Product quality detection method for intelligent manufacturing production line

Abstract

The invention discloses a product quality detection method for an intelligent manufacturing production line, and belongs to the field of product quality detection.The method comprises the steps that a production process library is constructed, and quality inspection indexes of all production processes are determined; determining the time sequence of each production process, and obtaining the correlation degree between the production processes based on the quality inspection indexes of the production processes; during production, after the mth production process is completed, detecting the current semi-finished product to obtain each quality inspection index value of the current semi-finished product; comparing each quality inspection index value of the current semi-finished product with the standard threshold value of each quality inspection index of the mth production process, when the quality inspection index is unqualified, judging that the current semi-finished product is a failure product, eliminating the failure product from the assembly line, and based on the quality inspection result of the current semi-finished product in the (1-m) th production process, judging that the current semi-finished product is a failure product. And updating the standard threshold value of the related quality inspection index of the previous production process. The problem that the productivity is wasted due to the fact that the quality inspection index threshold value cannot be adaptively adjusted based on specific working conditions in an existing method is solved.

Description

Technical Field

[0001] The present invention belongs to the field of product quality testing, and in particular relates to a product quality testing method for an intelligent manufacturing production line. Background Art

[0002] As the global manufacturing industry accelerates its transformation toward intelligent manufacturing, smart manufacturing production lines are gradually becoming a core driver of industrial upgrading. This production model, which integrates automation, digitalization, and intelligent technologies, has greatly improved production efficiency and flexibility, propelling the manufacturing industry toward a more flexible, efficient, and precise approach. However, while pursuing production efficiency, product quality control faces unprecedented challenges, and traditional quality inspection systems are no longer able to adapt to the development needs of intelligent manufacturing. In the field of quality control in intelligent production lines, traditional quality inspection methods play an important role in ensuring product quality. Currently, most intelligent production lines use fixed quality inspection thresholds for quality control. In each process, specific quality inspection indicator thresholds are pre-set, and the product is tested for various indicators using sensors and other testing equipment. If the test results are within the preset threshold range, the product is judged to have passed the quality inspection of this process and can enter the next process; if the threshold range is exceeded, it is judged to be unqualified. For example, in the patch process of electronic equipment manufacturing, a fixed threshold is set for the position accuracy of component placement. The deviation between the actual component placement position and the standard position is measured by a visual inspection system to determine whether the placement is qualified.

[0003] This existing method based on fixed quality inspection thresholds has certain advantages. On the one hand, its operation is relatively simple and direct, easy to understand and implement. Workers on the production line can quickly master quality inspection standards and processes, reducing training costs and time. In some production scenarios where the process is relatively stable and product quality requirements vary little, fixed thresholds can efficiently complete quality inspection work and ensure the continuity and efficiency of production. On the other hand, fixed thresholds can ensure the consistency of product quality to a certain extent. As long as the production process is stable, quality inspection according to the established thresholds can ensure that the products produced meet the pre-set quality standards and meet the market's requirements for basic product quality.

[0004] However, this approach also has significant drawbacks. First, fixed quality inspection thresholds lack adaptability to the complex and changing factors of the production process. Fluctuations in the production environment, such as changes in temperature and humidity, gradual aging and wear of equipment, and subtle differences between raw material batches, can all lead to changes in product quality characteristics, but fixed thresholds cannot be adjusted accordingly. In the food processing industry, changes in ambient temperature and humidity can affect indicators such as food moisture content. If the quality inspection threshold is fixed, it may result in previously qualified products being mistakenly classified as unqualified or unqualified products going undetected when the environment changes. Second, processes interact with each other. Even if the quality inspection indicators of a current process meet the standards, they may still affect the quality inspection indicators of subsequent processes, resulting in unqualified final products. However, since fixed thresholds cannot be dynamically adjusted based on these inter-process relationships, it is difficult to timely screen out potential unqualified products in the early stages of production, resulting in wasted resources and reduced production efficiency in subsequent processes. Furthermore, fixed thresholds fail to fully utilize the vast amount of data generated during the production process. In the era of intelligent production, production lines are accumulating vast amounts of quality inspection data, but fixed threshold methods fail to deeply mine and analyze this data to optimize and improve quality inspection standards. Summary of the Invention

[0005] In response to the above-mentioned deficiencies in the prior art, the present invention provides a product quality inspection method for an intelligent manufacturing production line, which solves the problem of waste of productivity caused by the inability of the quality inspection index thresholds in the prior art to be adaptively adjusted based on specific working conditions.

