A processing optimization method and system for scene-adaptive connector

Abstract

The present invention discloses a method and system for optimizing the processing of a scene-adaptive connector, and relates to the field of data processing technology. The method includes: obtaining application scene information of a target connector; obtaining a set of index bias weights of a preset quality index set; traversing the connector sequence for quality inspection to generate multiple sets of quality index values; generating L quality index value sequences; dividing the L quality index value sequences into water-immersed sequences to generate L sets of quality index value subsequences; generating L quality index means; generating target optimized processing parameters; and controlling processing equipment to produce and process the target connector. The method solves the technical problems of existing connector processing optimization, which rely on fixed process flows and processing parameters, are difficult to flexibly respond to complex and changeable application scenarios, and thus lead to poor quality and performance of the produced connectors and low production efficiency, and achieves the technical effect of improving the processing quality, performance and production efficiency of the connectors.

Description

Technical Field

[0001] The present application relates to fields related to data processing technology, and specifically to a processing optimization method and system for a scene-adaptive connector. Background Art

[0002] In the rapidly developing modern industrial system, connectors are an indispensable part of electronic equipment, and their quality and performance are directly related to the stability and reliability of the entire system. With the continuous advancement of science and technology and the emergence of diversified application scenarios, traditional connector processing methods can no longer meet complex and changing needs. Connectors play a vital role in automotive electronics, power automation, communication equipment, and electric transportation. In different application scenarios, the environmental conditions, signal transmission requirements, and mechanical properties of connectors are different, which poses severe challenges to the processing quality and production efficiency of connectors. However, traditional connector processing often relies on fixed process flows and processing parameters, and it is difficult to flexibly respond to changing application requirements, resulting in the connectors produced being difficult to achieve optimal quality, performance, and cost. It is impossible to take into account the quality requirements in all application scenarios, which causes connectors to be prone to failure in actual use, affecting user experience and equipment performance.

[0003] Therefore, at this stage, the technologies related to connector processing optimization rely on fixed process flows and processing parameters and are difficult to flexibly respond to complex and changing application scenarios, which in turn leads to technical problems such as poor quality and performance of the produced connectors and low production efficiency. Summary of the invention

[0004] The present application provides a processing optimization method and system for a scenario-adaptive connector, thereby solving the technical problems existing in the existing connector processing optimization, which rely on fixed process flow and processing parameters and are difficult to flexibly respond to complex and changeable application scenarios, thus resulting in poor quality and performance of the produced connectors and low production efficiency. The application achieves the technical effect of improving the processing quality, performance and production efficiency of the connector.

[0005] The present application provides a method for optimizing processing of a scene-adaptive connector, the method comprising: obtaining application scene information of a target connector; reconstructing a quality indicator bias weight based on the application scene information to obtain an indicator bias weight set of a preset quality indicator set; collecting a connector sequence processed on a production line within a preset processing window, traversing the connector sequence for quality inspection, and generating a plurality of quality indicator value sets; using the preset quality indicator set as an index, performing single indicator serialization processing on the plurality of quality indicator value sets in combination with the connector sequence to generate L quality indicator value sequences, wherein the preset quality indicator set includes L quality indicators; traversing the L quality indicator value sequences to perform a maximum quality indicator inspection. The method comprises the following steps: extracting the value of the L quality indicator, obtaining L maximum values ​​of the quality indicators, and dividing the L quality indicator value sequences into water-immersed sequences based on the L maximum values ​​of the quality indicators to generate L sets of quality indicator value subsequences, wherein the L sets of quality indicator value subsequences correspond to L sets of sub-processing windows; performing asynchronous mean identification of the same quality indicators on the L sets of quality indicator value subsequences to generate L quality indicator means; performing processing parameter pointing optimization on the production line based on the L quality indicator means and the indicator bias weight set to generate target optimized processing parameters; transmitting the target optimized processing parameters to the parameter control unit of the production line to control the processing equipment to perform production processing of the target connector.

[0006] In a possible implementation, the quality indicator bias weights are reconstructed based on the application scenario information to obtain an indicator bias weight set of a preset quality indicator set, and the following processing is performed: extracting a scenario feature set of the application scenario information, wherein the scenario feature set includes jitter capability constraints, connection mode complexity, and transmission rate constraints; matching an initial indicator bias weight set based on the scenario feature set; performing fault frequent item mining with the application scenario information as an index to obtain multiple connector fault records; performing same-indicator aggregation on the multiple connector fault records to generate a scenario indicator fault degree set; reconstructing the initial indicator bias weight set based on the scenario indicator fault degree set to generate the indicator bias weight set.

[0007] In a possible implementation, the L quality indicator value sequences are traversed to extract the maximum quality indicator value, and L quality indicator maximum values ​​are obtained. Based on the L quality indicator maximum values, the L quality indicator value sequences are divided into water-immersion sequences to generate L quality indicator value subsequence sets, and the following processing is performed: the L quality indicator maximum values ​​are respectively used as the starting points for water-immersion sequence division, and quality indicator value searches are performed on both sides of the L quality indicator value sequences according to a preset division threshold to obtain L first left-direction quality indicator values ​​and L first right-direction quality indicator values; it is determined whether the indicator value difference between the L quality indicator maximum values ​​and the L first left-direction quality indicator values ​​is greater than or equal to the preset division threshold. If so, the quality indicator value search in the left direction is stopped; if not, the L quality indicator maximum values ​​are used as the water-immersion sequence division starting points. water point, under the preset water injection amount, water will spread from the water point to the L first left-direction quality index values, and the L first left-direction quality index values ​​are added into the L first quality index value subsequences of the L quality index maximum values; when the difference between the L quality index maximum values ​​and the L first right-direction quality index values ​​is less than the preset division threshold, the L quality index maximum values ​​are taken as the water point, under the preset water injection amount, water will spread from the water point to the L first right-direction quality index values, and the L first right-direction quality index values ​​are added into the L first quality index value subsequences; when the quality index value search is stopped in both the left and right directions, the L first quality index value subsequences that have been searched are obtained; the water sequence is divided on both sides of the L first quality index value subsequences of the L quality index value sequence, and a set of L quality index value subsequences is generated.

