Intelligent analysis method and system for printing failure data

Abstract

The application discloses a kind of printing fault data intelligent analysis method and system, belong to fault monitoring technical field, this method first gathers raw materials in printing production, environment and image data constructs fault parameter library, combs fault and constructs risk fault library, analyzes the correlation of both to determine prepress wind evaluation parameter;Then calculate wind evaluation value, match the collection frequency of in-print wind evaluation parameter set;Through sliding time window, calculate in-print evaluation value and update collection frequency, according to risk fault early warning of fault evaluation value;Finally, construct maintenance personnel database, according to maintenance case and early warning fault analysis, select evaluation value, and screen out the most suitable maintenance personnel to handle fault;The application is convenient for positioning fault factor, quantifies risk assessment, realizes data acquisition dynamic adjustment and accurate early warning, efficiently allocates maintenance resources, so as to improve the quality and efficiency of printing production, enhance the comprehensive competitiveness of enterprise in market.

Description

Technical Field

[0001] This invention belongs to the field of fault monitoring technology, specifically relating to an intelligent analysis method and system for printing fault data. Background Technology

[0002] In the printing industry, with the advancement of technology and the growth of market demand, the scale of printing production is constantly expanding, the variety of products is becoming increasingly diverse, and the requirements for printing quality and production efficiency are also getting higher and higher. However, the existing printing fault handling technology has obvious shortcomings.

[0003] There are shortcomings in sorting out the correlation between various faults in printing production and raw materials, environment, and image data. It is difficult to accurately identify the key factors that may cause faults in the pre-press stage, thus affecting the ability to take targeted preventive measures in advance.

[0004] There are limitations in dynamically adjusting the data collection frequency according to the fault risk, making it difficult to monitor risk changes in the printing process in real time, which affects the ability to capture early signs of faults in a timely manner, and consequently affects the timeliness and accuracy of fault warnings.

[0005] There are shortcomings in the quantitative assessment of failure risks, making it difficult to intuitively reflect the probability of failures occurring. This affects the company's ability to rationally allocate production resources and preventive measures according to risk priorities, thereby affecting the overall efficiency and quality of printing production. To address this, we propose an intelligent analysis method and system for printing failure data. Summary of the Invention

[0006] The purpose of this invention is to provide an intelligent analysis method and system for printing fault data to solve the problems mentioned in the background art.

[0007] To achieve the above objectives, the present invention provides the following technical solution: an intelligent analysis method for printing fault data, comprising the following steps.

[0008] Step 1: Construct a fault parameter library and analyze the correlation between each risk fault in the fault parameter library and each fault parameter in the fault parameter library. Based on this, analyze the pre-press risk assessment parameters corresponding to each risk fault.

[0009] Step 2: Analyze the risk assessment value of the risk fault based on the prepress risk assessment parameters corresponding to the risk fault; construct the in-press risk assessment parameter set corresponding to the risk fault, and match the collection frequency of the corresponding in-press risk assessment parameter set according to the risk fault risk assessment value;

[0010] Step 3: Set a sliding time window, calculate the print quality assessment value of the print quality assessment parameter within the sliding time window, and update the collection frequency of the print quality assessment parameter according to the print quality assessment value; analyze the print quality assessment value of each print quality assessment parameter for risk faults to obtain the fault assessment value, and determine whether to issue a risk fault warning based on the fault assessment value;

[0011] Step 4: Build a maintenance personnel database. Based on the maintenance personnel's maintenance cases and the risk fault analysis corresponding to the current risk fault warning, the maintenance personnel with the highest evaluation value are selected as task dispatch personnel.

[0012] Preferably, in step one, the specific process of constructing the fault parameter library is as follows:

[0013] Collect raw material data, printing environment data, and printed image data from the current printing production process, and organize them to build a pre-press fault parameter database. Raw material data includes: paper thickness, ink flow, and plate abrasion resistance; printing environment data includes: workshop temperature and humidity, cleanliness, and air pressure; printed image data includes: number of colors, dot distribution density, and image and text detail.

[0014] We have compiled a risk failure database by identifying all printing failures that may result from prepress raw material data, printing environment data, and printing image data.