[0006] In order to achieve the above-mentioned purpose of the invention, the technical solution adopted by the present invention is: a product quality detection method for an intelligent manufacturing production line, comprising: Based on the product production process, build a production process library and determine the quality inspection indicators of each production process in the production process library; Determine the timing of each production process and obtain the degree of correlation between each production process based on the quality inspection indicators of each production process; Initialize the standard thresholds of each quality inspection indicator for each production process; During production, after the mth production process is completed, the current semi-finished product is tested to obtain the quality inspection index values ​​of the current semi-finished product; The quality inspection index values ​​of the current semi-finished product are compared with the standard threshold values ​​of the quality inspection indicators of the m-th production process. When any quality inspection indicator fails to meet the standards, the current semi-finished product is judged as a failed product and eliminated from the assembly line. Based on the quality inspection results of the current semi-finished product in the 1-m production processes, the standard threshold values ​​of the relevant quality inspection indicators of the previous production process are updated.

[0007] The beneficial effects of the present invention are: by constructing the degree of correlation between each production process, correlating the production quality between each production process, dynamically adjusting the standard threshold of the quality inspection index of each production process, and eliminating semi-finished products with a high probability of failure in the later stage in the early stage, the consumption of the production line can be reduced.

[0008] Furthermore, the degree of correlation between the production processes is obtained as follows: According to the time sequence of each production process, the production process association pair matrix is ​​constructed:

[0009] in, is the production process correlation pair matrix; is the self-correlation item of the first production process; For the first production process Related items of the production process; For the The self-correlation items of the production process; is the total number of production processes, and the production processes are numbered in chronological order; is a zero matrix; Conduct several production line simulations to obtain data sets of quality inspection indicators for each production process; Set the correlation degree of each correlation item in the production process correlation matrix to 1; According to the quality inspection index data set of each production process, calculate the The production process of the first Production process The degree of correlation; among them, and All are production process indexes. .

[0010] The beneficial effect of the above further scheme is: the production process association pair matrix is ​​a triangular matrix, which integrates the time sequence into the description of the production process pair to make a preliminary judgment on the influence relationship, that is, the previous process affects the subsequent process, which facilitates the subsequent solution of the specific influence degree.

[0011] Furthermore, the calculation The degree of correlation is: Get the Quality inspection indicators of the production process:

[0012] in, For the Quality inspection index set for the production process; For the The first quality inspection indicator of the production process; For the The second quality inspection indicator of the production process; For the The first production process Quality inspection indicators; For the The total number of quality inspection indicators of the production process; Get the Quality inspection indicators of the production process:

[0013] in, For the Quality inspection index set for the production process; For the The first quality inspection indicator of the production process; For the The second quality inspection indicator of the production process; For the The first production process Quality inspection indicators; For the The total number of quality inspection indicators of the production process; According to the quality inspection index set and quality inspection indicator set , make judgment on the impact of quality inspection indicators: From the quality inspection index set Select quality inspection indicator B from the quality inspection indicator set Select quality inspection indicator C; Based on the quality inspection indicator data sets of each production process, the change trends of quality inspection indicator B and quality inspection indicator C are depicted in the same line chart, with the production line simulation round as the horizontal axis and the quality inspection indicator value as the vertical axis; If the indicator types of quality inspection indicators B and C are the same, calculate the change convergence of quality inspection indicators B and C:

[0014] in, is the change convergence of quality inspection index B and quality inspection index C; Quality inspection indicator B Simulation to The change trend of the simulation is 1 for rising, 0 for remaining flat, and -1 for falling. Quality inspection indicator C Simulation to The change trend of the simulation is 1 when it increases, 0 when it remains the same, and -1 when it decreases. is the total number of simulations; is the simulation index; A function to determine whether the changing trends of two parameters are the same. If they are the same, it is 1, otherwise it is 0; If the indicator types of quality inspection indicators B and C are different, calculate the change deviation of quality inspection indicators B and C:

[0015] in, is the change deviation degree of quality inspection index B and quality inspection index C; A function to determine whether the changing trends of two parameters are opposite. If they are the same, it is 1, otherwise it is 0; When the change convergence or the change divergence is greater than the change threshold, the The quality inspection index C of the production process is subject to the The quality inspection index B of the first production process is affected, otherwise, the judgment of The quality inspection index C of the production process is not subject to the The quality inspection index B of the production process is affected; Repeated quality inspection indicators affect the judgment process and obtain the quality inspection indicator set Quality inspection indicators and quality inspection indicator sets The influence relationship of each quality inspection index in the Quality inspection indicators and quality inspection indicator sets The influence relationship of each quality inspection index in the construction of quality inspection index set Quality inspection indicator set The impact path; According to the quality inspection index set Quality inspection indicator set The impact path of degree of correlation.

[0016] The beneficial effect of the above further scheme is: by continuously performing simulations, the changing trend of subsequent indicators when the previous indicators change, the degree of convergence and divergence can be judged, and screening can be performed according to the change threshold, which can make up for the errors caused by data collection errors or accidental events.

[0017] Furthermore, the indicator types include positive indicators and negative indicators; the positive indicator is an indicator with a larger value and better quality; the negative indicator is an indicator with a larger value and worse quality.

[0018] The beneficial effects of the above further solution are: the classification of indicator types prepares for the subsequent adjustment of standard thresholds.

[0019] Furthermore, the quality inspection index set Quality inspection indicator set The expression of the impact path is:

[0020] in, Quality inspection indicator set Quality inspection indicator set The impact path; For the The first production process Quality inspection indicators; For the Impact on production process A set of quality inspection indicators; For the Quality inspection index index of the production process.

[0021] The beneficial effect of the above further solution is: obtaining the factors in the previous process that affect each quality inspection indicator of the current process, preparing for the subsequent calculation of the degree of influence.

[0022] Furthermore, the The expression of the degree of association is:

[0023]

[0024] in, for degree of association; For the The quality inspection indicators of the production process are The degree of influence of the first quality inspection index of the production process; For the The quality inspection indicators of the production process are The degree of influence of the second quality inspection index of the production process; For the The quality inspection indicators of the production process are The first production process The degree of influence of each quality inspection indicator; For the Quality inspection indicators of production process the extent of the impact; For the The first production process Quality inspection indicators; for The first quality inspection indicator is the extent of the impact; For the Impact on production process A set of quality inspection indicators; for The second quality inspection index is the extent of the impact; for Middle Quality inspection indicators the extent of the impact; for Middle Quality inspection indicators the extent of the impact; for The total number of elements; is the normalization function; for The weight of for Middle Quality inspection indicators The numerical value of affects the mean; is the total number of simulations; is the simulation index; for No. Simulation to Normalized value of the numerical change of the simulation; for Middle Quality inspection index Simulation to Normalized value of the numerical change of the simulation; for Middle Quality inspection indicators The numerical value of affects the mean; and Both The quality inspection indicator index in .

[0025] The beneficial effect of the above further scheme is: the degree of correlation is used to describe the degree of influence of the influencing factors in the previous process on the current process quality inspection indicators. When performing actual calculations, the data measurement thresholds for different indicators may be different. Therefore, normalization can be used to unify the degree changes into one standard, making the calculation of the degree of influence more accurate.

