index set acquisition unit uses terminal structural attribute characteristics (such as terminal material, shape, and size) and conductor structural attribute characteristics (such as conductor diameter, material, and number of strands) as screening conditions to extract crimping parameters that meet the constraints from historical crimping monitoring logs of similar lead wire assemblies, forming a sample crimping parameter set that includes parameters such as crimping position, crimping force, crimping time, crimping depth, and crimping speed. At the same time, it acquires historical crimping indices of lead wire assemblies under different sample crimping parameters, including electrical performance indices (such as contact resistance and insulation resistance) and mechanical strength indices (such as tensile strength and shear strength), and finally integrates them to form a sample crimping index set, providing training data support for building a crimping quality predictor.

[0044] The crimping quality predictor acquisition unit uses the sample crimping parameter set (covering parameters such as crimping position and crimping force) collected by the sample crimping index set acquisition unit as input layer data, and the corresponding historical crimping indexes (including electrical performance, mechanical strength, etc.) as output layer target values ​​to construct the training dataset of the feedforward neural network. Through the backpropagation algorithm, the weights and biases of each layer of the network are continuously adjusted so that the error between the crimping quality index predicted by the network and the actual historical index is gradually reduced until the training error reaches a preset threshold or the number of iterations reaches the upper limit. Finally, a crimping quality predictor that can accurately predict crimping quality based on the input crimping parameters is obtained, providing a quantitative prediction model for subsequent iterative optimization of crimping parameters.

[0045] In one possible implementation, the iterative optimization module 30 further includes:

[0046] The crimping parameter space acquisition unit is used to acquire the crimping parameter space of the crimping equipment, wherein the crimping parameters include crimping position, crimping force, crimping time, crimping depth and crimping speed.

[0047] An initial crimping parameter generation unit is used to randomly generate a number of initial crimping parameters within the crimping parameter space.

[0048] The optimal crimping parameter output unit is used to target the expected crimping requirements and, using the crimping quality predictor, iteratively optimize the crimping parameters based on the several initial crimping parameters to output the optimal crimping parameters.

[0049] Specifically, the crimping parameter space acquisition unit is used to acquire the crimping parameter space of the crimping equipment, where the crimping parameters include crimping position, crimping force, crimping time, crimping depth, and crimping speed. By analyzing the performance parameters of the crimping equipment, the value range, accuracy requirements, and constraints between the above-mentioned crimping parameters are determined, thereby constructing a complete crimping parameter space. This provides feasible parameter value ranges for subsequent generation of initial crimping parameters and iterative optimization of the crimping parameters.

[0050] The initial pressing parameter generation unit, based on the value range of parameters such as pressing position, pressing force, pressing time, pressing depth, and pressing speed determined by the pressing parameter space acquisition unit, independently generates parameter values ​​within the feasible domain of each parameter through a random algorithm, and combines them to form several sets of initial pressing parameters. These initial parameters can uniformly cover different regions of the pressing parameter space, providing diverse initial solutions for subsequent iterative optimization of pressing parameters, and ensuring that the optimization process can explore the optimal combination of pressing parameters from multiple dimensions.

[0051] The optimal crimping parameter output unit first inputs several initial crimping parameters into the crimping quality predictor to obtain the corresponding predicted crimping indices. Then, based on the expected crimping requirement index, it calculates the deviation of each predicted index and constructs an initial crimping parameter sequence by sorting the deviations from smallest to largest. The sequence is regarded as the initial solution. The first K best solutions and the last J worst solutions (J is N times K and N>10) are selected. The worst solutions are clustered based on the best solutions to obtain K solution sets. Within each solution set, the step size optimization coefficient is determined by calculating the mean deviation of the worst solutions and the deviation from the best solutions. The initial optimization step size is then adjusted to obtain the dynamic optimization step size. The worst solutions are adjusted with the best solutions as the adjustment direction to obtain the updated solution set. If the deviation of a worst solution in the updated solution set is less than or equal to that of a best solution, the best solution is replaced. After iterative optimization to a preset number of times, the best solution in the current solution set with the smallest sum of deviations is selected as the optimal crimping parameter to ensure that the crimping parameters meet the expected crimping requirement index.

