Intelligent assembly component whole process management system

Abstract

The application relates to the technical field of intelligent monitoring management, and particularly discloses a whole-process management system for intelligent assembly components, which aims to optimize production efficiency and product quality through real-time data acquisition and analysis, and comprises the following modules: a data acquisition module that monitors and records switching time of each process and quality inspection pass rate of each component in real time through sensors and automatic equipment integrated on a production line; a production plan evaluation module that calculates an abnormal coefficient of switching time, identifies bottlenecks or inefficient operations in the switching time, and dynamically adjusts the switching time based on specific values of influencing interference factors and their positive and negative nature; a production quality evaluation module that calculates a quality abnormal coefficient according to the quality inspection pass rate, judges quality fluctuation of each component; and an influence analysis and adjustment module that uses a gradient boosting regression model to predict the quality pass rate and dynamically adjusts the switching time of the process to eliminate potential quality problems.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent monitoring management, and particularly relates to an intelligent assembly component whole-process management system. BACKGROUND

[0002] In modern manufacturing industry, the production process of assembly components involves multiple complex processes and strict quality control standards. In order to ensure production efficiency and product quality, enterprises widely adopt automated equipment, sensor technology and data analysis tools. However, traditional production management systems rely on static plans and post-analysis, and it is difficult to monitor and adjust key parameters in real time, such as process switching time and quality inspection pass rate. These systems lack the ability to quickly respond to dynamic changes in the production process, resulting in low production efficiency and frequent quality problems.

[0003] The prior art has the following disadvantages:

[0004] The prior art has the following disadvantages: The prior art has the following disadvantages: Firstly, traditional production management systems usually rely on static schedules to arrange process switching, lacking real-time monitoring and dynamic adjustment capabilities for process switching time in actual production processes. This leads to bottlenecks or redundant time in some processes, but due to the lack of real-time data support, it is difficult to discover and optimize in a timely manner, thereby increasing unnecessary downtime and reducing overall production efficiency. Secondly, in terms of quality control, traditional methods mainly rely on periodic sampling and post-analysis, and cannot achieve real-time monitoring and immediate feedback. Quality problems can only be discovered after the product is completed and quality inspection is performed, resulting in defective products flowing into subsequent processes or even the market, affecting customer satisfaction. In addition, the quality fluctuates greatly between different batches, making it difficult to ensure the high quality consistency of each batch of products. Existing systems usually only provide single-dimensional data analysis, lacking comprehensive evaluation and prediction capabilities for the relationship between production efficiency and quality, making it difficult for management personnel to obtain comprehensive data support and make scientific decisions to optimize production processes. Due to the lack of forward-looking prediction models, enterprises can only take remedial measures after problems occur, and cannot prevent potential problems from occurring. These problems together result in low production efficiency and unstable product quality, restricting the competitiveness and development potential of enterprises. Therefore, an intelligent management system capable of real-time monitoring, dynamic adjustment and comprehensive analysis is urgently needed to solve these deficiencies. SUMMARY

[0005] The purpose of the present application is to provide an intelligent assembly component whole-process management system to solve the problems in the above background.

[0006] The purpose of the present application can be achieved by the following technical solutions:

[0007] An intelligent assembly component whole-process management system comprises:

[0008] A data acquisition module acquires switching time and quality inspection pass rate in real time during a component production monitoring period;

[0009] A production plan evaluation module analyzes switching time during the production process in the monitoring period, and evaluates production efficiency in the component production process according to the switching time;

[0010] A production quality evaluation module analyzes quality inspection pass rate during the production process in the monitoring period, and evaluates quality in the component production process according to fluctuation degree of the quality pass rate of each component;

[0011] A comprehensive analysis module analyzes production efficiency in the component production process and quality in the component production process, and evaluates quality consistency of each batch of assembly components in the production process according to an analysis result;

[0012] An influence analysis and adjustment module analyzes influence degree of production efficiency in the component production process on quality in the component production process according to the quality consistency of each batch of assembly components in the production process, and dynamically adjusts switching time in the production process according to the influence degree.

[0013] As a further scheme of the present application, the evaluation of production efficiency in the component production process specifically comprises:

[0014] During the component production monitoring period, switching time in the production process is acquired in real time, an abnormal coefficient of switching time of the whole production process is calculated according to fluctuation degree of different switching time, and it is judged whether the abnormal coefficient of switching time of the whole production process is greater than or equal to a preset threshold value; if yes, it is indicated that switching time of the production process is normal, and then production efficiency is normal; if no, it is indicated that switching time of the production process is abnormal, and then production efficiency is abnormal.

