Process mass production capability collaborative evaluation method and system

Abstract

The invention discloses a process mass production capability collaborative evaluation method and system, and belongs to the technical field of process management. The method comprises the following steps: collecting parameter data of each process link in each time period in real time, identifying and analyzing process cycle characteristics and material flow characteristics in the parameter data, evaluating the influence effect of each process link on the production capacity in each time period, carrying out grade division on the influence effect, judging the equipment operation state in the high-influence process link, and judging the equipment operation state in the high-influence process link. Whether hidden defects occur or not is determined, and prediction and early warning are carried out in advance; according to the method, general influencing process links are further analyzed, a potential optimization space is identified according to an analysis result, and parameter setting and process steps of the potential optimization space are dynamically adjusted, so that the overall production capacity and efficiency are improved, the efficiency and productivity of the production process are comprehensively improved, the downtime and resource waste are reduced, and the production cost is reduced. The high-quality mass production capability is ensured, the bottleneck and optimization space in the production process can be comprehensively reflected, and potential production problems are effectively prevented and solved.

Description

Technical Field

[0001] The present invention relates to the technical field of process management, and in particular to a collaborative evaluation method and system for process mass production capability. Background Art

[0002] Collaborative evaluation of process mass production capability refers to the comprehensive analysis and evaluation of the synergy of multiple process links during the production process to accurately measure the capacity of the entire production line. This evaluation method not only focuses on the performance of a single process parameter, but also considers the mutual influence and synergy between various process links, thereby providing a comprehensive capacity evaluation result. Through collaborative evaluation of process mass production capability, companies can identify potential bottlenecks and optimization space in the production process and develop more effective production management and optimization strategies. This method helps to improve production efficiency, ensure product quality, and reduce production costs, ultimately enhancing the company's competitiveness in the market.

[0003] The prior art has the following deficiencies:

[0004] Existing process mass production capacity assessment methods usually rely on the analysis of a single process parameter and lack consideration of the synergistic effects of multiple process links, resulting in inaccurate assessment results that cannot fully reflect the bottlenecks and optimization space in the production process. At the same time, if the synergistic effects between multiple process links cannot be accurately identified, hidden defects may exist in production. Hidden defects may cause equipment to gradually damage in long-term operation, which is difficult to detect and prevent in advance, and may have a profound negative impact on the production process and product quality, which is difficult to resolve in the short term.

[0005] Synergy refers to the fact that when multiple factors or components work together, their total effect is greater than the sum of their individual effects. In the production process, synergy means that the interaction between various process links can improve overall production efficiency and product quality, far exceeding the effect that can be achieved by optimizing each link separately. Summary of the invention

[0006] The purpose of the present invention is to provide a collaborative evaluation method and system for process mass production capability to address the deficiencies in the background technology.

[0007] In order to achieve the above object, the present invention provides the following technical solution: a collaborative evaluation method for process mass production capability, comprising the following steps:

[0008] S1: Divide each process step in the production process into several monitoring time periods, and collect parameter data of each process step in each time period in real time through sensors and data acquisition systems;

[0009] S2: Extract features from parameter data of each process link in each time period, identify and analyze process cycle features and material flow features in the parameter data, and evaluate the impact of each process link on production capacity in each time period;

[0010] S3: Based on the evaluation results, the process links are divided into high-impact process links, general-impact process links and low-impact process links according to their impact on production capacity in each time period, and the parameter data in the corresponding process links are divided into high-impact parameters, general-impact parameters and low-impact parameters;

[0011] S4: For high-impact process links, judge the operating status of the equipment in the high-impact process links, determine whether there are hidden defects, and make early predictions and warnings when hidden defects occur;

[0012] S5: Further analyze the effects of each process link on production capacity in each time period of the general process link, identify potential optimization space based on the analysis results, and dynamically adjust its parameter settings and process steps to improve overall production capacity and efficiency.

[0013] In a preferred embodiment, in S2, the process cycle characteristics and material flow characteristics in the parameter data are identified and analyzed, and the impact of each process link on the production capacity in each time period is evaluated, wherein an operation cycle fluctuation index is generated according to the process cycle characteristics in the parameter data, and a flow efficiency balance index of the process material is generated according to the material flow characteristics.

[0014] In a preferred embodiment, the method for obtaining the operation cycle fluctuation index is:

[0015] The time data of all equipment operations in each process link are acquired in real time, and the acquired operation time data is used to construct a time series, and the time series data is denoised; wavelet basis functions are selected for discrete wavelet transform, and the time series data is decomposed into wavelet coefficients of different scales. The wavelet coefficients at each scale are extracted, and their fluctuation characteristics at different frequencies are analyzed. The energy of the wavelet coefficients at each scale is calculated, where the energy formula is: = ; In the formula, is the energy of the jth scale, is the i-th wavelet coefficient of the j-th scale, is the number of wavelet coefficients of the jth scale. The mean and standard deviation of the energy of each scale are calculated respectively, and the mean and standard deviation of the energy of each scale are analyzed. The ratio of the standard deviation of the energy of each scale divided by its mean is calculated, that is, the operation cycle fluctuation index is calculated.