[0052] In one possible implementation, the optimal crimping parameter output unit further includes:

[0053] The predictive crimping index output unit is used to input the several initial crimping parameters into the crimping quality predictor and output several predicted crimping indices.

[0054] The deviation calculation unit is used to calculate the deviation of the predicted crimping indicators based on the expected crimping demand indicators, and obtain several deviations.

[0055] The parameter iteration optimization unit is used to iteratively optimize the crimping parameters based on the aforementioned deviations, and outputs the crimping parameters corresponding to the minimum deviation as the optimal crimping parameters.

[0056] Specifically, the predicted crimping index output unit takes several initial crimping parameters (including combinations of parameters such as crimping position, crimping force, crimping time, crimping depth, and crimping speed) randomly generated by the initial crimping parameter generation unit in the crimping parameter space, and inputs them sequentially into the crimping quality predictor obtained by training and converging a feedforward neural network. Through the calculation of the mapping relationship between crimping parameters and crimping quality indices by the predictor, the predicted crimping index corresponding to each initial crimping parameter is output, including the electrical performance index (such as contact resistance, insulation resistance, etc.) and mechanical strength index (such as tensile strength, shear strength, etc.) of the lead wire assembly, providing a quantitative basis for subsequent deviation calculation and iterative optimization of crimping parameters.

[0057] When implementing the deviation calculation unit, the first step is to obtain the determined expected crimping requirements, including electrical performance indicators (such as expected values ​​of contact resistance and insulation resistance) and mechanical strength indicators (such as expected values ​​of tensile strength and shear strength). Then, the predicted crimping index corresponding to each initial crimping parameter is obtained. For each predicted crimping index, the deviation is calculated by determining its absolute deviation (e.g., |predicted value - expected value|) or relative deviation (e.g., |predicted value - expected value| / expected value × 100%) from the corresponding expected crimping requirement. For example, if the expected contact resistance is 50mΩ and a predicted contact resistance is 55mΩ, the relative deviation is (55-50) / 50 × 100% = 10%. This process is repeated for all predicted crimping indices, providing quantified deviation data for subsequent parameter iteration and optimization.

[0058] When the parameter iterative optimization unit performs iterative optimization of the pressing parameters based on several deviations and outputs the optimal pressing parameters, the specific implementation method is as follows: First, the parameter sorting unit sorts several initial pressing parameters in ascending order of deviation, constructing an initial pressing parameter sequence. Next, the initial solution acquisition unit treats this sequence as the initial solution, selecting the first K solutions as excellent solutions and the last J solutions as inferior solutions, where J is N times K and N is greater than 10. Then, the solution set acquisition unit clusters the J inferior solutions based on the K excellent solutions to obtain K solution sets, where the smaller the deviation of the excellent solutions, the more inferior solutions are in the corresponding solution set. Subsequently, in the updated solution set acquisition unit, the mean deviation of multiple inferior solutions within each solution set is first calculated by the mean deviation acquisition unit, along with multiple deviation values ​​between the inferior and superior solutions. Then, the step size optimization coefficient acquisition unit sets the ratio of the deviation value to the mean deviation as the step size optimization coefficient. The initial optimization step size adjustment unit then adjusts the initial optimization step size based on these multiple step size optimization coefficients, obtaining the dynamic optimization step size for each solution set. Following this, using the superior solution as the adjustment direction, the inferior solutions within each solution set are adjusted according to the dynamic optimization step size, resulting in K updated solution sets. If, within the same updated solution set, the deviation of an inferior solution is less than or equal to the deviation of a superior solution, the inferior solution replacement unit replaces the superior solution with the inferior solution. Finally, the current solution set output unit performs iterative optimization until a preset number of iterations is reached, outputting K current solution sets. The current solution set with the smallest sum of deviations is selected as the optimal solution set, and the superior solutions within it are set as the optimal compression parameters.