[0008] In a possible implementation, a processing optimization method for a scene adaptive connector further performs the following processing: respectively extracting the time nodes of the left and right endpoints of each quality indicator value subsequence in the L quality indicator value subsequence sets to obtain L sub-processing window sets.

[0009] In a possible implementation, the L quality indicator value subsequence sets are respectively subjected to same-quality indicator asynchronous mean identification to generate L quality indicator means, and the following processing is also performed: the indicator value mean is calculated for each quality indicator value subsequence in the L quality indicator value subsequence sets to obtain L subsequence indicator mean sets; the indicator value fluctuation variance is calculated for each quality indicator value subsequence in the L quality indicator value subsequence sets to obtain L subsequence indicator fluctuation variance sets; the L subsequence indicator mean sets and the L subsequence indicator fluctuation variance sets are input into the indicator mean calculation formula for analysis to obtain L quality indicator means.

[0010] In a possible implementation, the L subsequence indicator mean value set and the L subsequence indicator fluctuation variance set are input into the indicator mean calculation formula for analysis, and the following processing is also performed: the indicator mean calculation formula is:

[0011] ;

[0012] in, is the mean value of the quality indicator, n is the number of quality indicator value subsequences in the quality indicator value subsequence set, is the subsequence index mean of the cth quality index value subsequence in the quality index value subsequence set, is the subsequence indicator fluctuation variance of the cth quality indicator value subsequence in the quality indicator value subsequence set.

[0013] In a possible implementation, the processing parameters of the production line are optimized based on the L quality indicator means and the indicator bias weight set to generate target optimized processing parameters, and the following processing is also performed: L quality indicator standard values ​​are obtained, and indicator deviations are calculated in combination with the L quality indicator means to obtain L quality indicator deviation values; preset processing parameters of the production line within the preset processing window are collected; the preset processing parameters are optimized based on the L quality indicator deviation values ​​and the indicator bias weight set to generate the target optimized processing parameters.

[0014] The present application also provides a processing optimization system for a scenario-adaptive connector, including: an application scenario information acquisition module, used to acquire application scenario information of a target connector; a quality indicator bias weight reconstruction module, used to reconstruct the quality indicator bias weight based on the application scenario information, and obtain an indicator bias weight set of a preset quality indicator set; a quality indicator value set generation module, used to collect a connector sequence processed on a production line within a preset processing window, traverse the connector sequence for quality inspection, and generate multiple quality indicator value sets; a quality indicator value sequence generation module, used to use a preset quality indicator set as an index, and perform single indicator serialization processing on the multiple quality indicator value sets in combination with the connector sequence to generate L quality indicator value sequences, wherein the preset quality indicator set includes L quality indicators; an immersion sequence division module, used to traverse the L quality indicators A maximum quality indicator value is extracted from a value sequence to obtain L maximum quality indicator values, and the L quality indicator value sequence is divided into a water-immersed sequence based on the L maximum quality indicator values ​​to generate L quality indicator value subsequence sets, wherein the L quality indicator value subsequence sets correspond to L sub-processing window sets; a quality indicator mean generation module is used to perform asynchronous mean recognition of the same quality indicators on the L quality indicator value subsequence sets respectively to generate L quality indicator means; a target optimized processing parameter generation module is used to perform processing parameter pointing optimization on the production line based on the L quality indicator means and the indicator bias weight set to generate target optimized processing parameters; a target connector production and processing module is used to transmit the target optimized processing parameters to the parameter control unit of the production line to control the processing equipment to perform production and processing of the target connector.

[0015] Through the processing optimization method and system of a scenario-adaptive connector proposed in this application, it is intended to obtain the application scenario information of the target connector; obtain the indicator bias weight set of the preset quality indicator set; traverse the connector sequence for quality inspection to generate multiple quality indicator value sets; generate L quality indicator value sequences; divide the L quality indicator value sequences into water-immersion sequences to generate L quality indicator value subsequence sets; generate L quality indicator means; generate target optimization processing parameters; and control the processing equipment to produce the target connector. The technical problem that the existing connector processing optimization relies on fixed process flow and processing parameters and is difficult to flexibly respond to complex and changeable application scenarios, which leads to poor quality and performance of the produced connectors and low production efficiency is solved, and the technical effect of improving the processing quality, performance and production efficiency of the connector is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solution of the embodiment of the present disclosure, the accompanying drawings of the embodiment of the present disclosure will be briefly introduced below. A flow chart is used in the present application to illustrate the operations performed by the system according to the embodiment of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.

[0017] Figure 1 A schematic diagram of a process flow of a scene-adaptive connector processing optimization method provided in an embodiment of the present application;

[0018] Figure 2 A schematic diagram of the structure of a processing optimization system for a scene-adaptive connector provided in an embodiment of the present application.

[0019] Explanation of the accompanying drawings: application scenario information acquisition module 10, quality indicator bias weight reconstruction module 20, quality indicator value set generation module 30, quality indicator value sequence generation module 40, immersion sequence division module 50, quality indicator mean generation module 60, target optimization processing parameter generation module 70, target connector production and processing module 80. DETAILED DESCRIPTION

[0020] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0021] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0022] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the terms "first\second" involved are merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "including" and "having" and any variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.