[0026] Furthermore, the relevant quality inspection indicators of the preceding production process are quality inspection indicators in the preceding production processes determined based on the influence path that have an impact on the unqualified quality inspection indicators of the current semi-finished product.

[0027] The beneficial effect of the above further solution is to prepare for the subsequent update of the standard threshold.

[0028] Furthermore, the standard thresholds of the relevant quality inspection indicators of the updated previous production process are specifically: If the unqualified quality inspection indicator of the current semi-finished product is an independent indicator, keep the standard thresholds of all quality inspection indicators of the previous production process unchanged. Otherwise, update the standard thresholds of the relevant quality inspection indicators of the previous production process:

[0029] in, For the Impact on production process Quality inspection indicator set The The updated standard threshold value of each quality inspection indicator; for Quality inspection index index in; For the Impact on production process A set of quality inspection indicators; For the The first production process Quality inspection indicators; The index of unqualified quality inspection indicators for the current semi-finished product; For the Update the standard threshold value of each quality inspection indicator; for The The standard threshold value before the quality inspection indicator is updated; For the The quality inspection index is The degree of influence of each quality inspection indicator; for The quality inspection indicators in The average impact of each quality inspection indicator; is the threshold adjustment coefficient used to balance the deviation of the impact degree distribution; is the scaling factor used to control the parameter adjustment step size; for The numerical value of for The standard threshold value of For judgment and The A binary parameter indicating whether the quality inspection indicators are of the same type. If they are of the same indicator type, it is 1; otherwise, it is 0. for The minimum value constraint of for The maximum value constraint.

[0030] The beneficial effects of the above further scheme are: when adjusting the standard threshold, the threshold constraints of the indicator items are taken into account, which is more in line with the actual working conditions. At the same time, the degree of influence of the influencing factors on the unqualified indicators is referred to during the adjustment, which can make the adjustment more accurate.

[0031] Furthermore, the independent indicator item is a quality inspection indicator item that has no impact on any quality inspection indicator of any previous production process.

[0032] The beneficial effects of the above further solution are: eliminating special indicator items, fully considering the correlation of process indicators, and avoiding erroneous adjustments. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 Flow chart of the method of the present invention.

[0034] Figure 2 This is a line chart showing the indicator change trend in an embodiment of the present invention. DETAILED DESCRIPTION

[0035] The specific embodiments of the present invention are described below to facilitate understanding of the present invention by those skilled in the art. However, it should be clear that the present invention is not limited to the scope of the specific embodiments. For those skilled in the art, as long as various changes are within the spirit and scope of the present invention as defined and determined by the appended claims, these changes are obvious, and all inventions and creations utilizing the concepts of the present invention are protected.

[0036] like Figure 1 As shown, in one embodiment of the present invention, a product quality detection method for an intelligent manufacturing production line includes: Based on the product production process, build a production process library and determine the quality inspection indicators of each production process in the production process library; Determine the timing of each production process and obtain the degree of correlation between each production process based on the quality inspection indicators of each production process; Initialize the standard thresholds of each quality inspection indicator for each production process; During production, after the mth production process is completed, the current semi-finished product is tested to obtain the quality inspection index values ​​of the current semi-finished product; The quality inspection index values ​​of the current semi-finished product are compared with the standard threshold values ​​of the quality inspection indicators of the m-th production process. When any quality inspection indicator fails to meet the standards, the current semi-finished product is judged as a failed product and eliminated from the assembly line. Based on the quality inspection results of the current semi-finished product in the 1-m production processes, the standard threshold values ​​of the relevant quality inspection indicators of the previous production process are updated.

[0037] In this embodiment, before implementation, it is necessary to first analyze the massive data accumulated in the production line to determine whether there is a correlation between the indicators of each process; if there is a correlation, based on the feedback of the quality inspection indicators in the subsequent process, it is determined whether the threshold value of the quality inspection indicator of the previous correlation is appropriate and adjusted, so as to screen out potential unqualified products in the early stage.