[0015] As a further scheme of the present application, the acquisition process of the abnormal coefficient of switching time comprises:

[0016] Switching time data of each process in the production process is acquired in real time, and each data represents switching time of the 1st process to the nth process;

[0017] The acquired switching time data is subjected to smoothing processing, including noise removal and difference operation;

[0018] Three parameters of the ARIMA model are determined: autoregressive order p, difference order d and moving average order q, and historical data is used for ARIMA model fitting, and the switching time prediction value at each time is calculated. ;

[0019] Among them, represents the switching time data prediction value at the t th moment, represents the collection moment; according to the difference between the actual switching time data and the prediction value , the residual error is calculated, and the calculation expression is: ; wherein, represents the residual error value at the t th moment; The residual error standard deviation value of all moments is calculated, and the residual error value and the residual error standard deviation value are calculated and processed to obtain the switching time abnormality coefficient, and the calculation expression is:

[0020] ;

[0021] Among them, represents the total number of collection time points, represents the residual error standard deviation value, represents the switching time abnormality coefficient.

[0022] As a further scheme of the present application: in the component production monitoring period, the quality inspection qualified rate of each component in the production process is collected in real time, the quality abnormality coefficient of the corresponding component when the quality is detected is calculated according to the fluctuation degree of the quality qualified rate of each component, and it is judged whether the quality abnormality coefficient of each component is greater than or equal to the preset threshold value, if yes, it means that the corresponding quality qualified rate has abnormal fluctuation, then the production quality of the component is poor, if not, it means that the corresponding quality qualified rate does not have abnormal fluctuation, then the production quality of the component is good.

[0023] As a further scheme of the present application: the acquisition process of the quality abnormality coefficient is:

[0024] In the component production monitoring period, the quality inspection qualified rate of each component in the production process is collected in real time according to the time sequence;

[0025] The quality qualified rate at the t th moment, the factors affecting the quality qualified rate and the quality detection result at the t th moment are obtained, , represents the collection moment;

[0026] ​​​The posterior probability distribution of the quality pass rate is inferred according to the Bayesian theorem, and the calculation expression is: ;

[0027] Wherein, represents the posterior probability distribution, represents the likelihood function of the quality detection result, represents the prior probability of the quality pass rate, represents a standardization constant;

[0028] The expected value of the posterior probability distribution of the quality pass rate obtained by Bayesian inference is calculated;

[0029] The quality anomaly coefficient is obtained by calculating the difference between the quality pass rate of each component and the expected pass rate, and the calculation expression is: ;

[0030] Wherein, represents the total number of collection time points, represents the standard deviation of the quality pass rate, represents the quality anomaly coefficient, represents the expected value of the posterior probability distribution of the quality pass rate obtained according to Bayesian inference.

[0031] As a further scheme of the application: the production efficiency in the component production process and the quality in the component production process are analyzed, and specifically include:

[0032] The switching time anomaly coefficient and the quality anomaly coefficient in the component production process are acquired, and the switching time anomaly coefficient and the quality anomaly coefficient are normalized and calculated, the quality consistency coefficient is calculated, and the quality consistency of each batch of assembled components in the production process is evaluated;

[0033] The calculation process of the consistency coefficient is:

[0034] The switching time anomaly coefficient and the quality anomaly coefficient in the production process of each component are collected;

[0035] The switching time anomaly coefficient and the quality anomaly coefficient are normalized and calculated;

[0036] For each component The distance between the component is calculated using the Euclidean distance, and the calculation expression is: ;

[0037] Wherein, represents the Euclidean distance between the component and the component , and a switching time abnormality coefficient of the component, a switching time abnormality coefficient of the component a normalized value of the switching time abnormality coefficient, a switching time abnormality coefficient of the component a normalized value of the switching time abnormality coefficient, a quality abnormality coefficient of the component a normalized value of the quality abnormality coefficient, a quality abnormality coefficient of the component a normalized value of the quality abnormality coefficient; for each component , the K nearest neighbors are selected, and the average distance between these neighbors and the component is calculated to obtain the quality consistency coefficient, and the calculation expression is: ;

[0038] wherein, a quality consistency coefficient of the component , represents the number of selected neighbors.