[0016] In a preferred embodiment, the method for obtaining the flow efficiency balance index is:

[0017] The material flow data is standardized so that each feature has the same scale, and the covariance matrix of the standardized data is calculated. The specific calculation expression is: C = Z; where C is the covariance matrix, n is the number of samples, Z is the standardized matrix, and T is the matrix transpose; the covariance matrix is ​​decomposed by eigenvalue to obtain eigenvalues ​​and eigenvectors. The eigenvalues ​​reflect the distribution of the data in the direction of the corresponding eigenvector. ;in: is the i-th eigenvalue of the covariance matrix, is the i-th eigenvector of the covariance matrix. For the largest eigenvector, the standardized data is projected into the new coordinate system and the principal component score of each sample is calculated. ;in, is the score of the ith principal component, vi is the eigenvector of the ith principal component, and the flow efficiency balance index is calculated based on the scores of the principal components to reflect the material flow efficiency and balance. The specific calculation expression is: ; where k is the maximum number of components selected, is the flow efficiency balance index.

[0018] In a preferred embodiment, the operation cycle fluctuation index and the flow efficiency balance index are converted into a first eigenvector, and the first eigenvector is used as the input of the machine learning model. The machine learning model uses each group of first eigenvectors to predict the impact effect value label of the production capacity of each process link in each time period as the prediction target, and takes minimizing the sum of prediction errors of the impact effect value labels of the production capacity of each process link in all time periods as the training target. The machine learning model is trained until the sum of prediction errors converges and the model training is stopped. The impact effect value of the production capacity of each process link in each time period is determined according to the model output results, wherein the machine learning model is a polynomial regression model.

[0019] In a preferred embodiment, in S3, the obtained impact effect value of the production capacity of each process link in each time period is compared with the gradient standard threshold, the gradient standard threshold includes a first standard threshold and a second standard threshold, and the first standard threshold is less than the second standard threshold, and the impact effect value of the production capacity of each process link in each time period is compared with the first standard threshold and the second standard threshold respectively;

[0020] If the impact effect value of the production capacity of the process link is greater than the second standard threshold, the corresponding process link is classified as a high-impact process link, and the parameter data in the corresponding process link is classified as a high-impact parameter;

[0021] If the impact effect value of the production capacity of the process link is greater than or equal to the first standard threshold and less than or equal to the second standard threshold, the corresponding process link is classified as a general impact process link, and the parameter data in the corresponding process link is classified as a high impact parameter;

[0022] If the impact effect value of the production capacity of the process link is less than the first standard threshold, the corresponding process link is classified as a low-impact process link, and the parameter data in the corresponding process link is classified as a high-impact parameter.

[0023] In a preferred embodiment, the random forest model is used to judge the operating status of the equipment in the high-impact process link to determine whether there are hidden defects, specifically:

[0024] When the corresponding process link is classified as a high-impact process link, analyze the high-impact parameters in the high-impact process link to obtain the operating data of the equipment, including temperature, pressure, vibration and speed;

[0025] Remove missing values ​​and outliers and perform normalization so that each parameter has the same dimension, and calculate the statistical characteristics of each parameter;

[0026] Combine the extracted features into a feature vector as the input of the model;

[0027] Build a random forest model using the training data;

[0028] Train the random forest model through training data and adjust the model parameters;

[0029] Use the trained random forest model to predict new data and use the model output as the predicted value of the equipment status.

[0030] In a preferred embodiment, the acquired equipment state prediction value is compared with the preset equipment state standard value. If the equipment state prediction value is greater than or equal to the equipment state standard value, the equipment operation state in the high-impact process link is a normal operation state, and no warning signal is generated at this time; if the equipment state prediction value is less than the equipment state standard value, the equipment operation state in the high-impact process link is an abnormal operation state, and a warning signal is generated at this time.

[0031] In a preferred embodiment, when the corresponding process link is divided into a general influencing process link, that is, the influencing effect value generated in a fixed time period is greater than or equal to the first standard threshold and less than or equal to the second standard threshold, the influencing effect values ​​generated in the subsequent fixed time period are collected, and a corresponding data set is established, and the mean and standard deviation of the data set are calculated, and the mean and standard deviation of the data set are analyzed to determine whether it is necessary to optimize the general influencing process link;

[0032] If the mean of the impact effect values ​​in the data set is greater than or equal to the reference threshold of the impact effect value mean, and the standard deviation of the impact effect value is less than the reference threshold of the standard deviation of the impact effect value, the operation status of the process link is stable, the impact on production capacity is positive, and no further optimization is required;

[0033] If the mean of the impact effect value is greater than or equal to the reference threshold of the mean of the impact effect value, and the standard deviation of the impact effect value is greater than or equal to the reference threshold of the standard deviation of the impact effect value, it means that the operating status of the process link has a positive impact on the production capacity, but there is fluctuation. Optimization is performed to reduce fluctuations and improve stability;

[0034] If the mean of the influencing effect value is less than the reference threshold of the mean of the influencing effect value, and the standard deviation of the influencing effect value is greater than or equal to the reference threshold of the standard deviation of the influencing effect value, the operation state of the process link is unstable and there is fluctuation, and it is comprehensively optimized to improve production capacity and reduce fluctuation;

[0035] If the mean of the influencing effect values ​​is less than the reference threshold of the mean of the influencing effect values, and the standard deviation of the influencing effect values ​​is less than the reference threshold of the standard deviation of the influencing effect values, the operating state of the process link is unstable and has small fluctuations, but has little impact on production capacity. Optimization is performed to increase its impact on production capacity.