[0023] The present application embodiment provides a method for optimizing the processing of a scene-adaptive connector, such as Figure 1 As shown, the method includes:

[0024] Step S100, obtain application scenario information of the target connector. Obtain key information of the scenario (environment, conditions, functional requirements, and connection method with other devices, etc.) in which the target connector is located in actual application. Specifically, application scenario information may include environmental conditions, such as ambient temperature, humidity, vibration, electromagnetic interference and other conditions of the connector; functional requirements, the specific functions that the connector needs to undertake in the system, such as signal transmission, current transmission, data exchange, etc.; connection method, the connection method between the connector and other devices or components, such as plug-in, welding, threaded connection, etc. Different connection methods require different processing and assembly technologies; space constraints, the installation space constraints of the connector in the device, including factors such as size, shape, layout, etc., which affect the design and processing accuracy of the connector; reliability requirements, the reliability requirements of the connector in the system, including service life, failure rate, maintenance difficulty and other aspects.

[0025] Step S200: reconstructing the quality indicator bias weight based on the application scenario information to obtain an indicator bias weight set of a preset quality indicator set. According to the application scenario information of the target connector, the importance of each quality indicator in the preset quality indicator set is evaluated and configured with a corresponding weight value, so as to obtain the indicator bias weight set of the preset quality indicator set. Specifically, a preset quality indicator set is determined, which usually covers the key parameters of the connector in terms of processing, performance, reliability, etc., such as dimensional accuracy, contact resistance, insulation performance, plugging and unplugging force, environmental resistance, etc. According to the application scenario information, the preset quality indicators are deeply analyzed to understand the importance and influence of different indicators in the scenario. For example, in a high temperature and high humidity environment, the environmental resistance and insulation performance of the connector may be particularly important; while in high-frequency signal transmission applications, indicators such as contact resistance and signal attenuation may be more critical. According to the analysis results of the application scenario information, each indicator in the preset quality indicator set is ranked in importance and assigned a corresponding weight value. The size of the weight value reflects the relative importance of the indicator in a specific application scenario. Finally, a set of quality indicator bias weight sets for specific application scenarios is obtained as input parameters of subsequent processing optimization methods to guide the intelligent optimization adjustment of processing parameters and the formulation of quality detection strategies.

[0026] In a possible implementation, step S200 further includes step S210, extracting a scenario feature set of the application scenario information, wherein the scenario feature set includes jitter capability constraints, connection method complexity, and transmission rate constraints. Extract key features related to connector performance, reliability, or failure mode from a given application scenario, including jitter capability constraints, connection method complexity, and transmission rate constraints. Specifically, the jitter capability constraint refers to the range of jitter (i.e., signal time deviation) that the connector can tolerate when transmitting a signal; the connection method complexity describes the difficulty or complexity of the connector connection method (such as the number of plug-in and unplug times, the number of connectors, the complexity of the connection structure, etc.), and a complex connection method may increase the risk of failure; the transmission rate constraint refers to the maximum data transmission rate that the connector can support, and a high transmission rate may require a higher level of signal integrity and a lower failure rate.

[0027] Step S220, matching the initial indicator bias weight set based on the scenario feature set. The initial indicator bias weight set may be a set of predefined weights used to represent the relative importance of different quality indicators in evaluating connector performance. By comparing the scenario feature set with a known application scenario feature library, a suitable initial weight set is matched and selected. Step S230, using the application scenario information as an index, performing frequent fault item mining to obtain multiple connector fault records. Using application scenario information (such as device type, working environment, operating conditions, etc.) as an index, frequently occurring failure modes or failure records related to a specific application scenario are mined from historical failure data, providing information about specific problems that the connector may encounter in the application scenario.

[0028] Step S240, aggregate the multiple connector fault records with the same index to generate a scenario index fault degree set. Aggregate the multiple mined connector fault records according to the same performance index (such as connection stability, transmission efficiency, durability, etc.), calculate the fault degree under each performance index (that is, the data obtained after analyzing the frequency and severity of the fault under the performance index, used to reflect the importance of the performance index), and the generated scenario index fault degree set reflects the fault condition of the connector in the application scenario under different performance indicators.

[0029] Optionally, taking the performance indicator as an index, the number of failures of each performance indicator in the multiple connector fault records and the failure coefficient of each failure are counted respectively (the failure coefficient is a coefficient obtained by technical personnel in this field after evaluation based on the failure repair cost and the degree of failure damage. The larger the failure coefficient, the greater the failure impact of the corresponding performance indicator when the connector fails), and the number of failures and the failure coefficient of each failure are weightedly calculated to obtain the failure influence coefficient of each performance indicator and a set of failure influence coefficients.

[0030] Furthermore, each fault influence coefficient in the fault influence coefficient set is respectively divided by the sum of the coefficients of the fault influence coefficient set to obtain the fault degree of each performance indicator, and generate the scenario indicator fault degree set.

[0031] Step S250, reconstructing the initial indicator bias weight set based on the scenario indicator failure degree set to generate the indicator bias weight set. According to the information provided in the scenario indicator failure degree set, the initial indicator bias weight set is reconstructed to generate the indicator bias weight set. Specifically, the weights of performance indicators with higher failure degrees are increased, while the weights of indicators with lower failure degrees are reduced, so that the reconstructed indicator bias weight set is closer to the needs of actual application scenarios and can more accurately reflect the relative importance of different quality indicators in evaluating connector performance.