[0059] In one possible implementation, the parameter iterative optimization unit further includes:

[0060] The parameter sorting unit is used to sort the initial crimping parameters according to the deviation degree from small to large, and construct the initial crimping parameter sequence.

[0061] The initial solution acquisition unit is used to regard the initial pressing parameter sequence as the initial solution, and select the first K solutions of the initial solution sequence as the optimal solution and the last J solutions as the inferior solution, where J is N times K and N is greater than 10.

[0062] The solution set acquisition unit is used to cluster J inferior solutions based on K superior solutions to obtain K solution sets. The smaller the deviation of the superior solutions, the more inferior solutions are in the solution set.

[0063] The updated solution set acquisition unit is used to adjust the inferior solutions in each solution set according to the dynamic optimization step size, with the optimal solution as the adjustment direction, within the K solution sets, to obtain K updated solution sets.

[0064] The inferior solution replacement unit is used to identify K updated solution sets. If the deviation of an inferior solution is less than or equal to the deviation of a superior solution within the same updated solution set, then the inferior solution replaces the superior solution.

[0065] The current solution set output unit is used to perform iterative optimization until a preset number of times is reached, output K current solution sets, and select the current solution set with the smallest sum of deviations as the optimal solution set, and set the optimal solution set as the optimal pressing parameters.

[0066] Specifically, the parameter sorting unit uses several deviations output by the deviation calculation unit as a basis to arrange the corresponding initial crimping parameters in ascending order of deviation value, forming an ordered sequence of initial crimping parameters. This results in a smaller deviation between the crimping quality prediction results of the preceding parameters and the expected crimping requirements, while the deviation of the following parameters is larger, thus providing an ordered set of parameters as a basis for subsequent initial solution acquisition and iterative optimization of crimping parameters.

[0067] The initial solution acquisition unit uses the initial pressing parameter sequence constructed by the parameter sorting unit as the initial solution set. According to the order of deviation from smallest to largest, the first K pressing parameters with smaller deviations in the sequence are determined as excellent solutions, and the last J pressing parameters with larger deviations are determined as inferior solutions. In this way, the initial solutions are divided into two groups with different qualities, which provides a basis for subsequent clustering and optimization adjustment of inferior solutions based on excellent solutions. The multiple relationship between J and K (J is N times K and N>10) ensures that the number of inferior solutions is much greater than that of excellent solutions, so that the parameter space can be explored more fully during the iterative optimization process.

[0068] The solution set acquisition unit uses the K optimal solutions output by the initial solution acquisition unit as cluster centers, with each optimal solution corresponding to a cluster. Then, it calculates the deviation distance (e.g., Euclidean or Manhattan distance) between each inferior solution and the K optimal solutions, assigning each inferior solution to the cluster corresponding to the optimal solution with the smallest deviation distance, thus forming K solution sets. During clustering, since the deviation of an optimal solution reflects the degree of matching between its pressing quality and the expected requirements, a smaller deviation indicates that the pressing parameters are closer to the optimal solution. Therefore, it attracts more inferior solutions with similar deviations to cluster in that cluster, resulting in a larger number of inferior solutions in that solution set than in the solution set corresponding to the optimal solution with larger deviations. In this way, clustering of inferior solutions based on optimal solutions is achieved, providing structured parameter grouping for subsequent iterative optimization of pressing parameters.