[0036] The present invention also provides a collaborative evaluation system for process mass production capability, including a process link parameter acquisition module, a production impact analysis module, an impact effect classification module, a defect warning module and an optimization adjustment module;

[0037] Process link parameter acquisition module: divides each process link in the production process into several monitoring time periods, and collects parameter data of each process link in each time period in real time through sensors and data acquisition systems;

[0038] Production impact analysis module: extracts features from parameter data of each process link in each time period, identifies and analyzes process cycle features and material flow features in parameter data, and evaluates the impact of each process link on production capacity in each time period;

[0039] Impact effect classification module: Based on the evaluation results, each process link is divided into high-impact process links, general-impact process links and low-impact process links according to the impact of each process link on production capacity in each time period, and the parameter data in the corresponding process link is divided into high-impact parameters, general-impact parameters and low-impact parameters;

[0040] Defect warning module: For high-impact process links, judge the operating status of the equipment in the high-impact process links, determine whether there are hidden defects, and make early predictions and warnings when hidden defects occur;

[0041] Optimization and adjustment module: further analyze the impact of each process link on production capacity in each time period of the general process link, identify potential optimization space based on the analysis results, and dynamically adjust its parameter settings and process steps to improve overall production capacity and efficiency.

[0042] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0043] 1. The present invention comprehensively analyzes and evaluates the parameter data of each process link in each time period by real-time monitoring and data collection of each process link in the production process. Through feature extraction and evaluation, high-impact, general-impact and low-impact process links are accurately divided, parameter settings and process steps are dynamically optimized, and hidden defects of equipment are predicted and warned in advance. The method and system can improve the overall efficiency and stability of the production process, reduce downtime and resource waste, and ensure high-quality mass production capabilities.

[0044] 2. The present invention makes up for the deficiency of the existing technology in relying on a single process parameter by introducing the evaluation of synergistic effects, and realizes the coordinated optimization among multiple process links. Through the data-driven machine learning model, it provides intelligent and automated decision support, making production management more scientific and accurate, effectively improving production capacity and product quality, reducing potential risks caused by hidden defects in equipment, and ensuring the continuity and reliability of the production process. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0046] Figure 1 The figure is a flow chart of the method of the present invention.

[0047] Figure 2 It is a system module diagram of the present invention. DETAILED DESCRIPTION

[0048] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0049] Example 1

[0050] See also Figure 1 As shown, a collaborative evaluation method for process mass production capability described in this embodiment includes the following steps:

[0051] S1: Divide each process step in the production process into several monitoring time periods, and collect parameter data of each process step in each time period in real time through sensors and data acquisition systems;

[0052] S2: Extract features from parameter data of each process link in each time period, identify and analyze process cycle features and material flow features in the parameter data, and evaluate the impact of each process link on production capacity in each time period;

[0053] S3: Based on the evaluation results, the process links are divided into high-impact process links, general-impact process links and low-impact process links according to their impact on production capacity in each time period, and the parameter data in the corresponding process links are divided into high-impact parameters, general-impact parameters and low-impact parameters;

[0054] S4: For high-impact process links, judge the operating status of the equipment in the high-impact process links, determine whether there are hidden defects, and make early predictions and warnings when hidden defects occur;

[0055] S5: Further analyze the effects of each process link on production capacity in each time period of the general process link, identify potential optimization space based on the analysis results, and dynamically adjust its parameter settings and process steps to improve overall production capacity and efficiency.

[0056] Among them, in S1, each process link in the production process is divided into several monitoring time periods, and the parameter data of each process link in each time period is collected in real time through sensors and data acquisition systems, specifically:

[0057] The entire production process is divided into several time periods according to certain time intervals. For example, every hour, every minute, or even every second can be used as a time period.

[0058] By refining the time period, the operating status and parameter changes of each process link at different time points can be monitored and analyzed more carefully, so as to discover potential problems and optimization space.

[0059] Determine the time interval: According to the characteristics of the production process and monitoring requirements, select the appropriate time interval for division. For example, for a high-speed production line, you can choose a time interval of seconds; for a slow production line, you can choose a time interval of minutes or hours.

[0060] Time synchronization: Ensure the time synchronization of all monitoring devices to ensure consistency and accuracy of data collection.

[0061] Sensors are installed in each process link of the production line, and parameter data in each time period is collected in real time through the data acquisition system.