[0015] Preferably, in step one, the specific process of analyzing the correlation between each risk fault in the fault parameter database and each fault parameter in the fault parameter database, and based on this, analyzing the pre-press risk assessment parameters corresponding to each risk fault, is as follows:

[0016] For each risk fault in the risk fault database, a Cartesian coordinate system is established between the risk fault and the fault parameter, where the risk fault level is the vertical axis and the fault parameter is the horizontal axis. Then, from the large amount of data recorded in the process log of printing production, the data points corresponding to the fault parameter and the risk fault are extracted one by one. These data points are marked one by one in the constructed Cartesian coordinate system to obtain the analysis scatter plot.

[0017] Divide the risk / failure severity into m intervals, denoted as C1, C2, ..., Cm; and divide the failure parameters into n intervals, denoted as A1, A2, ..., An; count the number of data points falling into the intersection of each risk / failure severity interval and the failure parameter interval, and mark them as follows: Where i represents the label of the risk level interval and j represents the label of the fault parameter interval; using the formula: To obtain the total number of data points in all intersecting regions. The joint probability of being in the risk fault level interval Ci and the fault parameter interval Aj The marginal probability of being in the risk failure level range Ci ; and the marginal probability of being in the fault parameter range Aj ;

[0018] The above , , as well as Using the mutual information formula: The parameter information value MI is obtained;

[0019] Each fault parameter and the fault parameter information value corresponding to the risk fault are compared with the corresponding threshold. If the fault parameter information value is greater than the corresponding threshold, the fault parameter corresponding to the fault parameter information value is marked as the prepress risk assessment parameter of this risk fault.

[0020] At the same time, based on the scatter plot of each fault parameter and the corresponding risk fault, the positive and negative correlations between the fault parameters and the corresponding risk faults are marked.

[0021] Preferably, in step two, the specific process of analyzing the risk assessment value of a risk fault based on the prepress risk assessment parameters corresponding to the risk fault is as follows:

[0022] For each risk failure, obtain the values ​​of various prepress risk assessment parameters corresponding to the risk failure, and mark the prepress risk assessment parameters that are positively correlated with the risk failure as follows: r = 1, 2, ..., P; where P is the total number of prepress risk assessment parameters positively correlated with risk failures; prepress risk assessment parameters negatively correlated with risk failures are labeled as... a = 1, 2, ..., K; where K is the total number of prepress risk assessment parameters that are negatively correlated with risk failures; after normalizing all prepress risk assessment parameter values, the formula is used: The reputation score FP is obtained, among which Preset weighting coefficients for prepress risk assessment parameters that are positively correlated with risk failures; Preset weighting coefficients for prepress risk assessment parameters that are negatively correlated with risk failures.

[0023] Preferably, in step two, a set of India-China risk assessment parameters corresponding to risk faults is constructed, and the collection frequency of the corresponding India-China risk assessment parameter set is matched according to the risk assessment value of the risk fault. The specific process is as follows:

[0024] All risk faults in the risk fault database are sorted in descending order according to their corresponding risk assessment values ​​to obtain a risk fault sorting sequence, and the risk fault sorting sequence is sent to the monitoring personnel's terminal.

[0025] Each type of risk fault has a corresponding set of print quality assessment parameters. Based on the risk fault sorting sequence of the current printing material, each risk fault is matched with the print quality assessment parameter set corresponding to all risk faults in turn, and the print quality assessment parameter set corresponding to each risk fault is output.

[0026] For each risk fault, several risk assessment value ranges are preset, and each risk assessment value range corresponds to a printing risk assessment parameter collection frequency. The risk assessment value of the risk fault is matched with all preset risk assessment value ranges, and its corresponding printing risk assessment parameter collection frequency is output. This printing risk assessment parameter collection frequency is used as the collection frequency of each risk assessment parameter in the printing risk assessment parameter set for that risk fault during the printing process.

[0027] Preferably, in step three, the specific process of setting a sliding time window, calculating the China-India rating value of the China-India rating parameter within the sliding time window, and updating the collection frequency of the China-India rating parameter based on the China-India rating value is as follows:

[0028] In the printing process, for each risk fault, the values ​​of each print risk assessment parameter in the print risk assessment parameter set are collected according to the collection frequency corresponding to the print risk assessment parameter set of the risk fault; thus obtaining the value of each print risk assessment parameter at each collection time.