[0038] The degree of correlation between the production processes is obtained as follows: According to the time sequence of each production process, the production process association pair matrix is ​​constructed:

[0039] in, is the production process correlation pair matrix; is the self-correlation item of the first production process; For the first production process Related items of the production process; For the The self-correlation items of the production process; is the total number of production processes, and the production processes are numbered in chronological order; is a zero matrix; Conduct several production line simulations to obtain data sets of quality inspection indicators for each production process; Set the correlation degree of each correlation item in the production process correlation matrix to 1; According to the quality inspection index data set of each production process, calculate the The production process of the first Production process The degree of correlation; among them, and All are production process indexes. .

[0040] In this embodiment, attention is paid to the influence of the preceding production process on the subsequent production process. Therefore, the production process association pair matrix is ​​a triangular matrix that takes the time sequence relationship into consideration.

[0041] The calculation The degree of correlation is: Get the Quality inspection indicators of the production process:

[0042] in, For the Quality inspection index set for the production process; For the The first quality inspection indicator of the production process; For the The second quality inspection indicator of the production process; For the The first production process Quality inspection indicators; For the The total number of quality inspection indicators of the production process; Get the Quality inspection indicators of the production process:

[0043] in, For the Quality inspection index set for the production process; For the The first quality inspection indicator of the production process; For the The second quality inspection indicator of the production process; For the The first production process Quality inspection indicators; For the The total number of quality inspection indicators of the production process; According to the quality inspection index set and quality inspection indicator set , make judgment on the impact of quality inspection indicators: From the quality inspection index set Select quality inspection indicator B from the quality inspection indicator set Select quality inspection indicator C; According to the quality inspection index data set of each production process, the production line simulation round is used as the horizontal axis and the quality inspection index value is used as the vertical axis. The change trend of quality inspection index B and quality inspection index C is depicted in the same line chart (such as Figure 2 shown); If the indicator types of quality inspection indicators B and C are the same, calculate the change convergence of quality inspection indicators B and C:

[0044] in, is the change convergence of quality inspection index B and quality inspection index C; Quality inspection indicator B Simulation to The change trend of the simulation is 1 when it increases, 0 when it remains the same, and -1 when it decreases. Quality inspection indicator C Simulation to The change trend of the simulation is 1 when it increases, 0 when it remains the same, and -1 when it decreases. is the total number of simulations; is the simulation index; A function to determine whether the changing trends of two parameters are the same. If they are the same, it is 1, otherwise it is 0; If the indicator types of quality inspection indicators B and C are different, calculate the change deviation of quality inspection indicators B and C:

[0045] in, is the change deviation degree of quality inspection index B and quality inspection index C; A function to determine whether the changing trends of two parameters are opposite. If they are the same, it is 1, otherwise it is 0; When the change convergence or the change divergence is greater than the change threshold, the The quality inspection index C of the production process is subject to the The quality inspection index B of the first production process is affected, otherwise, the judgment of The quality inspection index C of the production process is not subject to the The quality inspection index B of the production process is affected; Repeated quality inspection indicators affect the judgment process and obtain the quality inspection indicator set Quality inspection indicators and quality inspection indicator sets The influence relationship of each quality inspection index in the Quality inspection indicators and quality inspection indicator sets The influence relationship of each quality inspection index in the construction of quality inspection index set Quality inspection indicator set The impact path; According to the quality inspection index set Quality inspection indicator set The impact path of degree of correlation.

[0046] The indicator types include positive indicators and negative indicators; the positive indicator is an indicator with a larger value and better quality; the negative indicator is an indicator with a larger value and worse quality.

[0047] In this embodiment, not all indicators are correlated in actual working conditions, so it is necessary to analyze the changing trends between the quality inspection indicators of the preceding process and the subsequent process.

[0048] The quality inspection index set Quality inspection indicator set The expression of the impact path is:

[0049] in, Quality inspection indicator set Quality inspection indicator set The impact path; For the The first production process Quality inspection indicators; For the Impact on production process A set of quality inspection indicators; For the Quality inspection index index of the production process.