[0062] Real-time collection of parameter data can accurately reflect the dynamic changes of the production process, provide detailed data support, and provide a basis for subsequent analysis and decision-making.

[0063] Sensor installation: Install appropriate types of sensors at key locations in each process link, such as temperature sensors, pressure sensors, speed sensors, flow sensors, etc.

[0064] Data acquisition system: Using Industrial Internet of Things (IIoT) technology, various sensors are connected to the data acquisition system to achieve real-time data collection, transmission and storage.

[0065] Real-time monitoring: Through the real-time monitoring function of the data acquisition system, the operating status and parameter changes of each process link can be mastered at any time, and abnormal situations can be discovered in time.

[0066] S2: Extract features from the parameter data of each process link in each time period, identify and analyze the process cycle characteristics and material flow characteristics in the parameter data, and evaluate the impact of each process link on production capacity in each time period.

[0067] Feature extraction of parameter data of each process link in each time period is to extract representative indicators or characteristic values ​​from the original data. Through feature extraction, massive raw data can be simplified into key indicators, helping to better understand the important laws and characteristics of the production process.

[0068] Statistical features: Calculate the average value, maximum value, minimum value, standard deviation and other statistical indicators of parameter data in each time period. For example, the average temperature and maximum pressure in a certain time period.

[0069] Time characteristics: Extract the characteristics of parameters changing over time, such as rate of change, trend, periodicity, etc. For example, analyze the rising or falling trend of temperature in different time periods.

[0070] Frequency characteristics: Extract frequency components from data through spectrum analysis to identify periodic and fluctuation characteristics of data. For example, analyze the main frequency components in vibration data through Fourier transform.

[0071] Identify and analyze the process cycle characteristics and material flow characteristics in the parameter data, and evaluate the impact of each process link on the production capacity in each time period, wherein an operation cycle fluctuation index is generated according to the process cycle characteristics in the parameter data, and a process material flow efficiency balance index is generated according to the material flow characteristics;

[0072] Among them, the method for obtaining the operation cycle fluctuation index is:

[0073] The time data of all equipment operations in each process link are acquired in real time, and the acquired operation time data is used to construct a time series, and the time series data is denoised; wavelet basis functions (such as Daubechies wavelet, Haar wavelet, etc.) are selected for discrete wavelet transform, and the time series data is decomposed into wavelet coefficients of different scales (frequencies), and the wavelet coefficients at each scale are extracted, and the fluctuation characteristics at different frequencies are analyzed, and the energy of the wavelet coefficients at each scale is calculated. The energy reflects the intensity of the signal at the frequency, where the energy formula is: = ; In the formula, is the energy of the jth scale, is the i-th wavelet coefficient of the j-th scale, is the number of wavelet coefficients of the jth scale. The mean and standard deviation of the energy of each scale are calculated respectively, and the mean and standard deviation of the energy of each scale are analyzed. The ratio of the standard deviation of the energy of each scale divided by its mean is calculated, that is, the operation cycle fluctuation index is calculated.

[0074] The larger the operation cycle fluctuation index, the greater the fluctuation of the operation cycle of each process link in each time period, and the lower the stability of the production process. This means that the operating status of the process link varies greatly in different time periods, which may make it difficult for the production line to maintain a consistent rhythm and efficiency, and ultimately affect the overall production capacity.

[0075] High volatility increases the probability of abnormal situations in the production process, such as equipment failure, product quality problems, or production stagnation. These abnormal situations will lead to reduced operating efficiency of the production line, increased production costs, and may require additional resources for adjustment and repair, further affecting production capacity and the execution of production plans.

[0076] In addition, a larger operation cycle fluctuation index indicates that the synergy between process links is poor, making it difficult to achieve an optimized production process. The instability of the production link will lead to waste of resources and the emergence of production bottlenecks, making it difficult to effectively improve the overall production capacity and product quality of the production line. Therefore, reducing the operation cycle fluctuation index and maintaining the stable operation of each process link are of great significance to improving production capacity.

[0077] The method for obtaining the flow efficiency balance index is:

[0078] The material flow data is standardized so that each feature has the same scale. The covariance matrix of the standardized data is calculated to reflect the linear relationship between the features. The specific calculation expression is: C = Z; where C is the covariance matrix, n is the number of samples, Z is the standardized matrix, and T is the matrix transpose; perform eigenvalue decomposition on the covariance matrix to obtain eigenvalues ​​and eigenvectors. The eigenvalue reflects the distribution of the data in the direction of the corresponding eigenvector. ;in: is the i-th eigenvalue of the covariance matrix, is the i-th eigenvector of the covariance matrix. For the largest eigenvector, the standardized data is projected into the new coordinate system and the principal component score of each sample is calculated. ;in, is the score of the ith principal component, vi is the eigenvector of the ith principal component, and the flow efficiency balance index is calculated based on the scores of the principal components to reflect the material flow efficiency and balance. The specific calculation expression is: ; where k is the maximum number of components selected, is the flow efficiency balance index.