[0029] For each India-China reputation parameter, a sliding time window is set, and the mean, standard deviation, and range of the India-China reputation parameter within the sliding window are calculated. Then, different weight coefficients are assigned to the three, and the mean, standard deviation, and range of the India-China reputation parameter are multiplied by the corresponding weight coefficients and then summed to obtain the India-China evaluation value.

[0030] Multiple preset frequency update intervals for the India-China evaluation values ​​are used to continuously compare the frequency update intervals of the India-China evaluation values ​​in the current sliding window with those in the previous sliding window. Once a jump in the frequency update interval of the India-China evaluation values ​​in the current sliding window is detected, a frequency update operation is triggered.

[0031] Each India-China assessment value sampling frequency update interval corresponds to an India-China reputation parameter sampling update frequency. When the sampling update operation is triggered, the sampling update frequency corresponding to the sampling frequency update interval of the India-China assessment value in the current sliding window is used as the sampling frequency of the India-China reputation parameter for the next period of time.

[0032] The sampling frequency is then continuously updated according to the above-mentioned sampling frequency update process. The print evaluation value within each sliding time window is continuously calculated and the interval jump situation is judged. If the interval jump occurs again, the sampling frequency is updated again.

[0033] Preferably, in step three, the print assessment values ​​of each print risk parameter are analyzed to obtain the fault assessment value. The specific process for determining whether to issue a risk fault warning based on the fault assessment value is as follows:

[0034] For each India-China risk assessment parameter, multiple risk ranges for India-China assessment values ​​are set. Each risk range corresponds to a pre-defined risk assignment. The India-China assessment value corresponding to the current India-China risk assessment parameter is matched with all risk ranges for India-China assessment values, and the corresponding risk assignment is output. ; g = 1, 2, ..., H; g is the label of the India-China wind rating parameter, and H is the total number of India-China wind rating parameters;

[0035] Substitute the risk values ​​corresponding to each India-China risk assessment parameter for the risk fault into the preset formula model: Therefore, the evaluation value GP is obtained, where Assign preset weight coefficients to the risk assessment parameters for each India-China risk assessment parameter;

[0036] Using the fault assessment value as the vertical axis and time as the horizontal axis, a two-dimensional rectangular coordinate system is established. The value of the fault assessment value is calculated in real time and substituted into the two-dimensional rectangular coordinate system to obtain a series of fault assessment data points. Adjacent fault assessment data points are connected by curve smoothing to obtain the fault assessment change curve. A fault assessment threshold line is preset. The area enclosed by the fault assessment change curve exceeding the fault assessment threshold line and the fault assessment threshold line is marked as the risk assessment area.

[0037] The risk assessment area from the start time to the current time is accumulated to obtain the instant risk accumulation. An instant risk accumulation threshold is preset. If the instant risk accumulation is greater than or equal to the corresponding threshold, the risk fault warning is triggered.

[0038] Preferably, in step four, the specific process of constructing the maintenance personnel database is as follows:

[0039] Build a database of maintenance personnel to record their personal information, maintenance cases, and maintenance schedules; personal information includes: name, ID card number, personnel image, and contact information;

[0040] The maintenance case records contain the names of risky faults that maintenance personnel can repair, the number of each risky fault handled, the average maintenance time, and the maintenance performance score.

[0041] When a risk fault warning is issued, all risk faults corresponding to the warning are obtained. The personnel currently on the maintenance schedule are selected from the maintenance personnel database, their maintenance cases are extracted, and maintenance personnel who have handled all the risk faults that are warning are selected as the initial personnel.

[0042] Preferably, in step four, the specific process of analyzing the selection evaluation value of maintenance personnel and selecting the maintenance personnel with the highest selection evaluation value as task dispatch personnel is as follows:

[0043] For the initial selection personnel, the following parameters are considered in their maintenance cases: the number of risky faults handled (SLz), the average maintenance time (WTz), and the maintenance performance score (JXz), where z is the fault number (z=1, 2, ..., D), and D is the total number of risky faults. After normalizing the number of risky faults handled (SLz), the average maintenance time (WTz), and the maintenance performance score (JXz), the following formula is used: The dimension rating WP is obtained, where Mean time to repair for all risky faults The minimum value in the value, b is the loop variable used to sum the number of risk faults handled, b=1,2,...,D; , , These are the weighting coefficients for the number of risky faults handled, the average repair time, and the repair performance score, respectively.