[0069] When the solution set acquisition unit adjusts inferior solutions within K solution sets with the optimal solution as the adjustment direction and according to the dynamic optimization step size, the specific implementation is as follows: First, calculate the average deviation of inferior solutions within each solution set, as well as the deviation value between each inferior solution and its corresponding optimal solution; then, use the ratio of the deviation value to the average deviation value as the step size optimization coefficient, and calculate multiple step size optimization coefficients using multiple deviation values; then, the initial optimization step size adjustment unit adjusts the initial optimization step size according to these step size optimization coefficients to obtain the dynamic optimization step size for each solution set; finally, within each solution set, with the pressing parameters of the optimal solution as the target direction, iteratively adjust the parameters of the inferior solutions (such as pressing force, pressing depth, etc.) according to the dynamic optimization step size, thereby obtaining K updated solution sets and realizing the optimization adjustment of inferior solutions.

[0070] In its implementation, the inferior solution replacement unit first iterates through K updated solution sets. For each set, it compares the deviation of inferior solutions from their corresponding superior solutions. For each inferior solution within the same updated solution set, it obtains its deviation value and compares it with the deviation of the corresponding superior solution. If the deviation of an inferior solution is less than or equal to that of a superior solution, it indicates that the compression parameters corresponding to the inferior solution are closer to the expected compression requirements than the current superior solution. In this case, the inferior solution is replaced with a new superior solution, while retaining the parameter space position of the original superior solution as a reference. In this way, superior solutions are dynamically updated in each updated solution set, ensuring that the set of superior solutions always contains better compression parameters in the current iteration. This provides a better initial solution set for subsequent iterations, gradually approaching the optimal compression parameters.

[0071] In its implementation, the current solution set output unit first sets a preset number of iterations and then initiates an iterative optimization process. In each iteration, the solution set is updated and optimized using the solution set update acquisition unit and the inferior solution replacement unit. When the preset number of iterations is reached, the iteration process stops, and the K current solution sets are output. Next, the sum of the deviations of all solutions within each current solution set is calculated. By comparing these sums of deviations, the current solution set with the smallest sum of deviations is selected as the optimal solution set. Finally, optimal solutions are extracted from the optimal solution set and set as the optimal pressing parameters, completing the entire pressing parameter optimization process.

[0072] In one possible implementation, the updated solution set acquisition unit further includes:

[0073] The deviation mean acquisition unit is used to calculate the deviation mean of multiple inferior solutions and multiple deviation values ​​of the deviation between multiple inferior solutions and the superior solutions in each solution set.

[0074] The step size optimization coefficient acquisition unit is used to set the ratio of the deviation value to the mean deviation value as the step size optimization coefficient, and to calculate multiple step size optimization coefficients based on multiple deviation values.

[0075] The initial optimization step size adjustment unit is used to adjust the initial optimization step size according to the multiple step size optimization coefficients to obtain the dynamic optimization step size of each solution set.

[0076] Specifically, the deviation mean acquisition unit first traverses K solution sets. For each solution set, it collects the deviation data of all inferior solutions and calculates the mean deviation of inferior solutions in the solution set using the arithmetic mean method. At the same time, it subtracts the deviation of each inferior solution from the deviation of the corresponding superior solution to obtain the deviation deviation value between each inferior solution and the superior solution, forming multiple deviation value sets for the solution set, which provides a data basis for the subsequent calculation of the step size optimization coefficient.

[0077] The step-size optimization coefficient acquisition unit first obtains the deviation values ​​between multiple inferior solutions and superior solutions for each solution set, as well as the average deviation value of the inferior solutions in the solution set. Then, it divides each deviation value by the corresponding average deviation value to obtain the step-size optimization coefficient for each deviation value. For example, if the average deviation value of inferior solutions in a solution set is M, and the deviation value between one inferior solution and a superior solution is D, then the corresponding step-size optimization coefficient is D / M. In this way, multiple deviation values ​​are calculated one by one to obtain multiple step-size optimization coefficients for the solution set. These coefficients reflect the differences in the degree of deviation between different inferior solutions and superior solutions, providing a quantitative basis for subsequent dynamic adjustment of the optimization step-size.