[0079] The larger the flow efficiency balance index is, the higher the flow efficiency and balance of materials between different process links in each time period is. This indicates that the connection between each link in the production process is smoother, the material transmission speed between each process link is faster, and the path is more stable, reducing waiting time and the possibility of interruption.

[0080] A high flow efficiency balance index means that the overall operation efficiency of the production line is higher, the resources of each process link are more fully utilized, and the production rhythm is more coordinated. In this case, the production process can better maintain continuity and consistency, reduce bottlenecks and resource waste, and thus improve production capacity and output.

[0081] Therefore, the larger the flow efficiency balance index is, the more efficient and balanced the material flow in the production process is, and the more positive the impact on production capacity is, which helps to improve production efficiency, reduce costs, and improve product quality and delivery speed.

[0082] The operation cycle fluctuation index and the flow efficiency balance index are converted into the first eigenvector, and the first eigenvector is used as the input of the machine learning model. The machine learning model uses each group of first eigenvectors to predict the impact effect value label of the production capacity of each process link in each time period as the prediction target, and takes minimizing the sum of prediction errors of the impact effect value labels of the production capacity of each process link in all time periods as the training target. The machine learning model is trained until the sum of prediction errors converges and the model training is stopped. The impact effect value of the production capacity of each process link in each time period is determined according to the model output results, wherein the machine learning model is a polynomial regression model.

[0083] The method for obtaining the impact effect value of the production capacity of each process link in each time period is: from the first feature vector training data of the trained machine learning model, obtain the corresponding function expression: ; In the formula, is the output function of the model, is the operating cycle volatility index, is the flow efficiency balance index, and FK is the effect value of each process link on production capacity in each time period.

[0084] S3: Based on the evaluation results, each process link is divided into high-impact process links, general-impact process links and low-impact process links according to the impact of each process link on production capacity in each time period, and the parameter data in the corresponding process link is divided into high-impact parameters, general-impact parameters and low-impact parameters.

[0085] The obtained impact effect value of the production capacity of each process link in each time period is compared with the gradient standard threshold, the gradient standard threshold includes a first standard threshold and a second standard threshold, and the first standard threshold is less than the second standard threshold, and the impact effect value of the production capacity of each process link in each time period is compared with the first standard threshold and the second standard threshold respectively;

[0086] If the impact effect value of the production capacity of the process link is greater than the second standard threshold, the corresponding process link is classified as a high-impact process link, and the parameter data in the corresponding process link is classified as a high-impact parameter;

[0087] If the impact effect value of the production capacity of the process link is greater than or equal to the first standard threshold and less than or equal to the second standard threshold, the corresponding process link is classified as a general impact process link, and the parameter data in the corresponding process link is classified as a high impact parameter;

[0088] If the impact effect value of the production capacity of the process link is less than the first standard threshold, the corresponding process link is classified as a low-impact process link, and the parameter data in the corresponding process link is classified as a high-impact parameter.

[0089] In the present invention, by comparing the impact effect value of each process link with the gradient standard threshold, the impact degree of each process link in the production process can be systematically evaluated and classified. High-impact process links are usually key points in production and may be potential bottlenecks, requiring priority attention and optimization; general-impact process links have a moderate impact on production capacity and can be properly monitored and adjusted; low-impact process links have little impact on overall production capacity and can usually remain the same. This classification method helps to concentrate resources and energy on the links that have the greatest impact on production capacity, thereby improving overall production efficiency and stability.

[0090] S4: For high-impact process links, judge the operating status of the equipment in the high-impact process links, determine whether there are hidden defects, and make early predictions and warnings when hidden defects occur.

[0091] The random forest model is used to judge the operating status of equipment in high-impact process links and determine whether hidden defects occur. Specifically:

[0092] When the corresponding process link is classified as a high-impact process link, the high-impact parameters in the high-impact process link are analyzed to obtain the operating data of the equipment, including temperature, pressure, vibration and rotation speed.

[0093] Remove missing values ​​and outliers to ensure data quality. Perform normalization or standardization so that each parameter has the same dimension.

[0094] Calculate the statistical characteristics of each parameter, such as the mean, maximum, minimum, standard deviation, etc. Extract frequency domain features, such as the spectrum characteristics after Fourier transform.

[0095] The extracted features are combined into a feature vector as the input of the model.

[0096] Use the training data to build a random forest model. The random forest model contains multiple decision trees, each of which is built by random sampling of data and random selection of features.

[0097] Train the random forest model through training data and adjust model parameters such as the number of trees (n_estimators), maximum depth (max_depth), etc.

[0098] Use the trained random forest model to predict new data and use the model output as the predicted value of the equipment status.

[0099] The acquired equipment status prediction value is compared with the preset equipment status standard value. If the equipment status prediction value is greater than or equal to the equipment status standard value, it means that the equipment operation status in the high-impact process link is in normal operation, and the possibility of hidden defects is low. In this case, no warning signal is generated. If the equipment status prediction value is less than the equipment status standard value, it means that the equipment operation status in the high-impact process link is in abnormal operation, and the possibility of hidden defects is high. In this case, a warning signal is generated.