[0044] The number of unresolved risk faults currently being handled by the initial selection personnel is obtained. The task load RF is obtained by multiplying each unresolved risk fault by the average repair time of the corresponding risk fault and then summing the results.

[0045] For each shortlisted candidate, the corresponding dimensional assessment value WP and task load RF are normalized and then processed using the formula: The evaluation value XPZ is obtained, where d1 and d2 are preset weight coefficients.

[0046] The candidate with the highest evaluation score from all shortlisted candidates will be selected as the task dispatcher to handle the risk faults caused by this warning. After the task dispatcher successfully resolves the risk faults, a maintenance performance score will be conducted for each risk fault handled by the candidate, and the score will be recorded in the maintenance performance score record of the corresponding risk fault in the candidate's maintenance case.

[0047] Compared with the prior art, the beneficial effects of the present invention are:

[0048] (1) The intelligent analysis method and system for printing fault data collects data on raw materials, environment and images in printing production to build a fault parameter library, sorts out various faults to build a risk fault library, and analyzes the correlation between the two in depth. This process makes it easy to find the pre-press risk assessment parameters that are closely related to each risk fault, so that printing companies can clearly recognize the key factors that may cause faults in the early stage of production, and provide a strong basis for targeted fault prevention.

[0049] (2) The intelligent analysis method and system for printing failure data calculates the risk assessment value of risk failures based on the pre-press risk assessment parameters, quantifies the risk, and intuitively reflects the probability of failure occurrence; constructs a set of in-press risk assessment parameters based on the risk assessment value and matches the collection frequency to avoid blindly collecting data and achieve reasonable resource allocation; at the same time, risk failures are sorted by risk assessment value and sent to monitoring personnel to facilitate them to quickly understand the priority of failures, pay special attention to high-risk failures, provide clear risk reference for printing production, and help enterprises make scientific decisions and rationally arrange production resources and preventive measures.

[0050] (3) This intelligent analysis method and system for printing fault data calculates the print quality assessment value by setting a sliding time window, and updates the print quality assessment parameter collection frequency by jumping according to the interval of the assessment value, so as to realize the dynamic adjustment of the data collection frequency and timely capture the small fluctuations of faults; calculates the fault assessment value by comprehensively considering the mean, standard deviation and range of the print quality assessment parameters, and combines it with risk assignment analysis to accurately reflect the probability of risk fault occurrence; calculates the real-time risk product by establishing a two-dimensional rectangular coordinate system and comparing it with the preset threshold to achieve accurate early warning, so that enterprises can take measures in advance to reduce production interruption and quality problems, and ensure production continuity and product quality.

[0051] (4) The intelligent analysis method and system for printing fault data constructs a database of maintenance personnel and selects task dispatch personnel in combination with risk fault early warning; comprehensively considers the number of cases handled by maintenance personnel, average maintenance time and performance score and other multi-dimensional data, as well as the current task load, calculates the selection value to determine the most suitable maintenance personnel, and improves the efficiency of maintenance resource utilization; the maintenance performance score and record update after the task is completed provide continuous data support for the assessment of maintenance personnel's capabilities, which helps to optimize maintenance team management and ensure that faults are handled in a timely and effective manner. Attached Figure Description

[0052] Figure 1 This is a flowchart of the present invention. Detailed Implementation

[0053] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. Example

[0054] Please see Figure 1 This invention provides an intelligent analysis method for printing defect data, including:

[0055] Step 1: Construct a fault parameter library and analyze the correlation between each risk fault in the fault parameter library and each fault parameter in the fault parameter library. Based on this, analyze the pre-press risk assessment parameters corresponding to each risk fault. The specific process is as follows:

[0056] Data on raw materials, printing environment, and printed images from the current printing production process are collected. After preprocessing the collected data, a pre-press fault parameter database is constructed. Raw material data includes: paper thickness, ink flow, and plate abrasion resistance. Printing environment data includes: workshop temperature and humidity, cleanliness, and air pressure. Printed image data includes: number of colors, dot density, and image detail.