[0078] The initial optimization step size adjustment unit first obtains multiple step size optimization coefficients for each solution set. These coefficients are calculated as the ratio of the deviation value to the mean deviation. Then, an adjustment factor is determined by aggregating these step size optimization coefficients (e.g., calculating the average or weighted average). The deviation value reflects the degree of difference in deviation between inferior and superior solutions, while the mean deviation reflects the overall deviation level of inferior solutions within the solution set. Finally, the initial optimization step size is multiplied by this adjustment factor to obtain the dynamic optimization step size for each solution set. This allows the step size to adaptively adjust according to the distribution characteristics of inferior solutions within the solution set, thereby more accurately guiding inferior solutions towards iterative optimization.

[0079] Example 2, based on the same inventive concept as the crimping device for a lead wire assembly in the foregoing examples, such as... Figure 2 As shown, this application provides a crimping method for a lead wire assembly. The method and apparatus embodiments in this application are based on the same inventive concept. The method includes:

[0080] Step S100: Analyze the crimping requirements based on the expected working environment characteristics of the lead wire assembly and determine the expected crimping requirements indicators.

[0081] Step S200: Using the terminal structure attribute characteristics and conductor structure attribute characteristics of the lead wire assembly as constraints, and based on the sample training set collected from historical crimping monitoring logs, construct a crimping quality predictor.

[0082] Step S300: Taking the expected crimping requirement index as the target, the crimping quality predictor is used to iteratively optimize the crimping parameters in the crimping parameter space and output the optimal crimping parameters.

[0083] Step S400: Perform crimping control on the lead wire assembly according to the optimal crimping parameters.

[0084] Furthermore, the method is also used to achieve the following functions:

[0085] Obtain the expected static environment characteristics, expected dynamic environment characteristics, and expected usage duration of the lead wire component; using the expected usage duration as a yearly constraint and the expected static and dynamic environment characteristics as conditional constraints, retrieve the historical operation and maintenance logs of the lead wire component to obtain multiple sample crimping requirement indicators; perform anomaly removal and mean calculation on the multiple sample crimping requirement indicators to obtain the expected crimping requirement indicator.

[0086] Furthermore, the method is also used to achieve the following functions:

[0087] The expected static environmental characteristics include at least a temperature range, a temperature average, a humidity range, a humidity average, an air pressure range, and an air pressure average; the expected dynamic environmental characteristics include at least an amplitude range and a vibration frequency average.

[0088] Furthermore, the method is also used to achieve the following functions:

[0089] The crimping specifications should include at least the electrical performance and mechanical strength specifications of the lead wire assembly.

[0090] Furthermore, the method is also used to achieve the following functions:

[0091] Constrained by the terminal structure attributes and conductor structure attributes of the lead wire assembly, a sample crimping parameter set is collected based on the historical crimping monitoring logs of similar lead wire assemblies, and historical crimping indices of lead wire assemblies under different sample crimping parameters are obtained to obtain a sample crimping index set; using the sample crimping parameter set and sample crimping index set, a feedforward neural network is trained until convergence to obtain a crimping quality predictor.

[0092] Furthermore, the method is also used to achieve the following functions:

[0093] Obtain the crimping parameter space of the crimping equipment, wherein the crimping parameters include crimping position, crimping force, crimping time, crimping depth, and crimping speed; randomly generate several initial crimping parameters within the crimping parameter space; using the expected crimping requirement index as the target, utilize the crimping quality predictor to iteratively optimize the crimping parameters based on the several initial crimping parameters, and output the optimal crimping parameters.

[0094] Furthermore, the method is also used to achieve the following functions:

[0095] The initial crimping parameters are input into the crimping quality predictor, which outputs several predicted crimping indices. Based on the expected crimping demand index, the deviation of each of the predicted crimping indices is calculated to obtain several deviations. Based on the several deviations, the crimping parameters are iteratively optimized, and the crimping parameter corresponding to the minimum deviation is set as the optimal crimping parameter.