[0100] S5: Further analyze the effects of each process link on production capacity in each time period of the general process link, identify potential optimization space based on the analysis results, and dynamically adjust its parameter settings and process steps to improve overall production capacity and efficiency.

[0101] When the corresponding process link is classified as a general influencing process link, that is, the impact effect value generated within a fixed time period is greater than or equal to the first standard threshold and less than or equal to the second standard threshold, the impact effect values ​​generated in the subsequent fixed time period are collected, and the corresponding data set is established, and the mean and standard deviation of the data set are calculated. The mean and standard deviation of the data set are analyzed to determine whether it is necessary to optimize the general influencing process link.

[0102] If the mean of the effect values ​​in the data set is greater than or equal to the reference threshold of the mean of the effect values, and the standard deviation of the effect values ​​is less than the reference threshold of the standard deviation of the effect values, it means that the operation status of the process link is stable, the impact on production capacity is positive, and no further optimization is required; in this case, the operation effect of the process link is good, the impact on production capacity is positive and the fluctuation is small, indicating that the process parameter settings are relatively reasonable and the process steps are stable. It is necessary to maintain the existing process parameters and steps and continue to monitor to ensure continuous and stable operation.

[0103] If the mean of the effect value is greater than or equal to the reference threshold of the mean of the effect value, and the standard deviation of the effect value is greater than or equal to the reference threshold of the standard deviation of the effect value, it means that although the operating status of the process link has a positive impact on production capacity, there are fluctuations, and optimization is needed to reduce fluctuations and improve stability; although the overall performance of the process link is good, the fluctuations are large, which may lead to instability and production inconsistency. Parameter adjustment and process step optimization are needed to reduce fluctuations. Corresponding measures can be taken by analyzing the source of fluctuations (such as equipment instability, operating errors, etc.).

[0104] If the mean of the effect value is less than the reference threshold of the mean of the effect value, and the standard deviation of the effect value is greater than or equal to the reference threshold of the standard deviation of the effect value, it means that the operation status of the process link is not positive and fluctuates greatly, and it needs to be fully optimized to improve production capacity and reduce fluctuations; the overall performance of the process link is not positive and fluctuates greatly, indicating that there are major problems with the process parameter settings or process steps. It is necessary to comprehensively check and optimize the process parameters and steps to reduce fluctuations and improve overall efficiency. Through in-depth analysis of the data, the influencing factors can be found and targeted adjustments can be made.

[0105] If the mean of the effect value is less than the reference threshold of the mean of the effect value, and the standard deviation of the effect value is less than the reference threshold of the standard deviation of the effect value, it means that the operation status of the process link is not positive, and although the fluctuation is small, the impact on production capacity is small, and it needs to be optimized to improve its impact on production capacity. The overall performance of the process link is not ideal, but the fluctuation is small, which means that although the parameter settings or process steps are stable, the improvement of production capacity is limited. It is necessary to re-evaluate and optimize the process parameter settings and steps to find ways to improve production capacity. You can try to introduce new technologies or improve process flows.

[0106] In this embodiment, the parameter data of each process link in each time period is collected in real time through sensors and data acquisition systems; feature extraction is performed on the parameter data of each process link in each time period, the process cycle characteristics and material flow characteristics in the parameter data are identified and analyzed, and the impact of each process link on the production capacity in each time period is evaluated; according to the evaluation results, each process link is divided into high-impact process links, general-impact process links and low-impact process links according to the impact of each process link on the production capacity in each time period, and the parameter data in the corresponding process link is divided into high-impact parameters, general-impact parameters and low-impact parameters; for the high-impact process link, the operating status of the equipment in the high-impact process link is judged to determine whether it has hidden defects, and early prediction and warning are performed when hidden defects occur; the impact of each process link on the production capacity in each time period of the general-impact process link is further analyzed, potential optimization space is identified based on the analysis results, and its parameter settings and process steps are dynamically adjusted to improve the overall production capacity and efficiency. It can comprehensively monitor and analyze each process link in the production process, identify high-impact and general-impact process links and their parameters, predict and warn hidden defects of equipment in advance, and dynamically optimize parameter settings and process steps, thereby improving the overall efficiency, stability and capacity of the production process, reducing downtime and resource waste, and ensuring high-quality mass production capabilities.

[0107] Example 2

[0108] See also Figure 2As shown, a collaborative evaluation system for process mass production capability described in this embodiment includes a process link parameter acquisition module, a production impact analysis module, an impact effect classification module, a defect warning module and an optimization and adjustment module;

[0109] Process link parameter acquisition module: divides each process link in the production process into several monitoring time periods, and collects parameter data of each process link in each time period in real time through sensors and data acquisition systems;

[0110] Production impact analysis module: extracts features from parameter data of each process link in each time period, identifies and analyzes process cycle features and material flow features in parameter data, and evaluates the impact of each process link on production capacity in each time period;

[0111] Impact effect classification module: Based on the evaluation results, each process link is divided into high-impact process links, general-impact process links and low-impact process links according to the impact of each process link on production capacity in each time period, and the parameter data in the corresponding process link is divided into high-impact parameters, general-impact parameters and low-impact parameters;

[0112] Defect warning module: For high-impact process links, judge the operating status of the equipment in the high-impact process links, determine whether there are hidden defects, and make early predictions and warnings when hidden defects occur;

[0113] Optimization and adjustment module: further analyze the impact of each process link on production capacity in each time period of the general process link, identify potential optimization space based on the analysis results, and dynamically adjust its parameter settings and process steps to improve overall production capacity and efficiency.