[0032] Optionally, a weight reconstruction identifier is constructed, and the scenario indicator failure degree set and the initial indicator bias weight set are input into the weight reconstruction identifier for analysis to obtain the indicator bias weight set. By obtaining multiple sample scenario indicator failure degree sets, multiple sample initial indicator bias weight sets, and multiple sample indicator bias weight sets as training samples, the framework constructed based on the convolutional neural network is supervised and trained using the training samples. During the training, the two-to-one mapping relationship between the scenario indicator failure degree set and the initial indicator bias weight set and the indicator bias weight set is learned until the model output converges, and the trained weight reconstruction identifier is obtained. Thereby, the goal of adjusting the indicator bias weight situation according to the actual fault situation in different scenarios is achieved.

[0033] Step S300 , collecting a connector sequence processed by a production line within a preset processing window, traversing the connector sequence for quality inspection, and generating a plurality of quality indicator value sets. In the connector processing process, in order to monitor and evaluate the processing quality of the production line, a specific time period (i.e., preset processing window) is set as the quality inspection cycle. During this period, the production line processes multiple connectors, which form a sequence according to the processing order. Then, each connector is subjected to comprehensive quality inspection by quality inspection means. Specifically, the processing speed, processing capacity, inspection efficiency of the quality inspection equipment, and production plan arrangement of the production line are comprehensively considered, and the processing window is preset to facilitate regular quality inspection of the production line to ensure the stability and consistency of the processing quality. Within the preset processing window, the production line processes multiple connectors according to the established process flow, and collects and sorts them in a certain order (such as time sequence, batch sequence, etc.) to obtain a connector sequence. According to the use scenario and performance requirements of the connector, the quality indicators that need to be inspected are determined, which may include dimensional accuracy, contact resistance, insulation performance, insertion and extraction force, environmental resistance, etc. Each connector in the connector sequence is subjected to comprehensive quality inspection to obtain multiple quality indicator values. Finally, the inspection results of each quality indicator of each connector are sorted to generate multiple quality indicator sets, including information such as the connector number, inspection items, and inspection results, reflecting the overall quality level of the connector in the batch or processing window.

[0034] Step S400, using a preset quality indicator set as an index and combining the connector sequence to perform single indicator serialization processing on the multiple quality indicator value sets to generate L quality indicator value sequences, wherein the preset quality indicator set includes L quality indicators. According to the preset quality indicator set, the multiple indicator values ​​(i.e., multiple quality indicator value sets) obtained from the quality inspection of each connector are split and reorganized so that each quality indicator forms a separate numerical sequence arranged in the order of the connector sequence. If the preset quality indicator set includes L different quality indicators (such as dimensional accuracy, contact resistance, insulation performance, etc.), then the single indicator serialization processing will generate L corresponding quality indicator value sequences. The elements in each sequence are the specific detection values ​​of the connectors in the connector sequence under the quality indicator. Specifically, each quality indicator in the preset quality indicator set is used as an index to traverse the quality indicator value sets of all connectors. For each quality indicator, all detection values ​​under the indicator are extracted from the quality inspection results of all connectors and arranged in the order of the connector sequence. For each quality indicator, a separate quality indicator value sequence is generated. The elements in the sequence are the quality indicator values ​​arranged in the connector sequence. Finally, L quality indicator value sequences are obtained, corresponding to each quality indicator in the preset quality indicator set.

[0035] Step S500, traverse the L quality indicator value sequences to extract the maximum quality indicator value, obtain L quality indicator maximum values, and divide the L quality indicator value sequences into immersion sequences based on the L quality indicator maximum values ​​to generate L quality indicator value subsequence sets, wherein the L quality indicator value subsequence sets correspond to L sub-processing window sets. Waterlogged sequence partitioning is used to describe the segmentation of a sequence based on a certain threshold or condition, that is, each quality indicator value sequence is divided based on a certain partitioning standard. Specifically, for each quality indicator value sequence (a total of L), the quality indicator value sequence is traversed to extract the maximum quality indicator value therein, and L quality indicator maximum values ​​are obtained. Then, the L quality indicator value sequences are subjected to waterlogged sequence partitioning based on the L quality indicator maximum values. For example, a threshold is set based on the maximum value of each quality indicator (such as a certain percentage of the maximum value), and then the continuous quality indicator values ​​below the threshold are divided into a subsequence, and then each sequence is split into multiple subsequences to obtain L quality indicator value subsequence sets, each set corresponds to a quality indicator, and each quality indicator value subsequence set corresponds to L sub-processing window sets, that is, the data set contains information about the processing window (such as timestamp, batch number, etc.).

[0036] In a possible implementation, step S500 further includes step S510, wherein the L maximum values ​​of the quality indicators are respectively used as the starting point for dividing the waterlogging sequence, and the quality indicator values ​​are searched on both sides in the L quality indicator value sequence according to the preset division threshold value to obtain L first left-direction quality indicator values ​​and L first right-direction quality indicator values. The L maximum values ​​of the quality indicators are respectively used as the starting point for dividing the waterlogging sequence, that is, as the division point, and starting from each maximum value, the quality indicator values ​​are searched on both sides (left and right) of the sequence according to the preset division threshold value. In the left-direction search, the first quality indicator value whose difference with the maximum value is less than the preset division threshold value is found, and recorded as the first left-direction quality indicator value; in the right-direction search, the first quality indicator value whose difference with the maximum value is less than the preset division threshold value is also found, and recorded as the first right-direction quality indicator value.

[0037] Step S520, determine whether the difference between the L maximum values ​​of the quality indicators and the L first left-direction quality indicator values ​​is greater than or equal to a preset division threshold, if so, stop searching for the quality indicator value in the left direction; step S530, if not, take the L maximum values ​​of the quality indicators as the flooding point, and under the preset water injection amount, water will flow from the flooding point to the L first left-direction quality indicator values, and add the L first left-direction quality indicator values ​​to the L first quality indicator value subsequences of the L maximum values ​​of the quality indicators. If the difference between the maximum value and the first left-direction quality indicator value is greater than or equal to the preset division threshold, stop searching in the left direction, because the left part is sufficient to be distinguished from the maximum value, if the above condition is not met (i.e., the difference is less than the preset division threshold), the first left-direction quality indicator value and all values ​​on its left side (if any) are regarded as belonging to the same subsequence as the maximum value, i.e., the first quality indicator value subsequence, and these values ​​are added to the subsequence.