[0039] As a further scheme of the present application: the evaluation of the quality consistency of the components assembled in each batch in the production process specifically comprises:

[0040] the quality consistency coefficients of the components assembled in each batch in the production process are calculated respectively, and the quality consistency coefficients of the components assembled in each batch are compared with a preset threshold value;

[0041] it is judged whether the quality consistency coefficient is greater than or equal to the preset threshold value, if yes, the component quality is consistent, if not, the component quality is inconsistent.

[0042] As a further scheme of the present application: the analysis of the influence degree of the production efficiency in the component production process on the quality in the component production process specifically comprises:

[0043] according to the influence of the production efficiency in the component production process on the quality in the component production process, an influence interference factor is calculated, it is judged whether the influence interference factor in the component production process is greater than or equal to a preset threshold value, if yes, it indicates that the production efficiency in the component production process has an influence on the quality in the component production process, if not, it indicates that the production efficiency in the component production process has no influence on the quality in the component production process;

[0044] the acquisition process of the influence interference factor is:

[0045] the switching time abnormality coefficient and the quality abnormality coefficient in the production process of each component are collected;

[0046] the switching time abnormality coefficient and the quality abnormality coefficient are taken as input features, and the quality inspection pass rate in the production process of the component is taken as a target variable;

[0047] The switching time abnormality coefficient and the quality abnormality coefficient are normalized;

[0048] Based on the normalized switching time abnormality coefficient and the quality abnormality coefficient and the target variable quality qualified rate, a regression model is established by a gradient boosting regression algorithm for predicting the quality qualified rate;

[0049] Based on the trained gradient boosting regression model, the production efficiency of each batch component is calculated. The influence interference factor of quality is calculated, and the calculation expression is: ;

[0050] Wherein, The production efficiency of the first component is the production efficiency of the first component, The production efficiency of the first component is the production efficiency of the first component, The actual quality qualified rate of the first component is the actual quality qualified rate of the first component, The actual quality qualified rate of the first component is the actual quality qualified rate of the first component, The actual quality qualified rate of the first component is the actual quality qualified rate of the first component, The actual quality qualified rate of the first component is the actual quality qualified rate of the first component, The actual quality qualified rate of the first component is the actual quality qualified rate of the first component, The actual quality qualified rate of the first component is the actual quality qualified rate of the first component, The actual quality qualified rate of the first component is the actual quality qualified rate of the first component. As a further scheme of the application: the switching time of the process in the production process is dynamically adjusted according to the influence degree, which specifically includes:

[0051] If the influence interference factor is greater than or equal to the preset threshold value, it means that the switching time of the process in the production process has an impact on the quality qualified rate of the component, and the switching time of the process needs to be adjusted;

[0052] According to the specific value of the influence interference factor, the adjustment direction of the switching time of the affected process is determined, and the adjustment amplitude is calculated: for the positive influence factor, the switching time is reduced; for the negative influence factor, the switching time is increased, and the operation process of the automatic equipment is updated, and the new process switching time parameter is set.

[0053] The application has the following beneficial effects:

[0054]

[0055] ​(1) The present application accurately assesses the efficiency of the production process by real-time monitoring and recording the switching time of each process through the data acquisition module, and calculating the switching time abnormality coefficient using the production plan evaluation module. Specifically, the system can identify which process has abnormal fluctuations in switching time, and dynamically adjust the process switching time based on these abnormal situations. For example, when the switching time abnormality coefficient of a certain process is higher than the preset threshold, it indicates that there may be bottlenecks or inefficient operations in that process. At this time, the system will further analyze the specific values of the influencing factors and their positive and negative nature to determine whether to increase or decrease the switching time, and automatically update the relevant parameter settings on the production line, such as the operation process of automated equipment and the worker's job guide. This fine-tuned time management not only reduces unnecessary downtime, but also optimizes resource allocation and balances the workload of each process, thereby significantly improving overall production efficiency. In addition, through deep learning and model prediction of historical data, the system can provide early warning of potential production bottlenecks, allowing management personnel to take preventive measures before problems occur, further improving the overall operational efficiency and response speed of the production line.