[0114] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0115] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part 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 or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0116] It should be understood that the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. A and B can be singular or plural. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship, but it may also indicate an "and / or" relationship. Please refer to the context for specific understanding.

[0117] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0118] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0119] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.

Claims

1. A collaborative evaluation method for process mass production capability, characterized in that: The following steps are involved: S1: Divide each process step in the production process into several monitoring time periods, and collect parameter data of each process step in each time period in real time through sensors and data acquisition systems; S2: Extract features from parameter data of each process link in each time period, identify and analyze process cycle features and material flow features in the parameter data, and evaluate the impact of each process link on production capacity in each time period; S3: Based on the evaluation results, the process links are divided into high-impact process links, general-impact process links and low-impact process links according to their impact on production capacity in each time period, and the parameter data in the corresponding process links are divided into high-impact parameters, general-impact parameters and low-impact parameters; S4: For high-impact process links, judge the operating status of the equipment in the high-impact process links, determine whether there are hidden defects, and make early predictions and warnings when hidden defects occur; S5: Further analyze the effects of each process link on production capacity in each time period of the general process link, identify potential optimization space based on the analysis results, and dynamically adjust its parameter settings and process steps to improve overall production capacity and efficiency.

2. A collaborative evaluation method for process mass production capability according to claim 1, characterized in that: In S2, the process cycle characteristics and material flow characteristics in the parameter data are identified and analyzed, and the impact of each process link on production capacity in each time period is evaluated, which also includes: An operation cycle fluctuation index is generated according to the process cycle characteristics in the parameter data, and a flow efficiency balance index of the process material is generated according to the material flow characteristics.

3. A collaborative evaluation method for process mass production capability according to claim 2, characterized in that: The method for obtaining the operating cycle volatility index is: The time data of all equipment operations in each process link are acquired in real time, and the acquired operation time data is used to construct a time series, and the time series data is denoised; wavelet basis functions are selected for discrete wavelet transform, and the time series data is decomposed into wavelet coefficients of different scales. The wavelet coefficients at each scale are extracted, and their fluctuation characteristics at different frequencies are analyzed. The energy of the wavelet coefficients at each scale is calculated, where the energy formula is: = ; In the formula, is the energy of the jth scale, is the i-th wavelet coefficient of the j-th scale, is the number of wavelet coefficients of the jth scale. The mean and standard deviation of the energy of each scale are calculated respectively. The mean and standard deviation of the energy of each scale are analyzed, and the ratio of the standard deviation of the energy of each scale divided by its mean is calculated, that is, the operation cycle fluctuation index is calculated.

4. A collaborative evaluation method for process mass production capability according to claim 3, characterized in that: The method for obtaining the flow efficiency balance index is: The material flow data is standardized so that each feature has the same scale, and the covariance matrix of the standardized data is calculated. The specific calculation expression is: C = Z; where C is the covariance matrix, n is the number of samples, Z is the standardized matrix, and T is the matrix transpose; the covariance matrix is ​​decomposed by eigenvalue to obtain eigenvalues ​​and eigenvectors. The eigenvalues ​​reflect the distribution of the data in the direction of the corresponding eigenvector. ;in: is the i-th eigenvalue of the covariance matrix, is the i-th eigenvector of the covariance matrix. For the largest eigenvector, the standardized data is projected into the new coordinate system and the principal component score of each sample is calculated. ;in, is the score of the ith principal component, vi is the eigenvector of the ith principal component, and the flow efficiency balance index is calculated based on the scores of the principal components to reflect the material flow efficiency and balance. The specific calculation expression is: ; where k is the maximum number of components selected, is the flow efficiency balance index.

5. A collaborative evaluation method for process mass production capability according to claim 4, characterized in that: The operation cycle fluctuation index and the flow efficiency balance index are converted into the first eigenvector, and the first eigenvector is used as the input of the machine learning model. The machine learning model uses each group of first eigenvectors to predict the impact effect value label of the production capacity of each process link in each time period as the prediction target, and takes minimizing the sum of prediction errors of the impact effect value labels of the production capacity of each process link in all time periods as the training target. The machine learning model is trained until the sum of prediction errors converges and the model training is stopped. The impact effect value of the production capacity of each process link in each time period is determined according to the model output results, wherein the machine learning model is a polynomial regression model.