[0057] This study identifies all printing defects that can be caused by data such as prepress raw material data, printing environment data, and printing image data, and compiles a risk defect database. The risk defect database includes defects such as misregistration, uneven ink color, paper wrinkling, and halftone blurring.

[0058] For each risk fault in the risk fault database, a Cartesian coordinate system is established between the risk fault and the fault parameter, where the risk fault level is the vertical axis and the fault parameter is the horizontal axis. Then, from the large amount of data recorded in the process log of printing production, the data points corresponding to the fault parameter and the risk fault are extracted one by one. These data points are marked one by one in the constructed Cartesian coordinate system to obtain the analysis scatter plot.

[0059] Divide the risk of failure into m intervals, denoted as C1, C2, ..., Cn. m Simultaneously, the fault parameters are divided into n intervals, denoted as A1, A2, ..., A n Count the number of data points that fall within the intersection area of ​​each risk failure severity interval and failure parameter interval, and mark them as follows: Where i represents the label of the risk level interval and j represents the label of the fault parameter interval; using the formula: To obtain the total number of data points in all intersecting regions. The joint probability of being in the risk fault level interval Ci and the fault parameter interval Aj Marginal probability within the risk failure level range Ci ; and the marginal probability of being in the fault parameter range Aj ;

[0060] The above , , as well as Using the mutual information formula: The fault parameter information value MI is obtained, where the fault parameter information value represents the mutual information value between the fault parameter and the risk fault. The larger the fault parameter information value, the stronger the dependence between the fault parameter and the risk fault.

[0061] A preset fault parameter information threshold is set. The fault parameter information value corresponding to each fault parameter and risk fault is compared with the corresponding threshold. If the fault parameter information value is greater than the corresponding threshold, the fault parameter corresponding to the fault parameter information value is marked as the prepress risk assessment parameter of this risk fault.

[0062] At the same time, based on the scatter plot of each fault parameter and the corresponding risk fault, the positive and negative correlations between the fault parameters and the corresponding risk faults are marked.

[0063] It should be noted that by collecting data from various aspects of printing production to build a fault parameter database, sorting out various faults to build a risk fault database, and deeply analyzing the correlation between the two, it is possible to accurately identify the pre-press risk assessment parameters closely related to each risk fault. This allows for a clear understanding of the key factors that may cause faults in the early stages of printing production. At the same time, the determination of pre-press risk assessment parameters lays an important foundation for fault prediction and intelligent analysis in the subsequent printing process.

[0064] Step Two: Analyze the risk assessment value of the risk fault based on the pre-press risk assessment parameters corresponding to the risk fault; construct the in-press risk assessment parameter set corresponding to the risk fault, and match the collection frequency of the corresponding in-press risk assessment parameter set according to the risk fault's risk assessment value. The specific process is as follows:

[0065] For each risk failure, obtain the values ​​of various prepress risk assessment parameters corresponding to the risk failure, and mark the prepress risk assessment parameters that are positively correlated with the risk failure as follows: r = 1, 2, ..., P; where P is the total number of prepress risk assessment parameters positively correlated with risk failures; prepress risk assessment parameters negatively correlated with risk failures are labeled as... a = 1, 2, ..., K; where K is the total number of prepress risk assessment parameters that are negatively correlated with risk failures; after normalizing all prepress risk assessment parameter values, the formula is used: The reputation score FP is obtained, among which Preset weighting coefficients for prepress risk assessment parameters that are positively correlated with risk failures; Preset weighting coefficients for prepress risk assessment parameters that are negatively correlated with risk failures; the higher the risk assessment value, the greater the likelihood of a risk failure occurring.

[0066] All risk faults in the risk fault database are sorted in descending order according to their corresponding risk assessment values ​​to obtain a risk fault sorting sequence, and the risk fault sorting sequence is sent to the monitoring personnel's terminal.