[0050] described The expression of the degree of association is:

[0051]

[0052] in, for degree of association; For the The quality inspection indicators of the production process are The degree of influence of the first quality inspection index of the production process; For the The quality inspection indicators of the production process are The degree of influence of the second quality inspection index of the production process; For the The quality inspection indicators of the production process are The first production process The degree of influence of each quality inspection indicator; For the Quality inspection indicators of production process the extent of the impact; For the The first production process Quality inspection indicators; for The first quality inspection indicator is the extent of the impact; For the Impact on production process A set of quality inspection indicators; for The second quality inspection index is the extent of the impact; for Middle Quality inspection indicators the extent of the impact; for Middle Quality inspection indicators the extent of the impact; for The total number of elements; is the normalization function; for The weight of for Middle Quality inspection indicators The numerical value of affects the mean; is the total number of simulations; is the simulation index; for No. Simulation to Normalized value of the numerical change of the simulation; for Middle Quality inspection index Simulation to Normalized value of the numerical change of the simulation; for Middle Quality inspection indicators The numerical value of affects the mean; and Both The quality inspection indicator index in .

[0053] In this embodiment, the results of dividing the changes in the two influencing indicators in each simulation are summed and averaged. In a single division, when the change in indicator 1 is d1 and the change in indicator 2 is d2, due to the different representations of the indicators, some are counted as natural numbers and some are counted as percentages, they need to be normalized before the two are divided. This yields the proportional relationship between the changes in indicator 1 and indicator 2. The sum of multiple summations and averages provides a more accurate representation of this proportional relationship.

[0054] The relevant quality inspection indicators of the preceding production process are quality inspection indicators in the preceding production processes determined based on the influence path that have an impact on the unqualified quality inspection indicators of the current semi-finished product.

[0055] The standard thresholds of the relevant quality inspection indicators of the updated pre-production process are specifically: If the unqualified quality inspection indicator of the current semi-finished product is an independent indicator, keep the standard thresholds of all quality inspection indicators of the previous production process unchanged. Otherwise, update the standard thresholds of the relevant quality inspection indicators of the previous production process:

[0056] in, For the Impact on production process Quality inspection indicator set The The updated standard threshold value of each quality inspection indicator; for Quality inspection index index in; For the Impact on production process A set of quality inspection indicators; For the The first production process Quality inspection indicators; The index of unqualified quality inspection indicators for the current semi-finished product; For the Update the standard threshold value of each quality inspection indicator; for The The standard threshold value before the quality inspection indicator is updated; For the The quality inspection index is The degree of influence of each quality inspection indicator; for The quality inspection indicators in The average impact of each quality inspection indicator; is the threshold adjustment coefficient used to balance the deviation of the impact degree distribution; is the scaling factor used to control the parameter adjustment step size; for The numerical value of for The standard threshold value of For judgment and The A binary parameter indicating whether the quality inspection indicators are of the same type. If they are of the same indicator type, it is 1; otherwise, it is 0. for The minimum value constraint of for The maximum value constraint.

[0057] In this embodiment, when failed products appear in the subsequent process, in order to avoid the adjustment of the indicators beyond the range and to avoid concentrating the control risk on a single indicator, it is necessary to adjust the influential indicators of all the previous processes. With this adjustment method, the adjustment of each indicator threshold will be very small.

[0058] During the adjustment process, the parameters To share the risk, the expected adjustment amount of the unqualified indicator can be divided by the sum of the impact of the relevant quality inspection indicators of all previous production processes on the unqualified indicator: ;in, is the expected adjustment amount for the non-conforming indicator; For the previous The degree of influence of the production process on the unqualified indicators.

[0059] In this embodiment, during actual implementation, sometimes in order to be on the safe side and to reduce the adjustment risk, the adjustment ratio may be multiplied by a coefficient greater than 0 and less than 1 for control.