6. A collaborative evaluation method for process mass production capability according to claim 5, characterized in that: In S3, the obtained impact effect value of the production capacity of each process link in each time period is compared with the gradient standard threshold, the gradient standard threshold includes a first standard threshold and a second standard threshold, and the first standard threshold is less than the second standard threshold, and the impact effect value of the production capacity of each process link in each time period is compared with the first standard threshold and the second standard threshold respectively; If the impact effect value of the production capacity of the process link is greater than the second standard threshold, the corresponding process link is classified as a high-impact process link, and the parameter data in the corresponding process link is classified as a high-impact parameter; If the impact effect value of the production capacity of the process link is greater than or equal to the first standard threshold and less than or equal to the second standard threshold, the corresponding process link is classified as a general impact process link, and the parameter data in the corresponding process link is classified as a high impact parameter; If the impact effect value of the production capacity of the process link is less than the first standard threshold, the corresponding process link is classified as a low-impact process link, and the parameter data in the corresponding process link is classified as a high-impact parameter.

7. A collaborative evaluation method for process mass production capability according to claim 6, characterized in that: The random forest model is used to judge the operating status of equipment in high-impact process links and determine whether hidden defects occur. Specifically: When the corresponding process link is classified as a high-impact process link, analyze the high-impact parameters in the high-impact process link to obtain the operating data of the equipment, including temperature, pressure, vibration and speed; Remove missing values ​​and outliers and perform normalization so that each parameter has the same dimension, and calculate the statistical characteristics of each parameter; Combine the extracted features into a feature vector as the input of the model; Build a random forest model using the training data; Train the random forest model through training data and adjust the model parameters; Use the trained random forest model to predict new data and use the model output as the predicted value of the equipment status.

8. A collaborative evaluation method for process mass production capability according to claim 7, characterized in that: The acquired equipment status prediction value is compared with the preset equipment status standard value. If the equipment status prediction value is greater than or equal to the equipment status standard value, the equipment operation status in the high-impact process link is a normal operation state, and no warning signal is generated at this time; if the equipment status prediction value is less than the equipment status standard value, the equipment operation status in the high-impact process link is an abnormal operation state, and a warning signal is generated at this time.

9. A collaborative evaluation method for process mass production capability according to claim 8, characterized in that: When the corresponding process link is classified as a general influencing process link, that is, the influencing effect value generated in a fixed time period is greater than or equal to the first standard threshold and less than or equal to the second standard threshold, the influencing effect values ​​generated in the subsequent fixed time period are collected, and a corresponding data set is established, and the mean and standard deviation of the data set are calculated, and the mean and standard deviation of the data set are analyzed to determine whether it is necessary to optimize the general influencing process link; If the mean of the impact effect values ​​in the data set is greater than or equal to the reference threshold of the impact effect value mean, and the standard deviation of the impact effect value is less than the reference threshold of the standard deviation of the impact effect value, the operation status of the process link is stable, the impact on production capacity is positive, and no further optimization is required; If the mean of the impact effect value is greater than or equal to the reference threshold of the mean of the impact effect value, and the standard deviation of the impact effect value is greater than or equal to the reference threshold of the standard deviation of the impact effect value, it means that the operating status of the process link has a positive impact on the production capacity, but there is fluctuation. Optimization is performed to reduce fluctuations and improve stability; If the mean of the influencing effect value is less than the reference threshold of the mean of the influencing effect value, and the standard deviation of the influencing effect value is greater than or equal to the reference threshold of the standard deviation of the influencing effect value, the operation state of the process link is unstable and there is fluctuation, and it is comprehensively optimized to improve production capacity and reduce fluctuation; If the mean of the influencing effect values ​​is less than the reference threshold of the mean of the influencing effect values, and the standard deviation of the influencing effect values ​​is less than the reference threshold of the standard deviation of the influencing effect values, the operating state of the process link is unstable and has small fluctuations, but has little impact on production capacity. Optimization is performed to increase its impact on production capacity.

10. A process mass production capability collaborative evaluation system, used to implement a process mass production capability collaborative evaluation method according to any one of claims 1 to 9, characterized in that: It includes process parameter acquisition module, production impact analysis module, impact effect classification module, defect warning module and optimization and adjustment module; Process link parameter acquisition module: divides each process link in the production process into several monitoring time periods, and collects parameter data of each process link in each time period in real time through sensors and data acquisition systems; Production impact analysis module: extracts features from parameter data of each process link in each time period, identifies and analyzes process cycle features and material flow features in parameter data, and evaluates the impact of each process link on production capacity in each time period; Impact effect classification module: Based on the evaluation results, each process link is divided into high-impact process links, general-impact process links and low-impact process links according to the impact of each process link on production capacity in each time period, and the parameter data in the corresponding process link is divided into high-impact parameters, general-impact parameters and low-impact parameters; Defect warning module: For high-impact process links, judge the operating status of the equipment in the high-impact process links, determine whether there are hidden defects, and make early predictions and warnings when hidden defects occur; Optimization and adjustment module: further analyze the impact of each process link on production capacity in each time period of the general process link, identify potential optimization space based on the analysis results, and dynamically adjust its parameter settings and process steps to improve overall production capacity and efficiency.

Images