[0067] Each risk fault is pre-defined to have a corresponding set of in-press risk assessment parameters. Based on the risk fault ranking sequence of the current printing material, each risk fault is matched sequentially with the set of in-press risk assessment parameters corresponding to all risk faults, and the set of in-press risk assessment parameters corresponding to each risk fault is output. The method for analyzing the set of in-press risk assessment parameters corresponding to risk faults can be analyzed by analyzing the correlation of the pre-press risk assessment parameters of each risk fault in step one above.

[0068] For each risk fault, several risk assessment value ranges are preset, and each risk assessment value range corresponds to a printing risk assessment parameter collection frequency. The risk assessment value of the risk fault is matched with all preset risk assessment value ranges, and its corresponding printing risk assessment parameter collection frequency is output. This printing risk assessment parameter collection frequency is used as the collection frequency of each risk assessment parameter in the printing risk assessment parameter set for that risk fault during the printing process.

[0069] It should be noted that by calculating the risk assessment value, the probability of a risky failure is quantified, which can intuitively reflect the probability of the failure and provide a clear risk reference for printing production. Matching the collection frequency of the print risk assessment parameter set with the risk assessment value avoids blind data collection, saves system resources, and achieves reasonable resource allocation.

[0070] The risk faults are sorted by risk assessment value and sent to monitoring personnel, enabling them to quickly understand the priority of various risk faults in printing production and to focus on high-risk faults. At the same time, the set of risk assessment parameters and the collection frequency in printing are determined, which provides a key data foundation for subsequent real-time monitoring and fault early warning, and helps to discover potential faults in advance and take measures.

[0071] Step 3: Set a sliding time window, calculate the print quality assessment value of the print quality assessment parameters within the sliding time window, and update the collection frequency of the print quality assessment parameters based on the print quality assessment value; analyze the print quality assessment value of each print quality assessment parameter for risk faults to obtain the fault assessment value, and analyze based on the fault assessment value to determine whether to issue a risk fault warning. The specific process is as follows:

[0072] In the printing process, for each risk fault, the values ​​of each print risk assessment parameter in the print risk assessment parameter set are collected according to the collection frequency corresponding to the print risk assessment parameter set of the risk fault; thus obtaining the value of each print risk assessment parameter at each collection time.

[0073] For each India-China reputation parameter, a sliding time window is set, and the size and step size of the sliding time window are preset. The size and step size of the sliding time window can be set according to the collection frequency of the reputation parameter. The faster the collection frequency, the smaller the window size and step size can be to collect enough data.

[0074] Calculate the mean, standard deviation, and range of the India-China reputation parameters within the sliding window. Then, assign different weight coefficients to the three parameters. Multiply the mean, standard deviation, and range of the India-China reputation parameters by their respective weight coefficients and sum them up to obtain the India-China rating value.

[0075] Multiple frequency update intervals for the India-China evaluation value are preset, and the frequency update intervals of the India-China evaluation value in the current sliding window and the previous sliding window are continuously compared. Once a jump in the frequency update interval of the India-China evaluation value in the current sliding window is detected, that is, moving from one interval to another, the frequency update operation is triggered.

[0076] Each India-China assessment value frequency update interval corresponds to an India-China opinion assessment parameter collection update frequency. When the frequency update operation is triggered, the collection update frequency corresponding to the frequency update interval of the India-China assessment value in the current sliding window is used as the collection frequency of the India-China opinion assessment parameter for the next period of time. Subsequently, the collection frequency update process is continuously followed according to the above judgment process, constantly calculating the India-China assessment values ​​in each sliding time window and judging the interval jump situation. If the interval jump occurs again, the collection frequency is updated again.

[0077] For each India-China risk assessment parameter, multiple risk ranges for India-China assessment values ​​are set. Each risk range corresponds to a pre-defined risk assignment. The India-China assessment value corresponding to the current India-China risk assessment parameter is matched with all risk ranges for India-China assessment values, and the corresponding risk assignment is output. ; g = 1, 2, ..., H; g is the label of the India-China wind rating parameter, and H is the total number of India-China wind rating parameters;

[0078] Substitute the risk values ​​corresponding to each India-China risk assessment parameter for the risk fault into the preset formula model: Therefore, the evaluation value GP is obtained, where Each risk assessment parameter in the printing process is assigned a corresponding preset weight coefficient. Therefore, the larger the assessment value, the greater the probability that the risk fault will occur in the printing process.