[0038] Step S540, when the difference between the L maximum values ​​of the quality indicators and the L first rightward quality indicator values ​​is less than the preset division threshold, the L maximum values ​​of the quality indicators are taken as the flooding point, and water will flow from the flooding point to the L first rightward quality indicator values ​​under the preset water injection amount, and the L first rightward quality indicator values ​​are added to the L first quality indicator value subsequences; Step S550, when both the left and right directions stop searching for quality indicator values, the searched L first quality indicator value subsequences are obtained. Similarly, for the rightward search, if the difference between the maximum value and the first rightward quality indicator value is less than the preset division threshold, the first rightward quality indicator value and all values ​​to its right (if any) are added to the first quality indicator value subsequence, and when both the left and right searches stop, it means that the flooding sequence division of each quality indicator maximum value has been completed, and the searched L first quality indicator value subsequences are obtained.

[0039] Step S560, perform water-sequence partitioning on both sides of the L first quality indicator value subsequences of the L quality indicator value sequences, respectively, to generate a set of L quality indicator value subsequences. For each first quality indicator value subsequence, it is regarded as a central segment, and its two sides are the parts outside the subsequence but still in the original quality indicator value sequence, and these two sides are respectively called the left side and the right side. For the left and right parts of each first quality indicator value subsequence, the water-sequence partitioning method is applied again, that is, starting from a certain point on the left or right side (which may be the maximum value, minimum value, etc.), searching to both sides according to a preset partitioning threshold to identify a smaller subsequence, and recursively performing, that is, after a new subsequence is partitioned on the left or right side of a subsequence, continue to partition the two sides of these new subsequences until a certain stop condition is met. Conditions (such as the length of the subsequence is less than a certain threshold, the difference of all values ​​in the subsequence is less than a certain threshold, etc.), as the recursive partitioning proceeds, multiple subsequences of different levels are generated, covering different parts of the original quality indicator value sequence, and each subsequence represents a relatively independent or homogeneous fragment of the data. Finally, all the generated subsequences are collected to form a quality indicator value subsequence set, which includes all subsequences divided from the original L quality indicator value sequences, and each subsequence carries the identifier of its original sequence (to distinguish which quality indicator they come from) and its position information in the original sequence.

[0040] Step S600: performing quality indicator asynchronous mean identification on the L quality indicator value subsequence sets respectively to generate L quality indicator means.

[0041] Furthermore, step S600 also includes step S610, performing indicator value mean calculation on each quality indicator value subsequence in the L quality indicator value subsequence sets to obtain L subsequence indicator mean sets; step S620, performing indicator value fluctuation variance calculation on each quality indicator value subsequence in the L quality indicator value subsequence sets to obtain L subsequence indicator fluctuation variance sets; step S630, inputting the L subsequence indicator mean sets and the L subsequence indicator fluctuation variance sets into the indicator mean calculation formula for analysis to obtain L quality indicator means.

[0042] Furthermore, step S630 also includes that the indicator mean value calculation formula is:

[0043] The formula for calculating the mean value of the indicator is:

[0044] ;

[0045] in, is the mean value of the quality indicator, n is the number of quality indicator value subsequences in the quality indicator value subsequence set, is the subsequence index mean of the cth quality index value subsequence in the quality index value subsequence set, is the subsequence indicator fluctuation variance of the cth quality indicator value subsequence in the quality indicator value subsequence set.

[0046] Asynchronous mean recognition of the same quality index refers to performing mean recognition processing of different windows for each quality index (such as dimensional accuracy, contact resistance, etc.), that is, calculating the average value of the quality index value in each subsequence. Specifically, for each quality index (such as dimensional accuracy, contact resistance, etc.) in the set of L quality index value subsequences, traverse all its subsequences, calculate the mean of the quality index value in each subsequence, obtain the index mean of the subsequence, collect the index means of all subsequences, form the subsequence index mean set of the quality index, and obtain L subsequences. Indicator mean set; similarly, for each subsequence of each quality indicator, calculate the fluctuation variance of its quality indicator value to reflect the discrete degree of data points in the subsequence, collect the indicator fluctuation variances of all subsequences to form a subsequence indicator fluctuation variance set of the quality indicator, and obtain L subsequence indicator fluctuation variance sets; then input the L subsequence indicator mean sets and the L subsequence indicator fluctuation variance sets into the indicator mean calculation formula for analysis, that is, use the indicator mean and fluctuation variance to comprehensively evaluate the comprehensive performance of each quality indicator, and finally obtain the L quality indicator means.

[0047] Step S700, based on the L quality index means and the index bias weight set, the processing parameter orientation optimization of the production line is performed to generate target optimized processing parameters. The processing parameter orientation optimization of the production line is performed based on the L quality index means and the index bias weight set, specifically refers to establishing a multi-objective optimization model, taking the quality index mean as the optimization target, and taking the index bias weight as the constraint condition or the coefficient of the objective function, selecting a suitable optimization algorithm for solving (such as genetic algorithm, particle swarm optimization algorithm, simulated annealing algorithm, etc.) according to the complexity and scale of the optimization model, and performing multiple iterations in the specified search space. Each iteration will calculate the quality index mean according to the current processing parameter combination, and update the processing parameters according to the rules of the optimization algorithm to find the optimal processing parameter combination. After repeated iterations and optimization, a set of target optimized processing parameters will eventually be generated. This set of parameters can achieve the best performance of the production line under the premise of meeting production needs and quality requirements, that is, by adjusting the processing parameters of the production line (such as temperature, pressure, speed, time, etc.) to minimize the deviation between the quality index mean and the target value, and at the same time consider the index bias weight, so as to achieve the overall optimization of the production line.