[0056] (2) The present application ensures that the quality of each link in the production process can be accurately assessed and effectively controlled by real-time monitoring and analyzing the quality inspection pass rate of each component. The production quality evaluation module calculates the quality abnormality coefficient based on the quality inspection pass rate, thereby accurately determining the quality fluctuation of each component. Once the quality abnormality coefficient of a certain component exceeds the preset threshold, the system will immediately identify the quality problems existing in the production process of that component, and feed back these key information to the comprehensive analysis module. The comprehensive analysis module calculates the quality consistency coefficient of each batch of assembled components to comprehensively evaluate the quality stability of the entire production process. Based on the evaluation results, the influence analysis and adjustment module can dynamically adjust the process switching time to eliminate potential quality problems. For example, when a batch of components with low quality consistency coefficient is found, the system will automatically adjust the switching time of the relevant process to reduce the quality fluctuations caused by improper switching. This intelligent quality control not only improves the quality level of individual components, but also ensures the high consistency between batches of assembled components. In addition, by combining advanced algorithms such as Bayesian theorem and gradient boosting regression model, the system can more accurately predict and prevent the occurrence of quality problems, providing a more scientific and reliable quality management solution for enterprises, helping them achieve the goal of lean production and continuous improvement. BRIEF DESCRIPTION OF DRAWINGS

[0057] The present application will be further described below in conjunction with the accompanying drawings.

[0058] Figure 1 is a flowchart of an intelligent assembled component whole-process management system of the present application. DETAILED DESCRIPTION

[0059] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0060] Please refer to Figure 1 The present application is a kind of intelligent assembly component whole process management system, comprising:

[0061] The data acquisition module acquires the process switching time and the quality inspection pass rate in the production process in real time within the component production monitoring period.

[0062] The production plan evaluation module analyzes the process switching time in the production process within the monitoring period, and evaluates the production efficiency in the component production process according to the process switching time.

[0063] The production quality evaluation module analyzes the quality inspection pass rate in the production process within the monitoring period, and evaluates the quality in the component production process according to the fluctuation degree of the quality pass rate of each component.

[0064] The comprehensive analysis module analyzes the production efficiency in the component production process and the quality in the component production process, and evaluates the quality consistency of the assembly components in each batch in the production process according to the analysis result.

[0065] The influence analysis and adjustment module analyzes the influence degree of the production efficiency in the component production process on the quality in the component production process according to the quality consistency of the assembly components in each batch in the production process, and dynamically adjusts the process switching time in the production process according to the influence degree.

[0066] In the data acquisition module, the data acquisition module acquires the process switching time and the quality inspection pass rate in the production process in real time within the component production monitoring period, specifically comprising:

[0067] During the component production monitoring period, the data acquisition module first integrates with various sensors and automated equipment on the production line to obtain the switching time of each process in real time. These sensors and equipment are deployed at key process nodes and can accurately record the start and end timestamps of each process. When a process is completed and ready to enter the next process, the system automatically captures this moment and transmits these time data to the central database for storage and subsequent analysis. In addition, the system combines production plans and work order information to ensure that the collected data accurately corresponds to each specific production batch and component;

[0068] The data acquisition module is also responsible for real-time monitoring of the quality inspection link, recording the quality inspection pass rate of each component. At each quality detection point, inspection equipment (such as visual inspection systems) will check the components according to pre-set standards and automatically generate quality reports. The report contains information on whether the components pass each quality standard, as well as any existing defects or non-conformities. All this quality data is also transmitted to the central database in real time and associated with the corresponding production batch and process switching time data for subsequent analysis. In this way, the system can fully grasp the efficiency and quality of the production process, providing solid data support for subsequent evaluation and optimization.

[0069] In the production plan evaluation module, the process switching time during the monitoring period is analyzed, and the production efficiency of the component production process is evaluated according to the process switching time, including:

[0070] During the component production monitoring period, the process switching time in the production process is collected in real time, and the overall switching time anomaly coefficient of the production process is calculated according to the fluctuation degree of the switching time of different processes. Determine whether the overall switching anomaly coefficient of the production process is greater than or equal to the preset threshold value. If so, the switching time of the production process is normal, and the production efficiency is normal. If not, the switching time of the production process is abnormal, and the production efficiency is abnormal;

[0071] The acquisition process of the switching time anomaly coefficient is:

[0072] The switching time data of each process in the production process is collected in real time, and each data represents the switching time from the first process to the nth process;

[0073] The collected switching time data is processed for smoothing, including noise removal and difference operation;

[0074] Determine the three parameters of the ARIMA model: autoregressive order p, difference number d and moving average order q, and use historical data to fit the ARIMA model. The expression for calculating the switching time prediction value at each time is: ;

[0075] wherein, denotes the switching time data prediction value at the time point; denotes the collection time point;

[0076] According to the actual switching time data and the prediction value , a residual error is calculated, and the calculation expression is: ;

[0077] wherein, denotes the residual error value at the time point;

[0078] The standard deviation value of the residual error at all time points is calculated, and the residual error value and the standard deviation value of the residual error are calculated and processed to obtain a switching time abnormality coefficient, and the calculation expression is: ;

[0079] wherein, denotes the total number of collection time points, denotes the standard deviation value of the residual error, denotes the switching time abnormality coefficient;

[0080] It should be noted that the switching time abnormality coefficient reflects whether the switching time of the corresponding production process is normal. The greater the value of the switching time abnormality coefficient, the more likely it is that the switching time of the corresponding production process is abnormal.