[0096] Furthermore, the method is also used to achieve the following functions:

[0097] The initial pressing parameters are sorted according to their deviation from smallest to largest to construct an initial pressing parameter sequence. This initial pressing parameter sequence is considered the initial solution, and the first K solutions are selected as optimal solutions, and the last J solutions as inferior solutions, where J is N times K, and N is greater than 10. The J inferior solutions are clustered based on the K optimal solutions to obtain K solution sets, where the smaller the deviation of the optimal solutions, the more inferior solutions exist within each solution set. Within the K solution sets, the inferior solutions are adjusted according to the dynamic optimization step size, using the optimal solutions as the adjustment direction, to obtain K updated solution sets. The K updated solution sets are identified; if the deviation of an inferior solution is less than or equal to the deviation of an optimal solution within the same updated solution set, the inferior solution replaces the optimal solution. Iterative optimization is performed until a preset number of iterations is reached, outputting K current solution sets. The current solution set with the smallest sum of deviations is selected as the optimal solution set, and the optimal solutions of the optimal solution set are set as the optimal pressing parameters.

[0098] Furthermore, the method is also used to achieve the following functions:

[0099] Within each solution set, the mean deviation of multiple inferior solutions and multiple deviation values ​​of the deviation between multiple inferior solutions and the superior solutions are calculated respectively; the ratio of the deviation value to the mean deviation is set as the step size optimization coefficient, and multiple step size optimization coefficients are calculated based on the multiple deviation values; the initial optimization step size is adjusted according to the multiple step size optimization coefficients to obtain the dynamic optimization step size of each solution set.

[0100] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0101] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0102] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and variations fall within the scope of this application and its equivalents, this application intends to include such modifications and variations.

Claims

1. A crimping device for a lead wire assembly, characterized in that, The crimping device includes: The crimping requirement analysis module is used to analyze the crimping requirements based on the expected working environment characteristics of the lead wire assembly and determine the expected crimping requirement indicators. The quality predictor building module is used to build a crimping quality predictor based on a sample training set collected from historical crimping monitoring logs, constrained by the terminal structure attribute characteristics and conductor structure attribute characteristics of the lead wire assembly. The iterative optimization module is used to perform iterative optimization of the crimping parameters in the crimping parameter space using the crimping quality predictor, with the expected crimping requirement index as the target, and output the optimal crimping parameters. A crimping control module is used to control the crimping of the lead wire assembly according to the optimal crimping parameters. The crimping requirement analysis module also includes: The environmental feature acquisition unit is used to acquire the expected static environmental features, expected dynamic environmental features, and expected usage duration of the lead wire assembly. The crimping requirement indicator acquisition unit is used to retrieve the operation and maintenance logs of historical lead wire components and obtain multiple sample crimping requirement indicators, with the expected usage duration as the year limit constraint and the expected static environment characteristics and expected dynamic environment characteristics as the condition constraints. The expected crimping demand index acquisition unit is used to perform anomaly removal and mean calculation on the multiple sample crimping demand indices to obtain the expected crimping demand index. The quality predictor construction module also includes: The sample crimping index set acquisition unit is used to collect sample crimping parameter sets based on the historical crimping monitoring logs of similar lead wire assemblies, constrained by the terminal structure attribute characteristics and conductor structure attribute characteristics of the lead wire assembly, and to obtain the historical crimping indexes of the lead wire assembly under different sample crimping parameters, thus obtaining the sample crimping index set. The crimping quality predictor acquisition unit is used to train a feedforward neural network until convergence using the sample crimping parameter set and the sample crimping index set to obtain a crimping quality predictor.

2. The crimping device for a lead wire assembly according to claim 1, characterized in that, The environmental feature acquisition unit further includes: The expected static environmental characteristics include at least a temperature range, a temperature average, a humidity range, a humidity average, an air pressure range, and an air pressure average; the expected dynamic environmental characteristics include at least an amplitude range and a vibration frequency average.

3. The crimping device for a lead wire assembly according to claim 1, characterized in that, The expected crimping demand indicator acquisition unit also includes: The crimping specifications should include at least the electrical performance and mechanical strength specifications of the lead wire assembly.