[0048] In a possible implementation, step S700 further includes step S710, obtaining L quality index standard values, combining the L quality index mean values ​​to calculate index deviation, and obtaining L quality index deviation values. Based on historical data, the standard value of each quality index (such as dimensional accuracy, surface roughness, material strength, etc.) is determined, which represents the quality level that the products produced on the production line should achieve, and the index deviation is calculated in combination with the L quality index mean values. Specifically, the actual mean values ​​of the L quality indexes are obtained and compared with the respective quality index standard values ​​to calculate the index deviation value, and the L quality index deviation values ​​are obtained, which reflect the gap between the actual production results and the expected goals, which can be positive values ​​(indicating exceeding the standard) or negative values ​​(indicating below the standard). Step S720, collecting the preset processing parameters of the production line within the preset processing window. The preset processing parameters of the production line within the preset processing window are obtained, such as temperature, pressure, speed, time, etc. These processing parameters are key factors affecting product quality. Step S730, based on the L quality index deviation values ​​and the index bias weight set, the preset processing parameters are directed to optimize and generate the target optimized processing parameters. According to the L quality index deviation values ​​and the index bias weight set, one or more objective functions are defined, aiming to minimize the weighted sum of quality index deviations or a certain form of comprehensive deviation index. The preset processing parameters are decision variables, which are parameters that can be adjusted to optimize the production process. Specifically, the algorithm parameters are initialized. In each iteration, the objective function and constraints are calculated according to the values ​​of the current decision variables. The values ​​of the decision variables are updated according to the rules of the optimization algorithm to find a better solution. The iterative process is repeated until the stop condition is met (such as reaching the maximum number of iterations, the objective function value no longer changes significantly, etc.), and the output results of the optimization algorithm are analyzed, that is, the target optimized processing parameters are generated.

[0049] Step S800, the target optimized processing parameters are transmitted to the parameter control unit of the production line, and the processing equipment is controlled to produce the target connector. The target optimized processing parameters are transmitted to the parameter control unit of the production line by some means (such as network communication, data interface, etc.). After receiving the target optimized processing parameters, the parameter control unit will automatically set these parameters to the corresponding processing equipment, that is, the parameter control unit of the production line will monitor and adjust the processing equipment in real time according to the set target optimized processing parameters, including controlling key parameters such as processing speed, temperature, pressure, time, etc., to ensure the stability and consistency of the production process. The production line is usually also equipped with a feedback mechanism to monitor various data in the processing process (such as product quality indicators, equipment operation status, etc.) in real time, and compare them with the target optimized processing parameters. If there is a deviation, the control system will make timely adjustments to restore the optimal performance of the production line to ensure that the processing process is always in the best state.

[0050] In the above, refer to Figure 1 A method for optimizing the processing of a scene adaptive connector according to an embodiment of the present invention is described in detail. Figure 2 A processing optimization system for a scene-adaptive connector according to an embodiment of the present invention is described.

[0051] A processing optimization system for a scenario-adaptive connector according to an embodiment of the present invention is used to solve the technical problems of existing connector processing optimization, which is dependent on fixed process flow and processing parameters and is difficult to flexibly respond to complex and changeable application scenarios, thereby resulting in poor quality and performance of the produced connectors and low production efficiency, thereby achieving the technical effect of improving the processing quality, performance and production efficiency of the connectors. A processing optimization system for a scenario-adaptive connector includes: an application scenario information acquisition module 10, a quality indicator bias weight reconstruction module 20, a quality indicator value set generation module 30, a quality indicator value sequence generation module 40, an immersion sequence division module 50, a quality indicator mean generation module 60, a target optimization processing parameter generation module 70, and a target connector production and processing module 80.

[0052] An application scenario information acquisition module 10 is used to acquire application scenario information of a target connector; a quality indicator bias weight reconstruction module 20 is used to reconstruct quality indicator bias weights based on the application scenario information to obtain an indicator bias weight set of a preset quality indicator set; a quality indicator value set generation module 30 is used to collect a connector sequence processed on a production line within a preset processing window, traverse the connector sequence for quality inspection, and generate multiple quality indicator value sets; a quality indicator value sequence generation module 40 is used to use a preset quality indicator set as an index, combine the connector sequence to perform single indicator serialization processing on the multiple quality indicator value sets, and generate L quality indicator value sequences, wherein the preset quality indicator set includes L quality indicators; an immersion sequence division module 50 is used to traverse the L quality indicator value sequences for maximum quality indicator value. Extract, obtain L maximum values ​​of quality indicators, and divide the L quality indicator value sequences into water-immersed sequences based on the L maximum values ​​of quality indicators to generate L quality indicator value subsequence sets, wherein the L quality indicator value subsequence sets correspond to L sub-processing window sets; a quality indicator mean generation module 60, used to perform asynchronous mean identification of the L quality indicator value subsequence sets with the same quality indicators, and generate L quality indicator means; a target optimized processing parameter generation module 70, used to perform processing parameter pointing optimization on the production line based on the L quality indicator means and the indicator bias weight set, and generate target optimized processing parameters; a target connector production and processing module 80, used to transmit the target optimized processing parameters to the parameter control unit of the production line, and control the processing equipment to perform production and processing of the target connector.