[0081] In the production quality evaluation module, the quality inspection pass rate in the production process in the monitoring period is analyzed, and the quality of the component production process is evaluated according to the fluctuation degree of the quality pass rate of each component, specifically including:

[0082] In the component production monitoring period, the quality inspection pass rate of each component in the production process is collected in real time, and the quality abnormality coefficient of the corresponding component when the quality is detected is calculated according to the fluctuation degree of the quality pass rate of each component. If the quality abnormality coefficient of each component is greater than or equal to a preset threshold value, it means that the corresponding quality pass rate has abnormal fluctuations, and the production quality of the component is poor. If not, it means that the corresponding quality pass rate does not have abnormal fluctuations, and the production quality of the component is good;

[0083] The acquisition process of the quality abnormality coefficient is:

[0084] In the component production monitoring period, the quality inspection pass rate of each component in the production process is collected in real time according to the time sequence;

[0085] The quality pass rate at the time point and the factors affecting the quality pass rate and the quality detection result at the moment , denotes the collection moment; the posterior probability distribution of the quality pass rate is inferred according to the Bayesian theorem, and the calculation expression is: ;

[0086] wherein, denotes the posterior probability distribution, denotes the likelihood function of the quality detection result, denotes the prior probability of the quality pass rate, denotes a standardization constant;

[0087] the expectation value of the posterior probability distribution of the quality pass rate obtained by Bayesian inference is calculated;

[0088] the quality abnormality coefficient is obtained by calculating the difference between the quality pass rate of each component and the expected pass rate, and the calculation expression is: ;

[0089] wherein, denotes the total number of collection time points, denotes the standard deviation of the quality pass rate, denotes the quality abnormality coefficient, denotes the expectation value of the posterior probability distribution of the quality pass rate obtained according to the Bayesian inference.

[0090] It should be noted that the quality abnormality coefficient reflects whether the component product quality in the production process is abnormal, and the greater the value of the quality abnormality coefficient, the higher the degree of component quality abnormality.

[0091] In the comprehensive analysis module, the production efficiency in the component production process and the quality in the component production process are analyzed, and according to the analysis result, the quality consistency of each batch of assembled components in the production process is evaluated, specifically including:

[0092] The switching time abnormality coefficient and the quality abnormality coefficient in the component production process are obtained, and the switching time abnormality coefficient and the quality abnormality coefficient are normalized and calculated and processed, the quality consistency coefficient is calculated, and is used to evaluate the quality consistency of each batch of assembled components in the production process;

[0093] The calculation process of the consistency coefficient is:

[0094] The switching time abnormality coefficient and the quality abnormality coefficient in the production process of each component are collected;

[0095] The switching time abnormality coefficient and the quality abnormality coefficient are normalized and processed; for each component The Euclidean distance is used to calculate the distance between the component the distance between the components, ; wherein, denotes the Euclidean distance between the components and , and denote the components, denotes the normalized value of the switching time abnormality coefficient of the component , denotes the normalized value of the switching time abnormality coefficient of the component , denotes the normalized value of the quality abnormality coefficient of the component , denotes the normalized value of the quality abnormality coefficient of the component ;

[0096] for each component , the K nearest neighbors are selected, and the average distance between these neighbors and the component is calculated to obtain the quality consistency coefficient, and the calculation expression is: ;

[0097] wherein, denotes the quality consistency coefficient of the component , denotes the number of selected neighbors; the quality consistency of the components assembled in each batch in the production process is evaluated, specifically including:

[0098] the quality consistency coefficients of the components assembled in each batch in the production process are calculated respectively, and the quality consistency coefficients of the components assembled in each batch are compared with a preset threshold value;

[0099] it is judged whether the quality consistency coefficient is greater than or equal to the preset threshold value, if yes, the quality of the components is consistent, if not, the quality of the components is inconsistent.