4. The crimping device for a lead wire assembly according to claim 1, characterized in that, The iterative optimization module also includes: The crimping parameter space acquisition unit is used to acquire the crimping parameter space of the crimping equipment, wherein the crimping parameters include crimping position, crimping force, crimping time, crimping depth and crimping speed; An initial crimping parameter generation unit is used to randomly generate a number of initial crimping parameters within the crimping parameter space; The optimal crimping parameter output unit is used to target the expected crimping requirements and, using the crimping quality predictor, iteratively optimize the crimping parameters based on the several initial crimping parameters to output the optimal crimping parameters.

5. The crimping device for a lead wire assembly according to claim 4, characterized in that, The optimal crimping parameter output unit also includes: The predictive crimping index output unit is used to input the several initial crimping parameters into the crimping quality predictor and output several predicted crimping indices. The deviation calculation unit is used to calculate the deviation of the several predicted crimping indicators based on the expected crimping demand indicators, and obtain several deviations. The parameter iteration optimization unit is used to iteratively optimize the crimping parameters based on the aforementioned deviations, and outputs the crimping parameters corresponding to the minimum deviation as the optimal crimping parameters.

6. The crimping device for a lead wire assembly according to claim 5, characterized in that, The parameter iteration optimization unit further includes: The parameter sorting unit is used to sort the initial crimping parameters according to the deviation degree from small to large, and construct the initial crimping parameter sequence based on the deviation degree. The initial solution acquisition unit is used to regard the initial pressing parameter sequence as the initial solution, and select the first K solutions of the initial solution sequence as the optimal solution and the last J solutions as the inferior solution, where J is N times K and N is greater than 10; The solution set acquisition unit is used to cluster J inferior solutions based on K superior solutions to obtain K solution sets. The smaller the deviation of the superior solutions, the more inferior solutions are in the solution set. The solution set acquisition unit is used to adjust the inferior solutions in each solution set according to the dynamic optimization step size, with the optimal solution as the adjustment direction, within the K solution sets to obtain K updated solution sets; The inferior solution replacement unit is used to identify K updated solution sets. If the deviation of the inferior solution is less than or equal to the deviation of the superior solution within the same updated solution set, the inferior solution is used to replace the superior solution. The current solution set output unit is used to perform iterative optimization until a preset number of times is reached, output K current solution sets, and select the current solution set with the smallest sum of deviations as the optimal solution set, and set the optimal solution set as the optimal pressing parameters.

7. The crimping device for a lead wire assembly according to claim 6, characterized in that, The updated solution set acquisition unit further includes: The deviation mean acquisition unit is used to calculate the deviation mean of multiple inferior solutions and multiple deviation values ​​of the deviation between multiple inferior solutions and the superior solutions in each solution set. The step size optimization coefficient acquisition unit is used to set the ratio of the deviation value to the mean deviation as the step size optimization coefficient, and to calculate multiple step size optimization coefficients based on multiple deviation values. The initial optimization step size adjustment unit is used to adjust the initial optimization step size according to the multiple step size optimization coefficients to obtain the dynamic optimization step size of each solution set.

8. A crimping method for a lead wire assembly, characterized in that, The crimping method is implemented using a crimping device for a lead wire assembly as described in any one of claims 1-7, and the crimping method includes: Based on the expected operating environment characteristics of the lead wire assembly, the crimping requirements are analyzed to determine the expected crimping requirements indicators. Using the terminal structure attributes and conductor structure attributes of the lead wire assembly as constraints, and based on the sample training set collected from historical crimping monitoring logs, a crimping quality predictor is constructed. With the expected crimping requirements as the target, the crimping quality predictor is used to iteratively optimize the crimping parameters in the crimping parameter space and output the optimal crimping parameters. The crimping of the lead wire assembly is controlled according to the optimal crimping parameters.

Images