[0053] The specific configuration of the quality indicator bias weight reconstruction module 20 will be described in detail below. The quality indicator bias weight reconstruction module 20 further includes: extracting a scenario feature set of the application scenario information, wherein the scenario feature set includes jitter capability constraints, connection mode complexity, and transmission rate constraints; matching the initial indicator bias weight set based on the scenario feature set; using the application scenario information as an index, performing fault frequent item mining to obtain multiple connector fault records; performing same-indicator aggregation on the multiple connector fault records to generate a scenario indicator fault degree set; reconstructing the initial indicator bias weight set based on the scenario indicator fault degree set to generate the indicator bias weight set.

[0054] The specific configuration of the flooding sequence division module 50 will be described in detail below. The flooding sequence division module 50 may further include: taking the L maximum values ​​of the quality indicators as the starting point for flooding sequence division, searching for quality indicator values ​​in both directions in the L quality indicator value sequence according to a preset division threshold, and obtaining L first left-direction quality indicator values ​​and L first right-direction quality indicator values; judging whether the indicator value difference between the L maximum values ​​of the quality indicators and the L first left-direction quality indicator values ​​is greater than or equal to the preset division threshold, and if so, stopping searching for quality indicator values ​​in the left direction; if not, taking the L maximum values ​​of the quality indicators as the flooding point, and under the preset water injection amount, water will flow from the flooding point to the L first left-direction quality indicator values, and the L first left-direction quality indicator values ​​will be filled. The value is added into the L first quality indicator value subsequences of the L quality indicator maximum values; when the difference between the L quality indicator maximum values ​​and the L first right-direction quality indicator values ​​is less than a preset division threshold, the L quality indicator maximum values ​​are taken as the flooding point, and under a preset water injection amount, water will flow from the flooding point to the L first right-direction quality indicator values, and the L first right-direction quality indicator values ​​are added into the L first quality indicator value subsequences; when the quality indicator value search in both the left and right directions is stopped, the searched L first quality indicator value subsequences are obtained; the flooding sequence is divided on both sides of the L first quality indicator value subsequences of the L quality indicator value sequence, and a set of L quality indicator value subsequences is generated.

[0055] The specific configuration of the immersion sequence division module 50 will be described in detail below. The immersion sequence division module 50 may further include: respectively extracting the time nodes of the left and right endpoints of each quality index value subsequence in the L quality index value subsequence sets to obtain L sub-processing window sets.

[0056] The specific configuration of the quality indicator mean value generation module 60 will be described in detail below. The quality indicator mean value generation module 60 further includes: performing indicator value mean calculation on each quality indicator value subsequence in the L quality indicator value subsequence sets to obtain L subsequence indicator mean value sets; performing indicator value fluctuation variance calculation on each quality indicator value subsequence in the L quality indicator value subsequence sets to obtain L subsequence indicator fluctuation variance sets; inputting the L subsequence indicator mean value sets and the L subsequence indicator fluctuation variance sets into the indicator mean value calculation formula for analysis to obtain L quality indicator means.

[0057] The specific configuration of the quality indicator mean value generation module 60 will be described in detail below. The quality indicator mean value generation module 60 further includes: the indicator mean value calculation formula is:

[0058] ;

[0059] in, is the mean value of the quality indicator, n is the number of quality indicator value subsequences in the quality indicator value subsequence set, is the subsequence index mean of the cth quality index value subsequence in the quality index value subsequence set, is the subsequence indicator fluctuation variance of the cth quality indicator value subsequence in the quality indicator value subsequence set.

[0060] The specific configuration of the target optimization processing parameter generation module 70 will be described in detail below. The target optimization processing parameter generation module 70 may further include: obtaining L quality index standard values, performing index deviation calculation in combination with the L quality index mean values, and obtaining L quality index deviation values; collecting the preset processing parameters of the production line within the preset processing window; and performing directional optimization on the preset processing parameters based on the L quality index deviation values ​​and the index bias weight set to generate the target optimization processing parameters.

[0061] A scene adaptive connector processing optimization system provided by an embodiment of the present invention can execute a scene adaptive connector processing optimization method provided by any embodiment of the present invention, and has functional modules and beneficial effects corresponding to the execution method.

[0062] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules may be used and run on the user terminal and / or server, and the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0063] The above specific implementations do not constitute a limitation on the protection scope of this application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of this application should be included in the protection scope of this application.

Claims

1. A method for optimizing the processing of a scene-adaptive connector, characterized in that: include: Get the application scenario information of the target connector; Reconstructing the quality indicator bias weights based on the application scenario information to obtain an indicator bias weight set of a preset quality indicator set; Collecting a connector sequence processed by a production line within a preset processing window, traversing the connector sequence for quality inspection, and generating a plurality of quality indicator value sets; Taking a preset quality indicator set as an index, performing single indicator serialization processing on the multiple quality indicator value sets in combination with the connector sequence to generate L quality indicator value sequences, wherein the preset quality indicator set includes L quality indicators; Traversing the L quality indicator value sequences to extract the maximum quality indicator value, obtaining L quality indicator maximum values, and performing water immersion sequence division on the L quality indicator value sequences based on the L quality indicator maximum values ​​to generate L quality indicator value subsequence sets, wherein the L quality indicator value subsequence sets correspond to L sub-processing window sets; Respectively performing the same-quality-indicator asynchronous mean identification on the L quality indicator value subsequence sets to generate L quality indicator means; Based on the mean values ​​of the L quality indicators and the indicator bias weight set, the processing parameter orientation optimization of the production line is performed to generate target optimized processing parameters; Transmitting the target optimized processing parameters to the parameter control unit of the production line to control the processing equipment to perform production processing of the target connector; Respectively performing the same-quality-indicator asynchronous mean identification on the L quality indicator value subsequence sets to generate L quality indicator means, including: Calculating the mean value of each quality indicator value subsequence in the L quality indicator value subsequence sets to obtain a set of L subsequence indicator means; Calculating the index value fluctuation variance for each quality index value subsequence in the L quality index value subsequence sets to obtain L subsequence index fluctuation variance sets; The L subsequence indicator mean value set and the L subsequence indicator fluctuation variance set are input into the indicator mean calculation formula for analysis to obtain L quality indicator means.