[0100] It should be noted that the quality consistency coefficient reflects the quality consistency of the components assembled in each batch in the production process, and the greater the value of the quality consistency coefficient, the higher the quality consistency of the corresponding assembled components.

[0101] In the influence analysis and adjustment module, according to the quality consistency of the components assembled in each batch in the production process, the influence degree of the production efficiency of the components in the production process on the quality of the components in the production process is analyzed, and the switching time of the process in the production process is dynamically adjusted according to the influence degree, specifically including:

[0102] Based on the impact of production efficiency on the quality of component production, the interference factor is calculated. It is then determined whether the interference factor is greater than or equal to a preset threshold. If it is, it indicates that the production efficiency affects the quality of component production. If not, it indicates that the production efficiency does not affect the quality of component production.

[0103] The process for obtaining the interference factor is as follows:

[0104] Collect the changeover time anomaly coefficient and quality anomaly coefficient during the production process of each component;

[0105] Using the switching time anomaly coefficient and the quality anomaly coefficient as input features, and collecting the quality inspection pass rate during the component production process as the target variable;

[0106] Normalize the switching time anomaly coefficient and the quality anomaly coefficient;

[0107] Based on the normalized switching time anomaly coefficient and quality anomaly coefficient, as well as the target variable quality pass rate, a regression model is established using the gradient enhancement regression algorithm to predict the quality pass rate.

[0108] Based on the trained gradient-enhanced regression model, the interference factor of production efficiency on quality for each batch of components is calculated, and the calculation expression is as follows: ;in, Indicates the first The impact of production efficiency of individual components on quality is influenced by several factors. Indicates components, Indicates the first The actual quality pass rate of each component The quality pass rate is predicted based on the normalized switching time anomaly coefficient and quality anomaly coefficient. Representation of components The standardized value of the switching time anomaly coefficient. Representation of components The standardized value of the quality anomaly coefficient is determined; it is determined whether the influencing interference factor is greater than or equal to the preset threshold. If so, it indicates that the process changeover time in the corresponding production process affects the quality pass rate of the component, and the process changeover time needs to be adjusted; based on the specific value of the influencing interference factor, the direction of the changeover time adjustment for the affected process is determined, and the adjustment range is calculated: for positive influencing factors, the changeover time is reduced; for negative influencing factors, the changeover time is increased, and the operation process of the automated equipment is updated, and new process changeover time parameters are set.

[0109] The working principle of the present application: the system includes a data acquisition module, a production plan evaluation module, a production quality evaluation module, a comprehensive analysis module and an influence analysis and adjustment module. The data acquisition module integrates sensors and automated equipment on the production line, real-time acquires and records the switching time of each process and the quality inspection pass rate of each component, and transmits these data to the central database for storage and association. The production plan evaluation module analyzes the process switching time, calculates the switching time abnormality coefficient to evaluate the production efficiency; the production quality evaluation module calculates the quality abnormality coefficient according to the quality inspection pass rate to evaluate the product quality fluctuation. The comprehensive analysis module calculates the quality consistency coefficient by normalizing the switching time abnormality coefficient and the quality abnormality coefficient, and evaluates the quality consistency of each batch of assembled components. The influence analysis and adjustment module calculates the influence interference factor based on the gradient boosting regression model, evaluates the influence degree of production efficiency on quality, and dynamically adjusts the process switching time accordingly. Specifically, when the influence interference factor exceeds the preset threshold, the system will determine the adjustment direction (increase or decrease the switching time) according to its positive and negative nature, and realize parameter adjustment by updating the operation process of the automated equipment, so as to ensure that the production process runs efficiently while maintaining high quality standards. Through fine data management and intelligent adjustment strategy, the system realizes the dual optimization of production efficiency and product quality.

[0110] The above formulas are dimensionless values calculated, and the formulas are obtained by collecting a large amount of data to simulate the most recent real situation, and the preset parameters in the formula are set by a person skilled in the art according to the actual situation.

[0111] The above-described embodiments can be implemented in part or in whole through software, hardware, firmware or any combination thereof. When implemented in software, the above-described embodiments can be implemented in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded into a computer, all or part of the processes described in the embodiments of the present application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions can be transferred from one website, computer, server or data center to another website, computer, server or data center through wired or wireless (such as infrared, wireless, microwave, etc.) manner. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. containing one or more available medium collections. The available medium can be a magnetic medium (such as a floppy disk, a hard disk, a magnetic tape), an optical medium (such as a DVD), or a semiconductor medium. The semiconductor medium can be a solid state disk.