[0060] The independent indicator item is a quality inspection indicator item that has no impact on any quality inspection indicator of any previous production process.

Claims

1. A product quality detection method for an intelligent manufacturing production line, characterized in that: include: Based on the product production process, build a production process library and determine the quality inspection indicators of each production process in the production process library; Determine the timing of each production process and obtain the degree of correlation between each production process based on the quality inspection indicators of each production process; Initialize the standard thresholds of each quality inspection indicator for each production process; During production, after the mth production process is completed, the current semi-finished product is tested to obtain the quality inspection index values ​​of the current semi-finished product; The quality inspection index values ​​of the current semi-finished product are compared with the standard threshold values ​​of the quality inspection indicators of the m-th production process. When any quality inspection indicator fails to meet the standards, the current semi-finished product is judged as a failed product and eliminated from the assembly line. Based on the quality inspection results of the current semi-finished product in the 1-m production processes, the standard threshold values ​​of the relevant quality inspection indicators of the previous production process are updated.

2. The product quality detection method of the intelligent manufacturing production line according to claim 1 is characterized in that: The degree of correlation between the production processes is obtained as follows: According to the time sequence of each production process, the production process association pair matrix is ​​constructed: in, is the production process correlation pair matrix; is the self-correlation item of the first production process; For the first production process Related items of the production process; For the The self-correlation items of the production process; is the total number of production processes, and the production processes are numbered in chronological order; is a zero matrix; Conduct several production line simulations to obtain data sets of quality inspection indicators for each production process; Set the correlation degree of each correlation item in the production process correlation matrix to 1; According to the quality inspection index data set of each production process, calculate the The production process of the first Production process The degree of correlation; among them, and All are production process indexes. .

3. The product quality detection method of the intelligent manufacturing production line according to claim 2 is characterized in that: The calculation The degree of correlation is: Get the Quality inspection indicators of the production process: in, For the Quality inspection index set for the production process; For the The first quality inspection indicator of the production process; For the The second quality inspection indicator of the production process; For the The first production process Quality inspection indicators; For the The total number of quality inspection indicators of the production process; Get the Quality inspection indicators of the production process: in, For the Quality inspection index set for the production process; For the The first quality inspection indicator of the production process; For the The second quality inspection indicator of the production process; For the The first production process Quality inspection indicators; For the The total number of quality inspection indicators of the production process; According to the quality inspection index set and quality inspection indicator set , make judgment on the impact of quality inspection indicators: From the quality inspection index set Select quality inspection indicator B from the quality inspection indicator set Select quality inspection indicator C; Based on the quality inspection indicator data sets of each production process, the change trends of quality inspection indicator B and quality inspection indicator C are depicted in the same line chart, with the production line simulation round as the horizontal axis and the quality inspection indicator value as the vertical axis; If the indicator types of quality inspection indicators B and C are the same, calculate the change convergence of quality inspection indicators B and C: in, is the change convergence of quality inspection index B and quality inspection index C; For quality inspection indicator B Simulation to The change trend of the simulation is 1 when it increases, 0 when it remains the same, and -1 when it decreases. For quality inspection indicator C Simulation to The change trend of the simulation is 1 when it increases, 0 when it remains the same, and -1 when it decreases. is the total number of simulations; is the simulation index; A function to determine whether the changing trends of two parameters are the same. If they are the same, it is 1, otherwise it is 0; If the indicator types of quality inspection indicators B and C are different, calculate the change deviation of quality inspection indicators B and C: in, is the change deviation degree of quality inspection index B and quality inspection index C; A function to determine whether the changing trends of two parameters are opposite. If they are the same, it is 1, otherwise it is 0; When the change convergence or the change divergence is greater than the change threshold, the The quality inspection index C of the production process is subject to the The quality inspection index B of the first production process is affected, otherwise, the judgment of The quality inspection index C of the production process is not subject to the The quality inspection index B of the production process is affected; Repeated quality inspection indicators affect the judgment process and obtain the quality inspection indicator set Quality inspection indicators and quality inspection indicator sets The influence relationship of each quality inspection index in the Quality inspection indicators and quality inspection indicator sets The influence relationship of each quality inspection index in the construction of quality inspection index set Quality inspection index set The impact path; According to the quality inspection index set Quality inspection index set The impact path of degree of correlation.