2. A method for optimizing the processing of a scene-adaptive connector according to claim 1, characterized in that: Reconstructing the quality indicator bias weight based on the application scenario information to obtain an indicator bias weight set of a preset quality indicator set includes: Extracting a scenario feature set of the application scenario information, wherein the scenario feature set includes jitter capability constraint, connection mode complexity and transmission rate constraint; Matching an initial indicator bias weight set based on the scene feature set; Using the application scenario information as an index, frequent fault item mining is performed to obtain multiple connector fault records; Aggregating the multiple connector fault records with the same index to generate a scenario index fault degree set; The initial indicator bias weight set is reconstructed based on the scenario indicator failure degree set to generate the indicator bias weight set.

3. The method for optimizing the processing of a scene-adaptive connector according to claim 1, characterized in that: Traversing the L quality indicator value sequences to extract the maximum quality indicator value, obtaining L quality indicator maximum values, and performing water-immersion sequence division on the L quality indicator value sequences based on the L quality indicator maximum values ​​to generate L quality indicator value subsequence sets, including: The L maximum values ​​of the quality indicators are respectively used as the starting points for dividing the water immersion sequence, and the quality indicator values ​​are searched on both sides in the L quality indicator value sequences according to the preset division threshold value to obtain L first left-direction quality indicator values ​​and L first right-direction quality indicator values; Determine whether the difference between the L maximum values ​​of the quality indicators and the L first left-direction quality indicator values ​​is greater than or equal to a preset division threshold, and if so, stop searching for the quality indicator value in the left direction; If not, the maximum values ​​of the L quality indicators are taken as the flooding point, and water will flow from the flooding point to the L first left-direction quality indicator values ​​under the preset water injection volume, and the L first left-direction quality indicator values ​​are added to the L first quality indicator value subsequence of the L maximum values ​​of the quality indicators; When the difference between the L maximum values ​​of the quality indicators and the L first rightward quality indicator values ​​is less than a preset division threshold, the L maximum values ​​of the quality indicators are taken as the flooding point, and water will flow from the flooding point to the L first rightward quality indicator values ​​under a preset water injection amount, and the L first rightward quality indicator values ​​are added to the L first quality indicator value subsequence; When both the left and right directions stop searching for the quality indicator value, L first quality indicator value subsequences that have been searched are obtained; Perform water-soaking sequence division on both sides of the L first quality indicator value subsequences of the L quality indicator value sequences respectively to generate L quality indicator value subsequence sets.

4. A method for optimizing the processing of a scene-adaptive connector according to claim 3, characterized in that: The time nodes of the left and right endpoints of each quality indicator value subsequence in the L quality indicator value subsequence sets are respectively extracted to obtain L sub-processing window sets.

5. The method for optimizing the processing of a scene-adaptive connector according to claim 1, characterized in that: include: The formula for calculating the mean value of the indicator is: ； in, is the mean value of the quality indicator, n is the number of quality indicator value subsequences in the quality indicator value subsequence set, is the subsequence index mean of the cth quality index value subsequence in the quality index value subsequence set, is the subsequence indicator fluctuation variance of the cth quality indicator value subsequence in the quality indicator value subsequence set.

6. The method for optimizing the processing of a scene-adaptive connector according to claim 1, characterized in that: Based on the L quality index means and the index bias weight set, the production line is optimized for processing parameters to generate target optimized processing parameters, including: Obtain L quality indicator standard values, and calculate indicator deviations based on the L quality indicator mean values ​​to obtain L quality indicator deviation values; Collecting preset processing parameters of the production line within the preset processing window; Based on the L quality index deviation values ​​and the index bias weight set, the preset processing parameters are optimized to generate the target optimized processing parameters.

7. A processing optimization system for a scene-adaptive connector, characterized in that: The system is used to implement a method for optimizing the processing of a scene-adaptive connector according to any one of claims 1 to 6, and the system comprises: An application scenario information acquisition module, used to acquire application scenario information of a target connector; A quality indicator bias weight reconstruction module, used to reconstruct the quality indicator bias weight based on the application scenario information to obtain an indicator bias weight set of a preset quality indicator set; A quality indicator value set generation module, used to collect a connector sequence processed by a production line within a preset processing window, traverse the connector sequence to perform quality inspection, and generate multiple quality indicator value sets; A quality indicator value sequence generation module, used to perform single indicator serialization processing on the multiple quality indicator value sets in combination with the connector sequence using a preset quality indicator set as an index to generate L quality indicator value sequences, wherein the preset quality indicator set includes L quality indicators; a water immersion sequence division module, used for traversing the L quality index value sequences to extract the maximum quality index value, obtaining L quality index maximum values, and performing water immersion sequence division on the L quality index value sequences based on the L quality index maximum values ​​to generate L quality index value subsequence sets, wherein the L quality index value subsequence sets correspond to L sub-processing window sets; A quality indicator mean value generation module is used to perform asynchronous mean value recognition of the same quality indicator on the L quality indicator value subsequence sets respectively to generate L quality indicator means; A target optimized processing parameter generation module is used to perform processing parameter orientation optimization on the production line based on the L quality index means and the index bias weight set to generate target optimized processing parameters; The target connector production and processing module is used to transmit the target optimized processing parameters to the parameter control unit of the production line, and control the processing equipment to perform the production and processing of the target connector.

Images