[0112] It should be understood that the term "and / or" herein merely describes the association relationship of the associated objects, which means that there can be three relationships, for example, A and / or B, which can represent three cases of A alone, A and B together, and B alone, where A and B can be singular or plural. In addition, the character " / " herein generally represents that the associated objects before and after it are in an "or" relationship, but it can also represent an "and / or" relationship, which can be understood according to the context before and after it.

[0113] It should be understood that in various embodiments of the present application, the size of the sequence number of the above-mentioned processes does not mean the order of execution, and the execution order of the processes should be determined according to its function and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0114] The above describes one embodiment of the present application in detail, but the content described is only the preferred embodiment of the present application, and cannot be considered as limiting the scope of the implementation of the present application. Any equivalent changes and improvements made within the scope of the present application are still within the scope of the patent coverage of the present application.

Claims

1. An intelligent assembly component lifecycle management system, characterized by, The application relates to a production process monitoring method for assembly components, which comprises the following steps: a data acquisition module is used for collecting process switching time and quality inspection pass rate in a production process in real time in a component production monitoring period; a production plan evaluation module is used for analyzing process switching time in the production process in the monitoring period, and evaluating production efficiency in the component production process according to the process switching time; a production quality evaluation module is used for analyzing quality inspection pass rate in the production process in the monitoring period, and evaluating quality in the component production process according to the fluctuation degree of the quality pass rate of each component; a comprehensive analysis module is used for analyzing production efficiency in the component production process and quality in the component production process, and evaluating quality consistency of each batch of assembly components in the production process according to an analysis result; an influence analysis and adjustment module is used for analyzing the influence degree of production efficiency in the component production process on quality in the component production process according to the quality consistency of each batch of assembly components in the production process, and dynamically adjusting process switching time in the production process according to the influence degree; the analysis of the influence degree of production efficiency in the component production process on quality in the component production process specifically comprises the following steps: an influence interference factor is calculated according to the influence of production efficiency in the component production process on quality in the component production process, and it is judged whether the influence interference factor in the component production process is greater than or equal to a preset threshold value; if yes, it is indicated that production efficiency in the component production process has an influence on quality in the component production process, and if no, it is indicated that production efficiency in the component production process has no influence on quality in the component production process; the acquisition process of the influence interference factor is as follows: switching time abnormality coefficients and quality abnormality coefficients in the production process of each component are collected; the switching time abnormality coefficients and the quality abnormality coefficients are taken as input characteristics, and quality inspection pass rate in the production process of the component is taken as a target variable; the switching time abnormality coefficients and the quality abnormality coefficients are normalized; a regression model is established by using a gradient boosting regression algorithm based on the normalized switching time abnormality coefficients and the quality abnormality coefficients and the target variable of the quality pass rate, and the regression model is used for predicting the quality pass rate; an influence interference factor of production efficiency of each batch of components on quality is calculated based on the trained gradient boosting regression model, and the calculation expression is as follows: ； wherein, represents the production efficiency of the member, represents the member, represents the actual quality pass rate of the member, is the predicted quality pass rate based on the normalized switching time abnormality coefficient and the quality abnormality coefficient, represents the member represents the member ​​​​ 2. The intelligent assembly component lifecycle management system of claim 1, wherein, the evaluation of production efficiency in the component production process specifically comprises the following steps: in the component production monitoring period, process switching time in the production process is collected in real time, a switching time abnormality coefficient of the whole production process is calculated according to the fluctuation degree of different process switching time, and it is judged whether the switching abnormality coefficient of the whole production process is greater than or equal to a preset threshold value; if yes, it is indicated that the switching time of the production process is normal, and then the production efficiency is normal, and if no, it is indicated that the switching time of the production process is abnormal, and then the production efficiency is abnormal.

3. The intelligent assembly component lifecycle management system of claim 2, wherein, the acquisition process of the switching time abnormality coefficient is as follows: switching time data of each process in the production process is collected in real time, and each data represents switching time of the first process to the nth process; The collected switching time data is smoothed, including removing noise and differential operation; Three parameters of the ARIMA model are determined: the autoregressive order p, the difference order d and the moving average order q, and the historical data is used for ARIMA model fitting, and the expression of the switching time prediction value at each time is calculated ; wherein, represents the switching time data prediction value at the th time, represents the collection time; according to the difference between the actual switching time data and the prediction value , the residual is calculated, and the calculation expression is: ; wherein, represents the residual value at the th time; the residual standard deviation value of all times is calculated, and the residual value and the residual standard deviation value are calculated and processed to obtain the switching time abnormality coefficient, and the calculation expression is: ; wherein, represents the total number of collection time points, represents the residual standard deviation value, represents the switching time abnormality coefficient.