4. The product quality detection method of the intelligent manufacturing production line according to claim 3 is characterized in that: The indicator types include positive indicators and negative indicators; the positive indicator is an indicator with a larger value and better quality; the negative indicator is an indicator with a larger value and worse quality.

5. The product quality detection method of the intelligent manufacturing production line according to claim 3 is characterized in that: The quality inspection index set Quality inspection index set The expression of the impact path is: in, Quality inspection indicator set Quality inspection index set The impact path; For the The first production process Quality inspection indicators; For the Impact on production process A set of quality inspection indicators; For the Quality inspection index index of the production process.

6. The product quality detection method of the intelligent manufacturing production line according to claim 3 is characterized in that: described The expression of the degree of association is: in, for degree of association; For the The quality inspection indicators of the production process are The degree of influence of the first quality inspection index of the production process; For the The quality inspection indicators of the production process are The degree of influence of the second quality inspection index of the production process; For the The quality inspection indicators of the production process are The first production process The degree of influence of each quality inspection indicator; For the Quality inspection indicators of production process the extent of the impact; For the The first production process Quality inspection indicators; for The first quality inspection indicator is the extent of the impact; For the Impact on production process A set of quality inspection indicators; for The second quality inspection index is the extent of the impact; for Middle Quality inspection indicators the extent of the impact; for Middle Quality inspection indicators the extent of the impact; for The total number of elements; is the normalization function; for The weight of for Middle Quality inspection indicators The numerical value of affects the mean; is the total number of simulations; is the simulation index; for No. Simulation to Normalized value of the numerical change of the simulation; for Middle Quality inspection index Simulation to Normalized value of the numerical change of the simulation; for Middle Quality inspection indicators The numerical value of affects the mean; and Both The quality inspection indicator index in .

7. The product quality detection method of the intelligent manufacturing production line according to claim 1 is characterized in that: The relevant quality inspection indicators of the preceding production process are quality inspection indicators in the preceding production processes determined based on the influence path that have an impact on the unqualified quality inspection indicators of the current semi-finished product.

8. The product quality detection method of the intelligent manufacturing production line according to claim 1 is characterized in that: The standard thresholds of the relevant quality inspection indicators of the updated pre-production process are specifically: If the unqualified quality inspection indicator of the current semi-finished product is an independent indicator, keep the standard thresholds of all quality inspection indicators of the previous production process unchanged. Otherwise, update the standard thresholds of the relevant quality inspection indicators of the previous production process: in, For the Impact on production process Quality inspection indicator set The The updated standard threshold value of each quality inspection indicator; for Quality inspection index index in; For the Impact on production process A set of quality inspection indicators; For the The first production process Quality inspection indicators; The index of unqualified quality inspection indicators for the current semi-finished product; For the Update the standard threshold value of each quality inspection indicator; for The The standard threshold value before the quality inspection indicator is updated; For the The quality inspection index is The degree of influence of each quality inspection indicator; for The quality inspection indicators in The average impact of each quality inspection indicator; is the threshold adjustment coefficient used to balance the deviation of the impact degree distribution; is the scaling factor used to control the parameter adjustment step size; for The value of for The standard threshold value of For judgment and The A binary parameter indicating whether the quality inspection indicators are of the same type. If they are of the same indicator type, it is 1; otherwise, it is 0. for The minimum value constraint of for The maximum value constraint.

9. The product quality detection method of the intelligent manufacturing production line according to claim 8, characterized in that: The independent indicator item is a quality inspection indicator item that has no impact on any quality inspection indicator of any previous production process.

Images