4. The intelligent assembly component lifecycle management system of claim 1, wherein, The evaluation component quality in the production process, specifically including: In the component production monitoring period, the quality inspection pass rate of each component in the production process is collected in real time, the quality anomaly coefficient of the corresponding component when the quality is detected is calculated according to the fluctuation degree of the quality pass rate of each component, and it is judged whether the quality anomaly coefficient of each component is greater than or equal to the preset threshold value, if yes, it means that the corresponding quality pass rate has abnormal fluctuation, then the production quality of the component is poor, if not, it means that the corresponding quality pass rate has no abnormal fluctuation, then the production quality of the component is good.

5. The intelligent assembly component lifecycle management system of claim 4, wherein, The acquisition process of the quality anomaly coefficient is: In the component production monitoring period, the quality inspection pass rate of each component in the production process is collected in real time according to the time sequence; Get the Quality pass rate at any time Factors affecting the quality pass rate and the Quality inspection results at any time , Indicates the time of data collection; The posterior probability distribution of the quality pass rate is inferred according to Bayes theorem, and the calculation expression is: ； wherein, represents a posterior probability distribution, represents a likelihood function of the quality detection result, represents a prior probability of the quality pass rate, represents a normalization constant; The expected value of the posterior probability distribution of the quality pass rate obtained by Bayes inference is calculated; The quality abnormality coefficient is obtained by calculating the difference between the quality qualification rate of each component and the expected qualification rate, and the calculation expression is: ; wherein, denotes the total number of collection time points, denotes the standard deviation of the quality pass rate, denotes the quality abnormality coefficient, denotes the expected value of the posterior probability distribution of the quality pass rate according to Bayesian inference.

6. The intelligent assembly component lifecycle management system of claim 1, wherein, The analysis of the production efficiency in the component production process and the quality in the component production process, specifically including: The switching time anomaly coefficient and the quality anomaly coefficient in the component production process are obtained, the switching time anomaly coefficient and the quality anomaly coefficient are normalized and calculated, the quality consistency coefficient is calculated, and the quality consistency of each batch of assembled components in the production process is evaluated; The calculation process of the consistency coefficient is: The switching time anomaly coefficient and the quality anomaly coefficient in the production process of each component are collected; The switching time anomaly coefficient and the quality anomaly coefficient are normalized; For each component The distance between the component and the point is calculated using the Euclidean distance, the expression being: For each component ； wherein denotes a component and a component between the Euclidean distances, and denote components, denotes a component of the switching time anomaly coefficient, denotes a component of the switching time anomaly coefficient, denotes a component of the quality anomaly coefficient, denotes a component of the quality anomaly coefficient; For each component , the K nearest neighbors are selected and the average distance between these neighbors and the component is calculated, resulting in a quality consistency coefficient, calculated as: ; wherein, a quality consistency coefficient of the component, a quality consistency coefficient of the component, denotes the number of selected neighbors.

7. The intelligent assembly component lifecycle management system of claim 6, wherein, The evaluation of the quality consistency of each batch of assembled components in the production process, specifically including: The quality consistency coefficient of each batch of assembled components in the production process is calculated respectively, and the quality consistency coefficient of each batch of assembled components is compared with the preset threshold value; It is judged whether the quality consistency coefficient is greater than or equal to the preset threshold value, if yes, the component quality is consistent, if not, the component quality is inconsistent.

8. The intelligent assembly component lifecycle management system of claim 1, wherein, The switching time of the process in the production process is dynamically adjusted according to the influence degree, specifically including: It is judged whether the influence interference factor is greater than or equal to the preset threshold value, if yes, it means that the switching time of the process in the production process has an impact on the quality pass rate of the component, and the switching time of the process is adjusted; According to the specific value of the influence interference factor, the adjustment direction of the switching time of the affected process is determined, the adjustment amplitude is calculated: for the positive influence factor, the switching time is reduced; for the negative influence factor, the switching time is increased, and the operation process of the automatic equipment is updated, and the new process switching time parameter is set.

Images