[0079] Using the fault assessment value as the vertical axis and time as the horizontal axis, a two-dimensional rectangular coordinate system is established. The value of the fault assessment value is calculated in real time and substituted into the two-dimensional rectangular coordinate system to obtain a series of fault assessment data points. Adjacent fault assessment data points are connected by curve smoothing to obtain the fault assessment change curve. A fault assessment threshold line is preset. The area enclosed by the fault assessment change curve exceeding the fault assessment threshold line and the fault assessment threshold line is marked as the risk assessment area.

[0080] The risk assessment area from the start time to the current time is accumulated to obtain the instant risk accumulation. An instant risk accumulation threshold is preset. If the instant risk accumulation is greater than or equal to the corresponding threshold, the risk fault warning is triggered.

[0081] It should be noted that by calculating the print quality evaluation value through a sliding time window and updating the collection frequency of the print quality evaluation parameters based on the jumps in the interval of the evaluation value, the data collection frequency can be dynamically adjusted. This allows for a more flexible response to changes in the risk and fault status during the printing process and timely capture of subtle fluctuations that may indicate the occurrence of faults.

[0082] The printing quality assessment value is calculated by comprehensively considering the mean, standard deviation, and range of the printing quality assessment parameters. This fully takes into account the central tendency, dispersion, and range of variation of the data. Combined with risk assignment analysis, the failure assessment value is obtained. The failure assessment value reflects the probability of risk failures occurring in the printing process. This helps printing companies take measures in advance to reduce the possibility of failures and reduce production interruptions and quality problems caused by failures.

[0083] By establishing a two-dimensional rectangular coordinate system, the system calculates the real-time risk accumulation and compares it with a preset threshold to trigger risk and fault warnings. This allows for a clear and intuitive display of the changing trends of risk and fault over time. Once the real-time risk accumulation exceeds the threshold, the system immediately issues a warning, giving printing companies sufficient time to take countermeasures.

[0084] Step four: Construct a maintenance personnel database. Based on the maintenance personnel's maintenance cases and the risk fault analysis corresponding to the current risk fault warning, the maintenance personnel with the highest evaluation scores are selected as task dispatch personnel. The specific process is as follows:

[0085] A database of maintenance personnel is established to record their personal information, maintenance cases, and maintenance schedules. Personal information includes: name, ID card number, personnel image, and contact information. Maintenance cases store information such as the names of risky faults that maintenance personnel can repair, the number of each risky fault handled, average maintenance time, and maintenance performance scores.

[0086] When a risk fault warning is issued, all risk faults corresponding to the warning are obtained. The personnel currently on the maintenance schedule are selected from the maintenance personnel database, their maintenance cases are extracted, and maintenance personnel who have handled all the risk faults with warnings are selected as the initial personnel.

[0087] For the initial selection personnel, the following parameters are considered in their maintenance cases: the number of risky faults handled (SLz), the average maintenance time (WTz), and the maintenance performance score (JXz), where z is the fault number (z=1, 2, ..., D), and D is the total number of risky faults. After normalizing the number of risky faults handled (SLz), the average maintenance time (WTz), and the maintenance performance score (JXz), the following formula is used: The dimension rating WP is obtained, where Mean time to repair for all risky faults The minimum value in the value, b is the loop variable used to sum the number of risk faults handled, b=1,2,...,D; , , These are the weighting coefficients for the number of risky faults handled, the average repair time, and the repair performance score, respectively.

[0088] The number of unresolved risk faults currently being handled by the initial selection personnel is obtained. The task load RF is obtained by multiplying each unresolved risk fault by the average repair time of the corresponding risk fault and then summing the results.

[0089] For each shortlisted candidate, the corresponding dimensional assessment value WP and task load RF are normalized and then processed using the formula: The selection evaluation value XPZ is obtained, where d1 and d2 are preset weight coefficients. The larger the selection evaluation value of the preliminary selection personnel, the more suitable the preliminary selection personnel are to be assigned to handle the risk faults corresponding to this fault warning.

[0090] The candidate with the highest evaluation score from all shortlisted candidates will be selected as the task dispatcher to handle the risk faults generated by this warning. After the task dispatcher successfully resolves the risk faults, a maintenance performance score will be conducted for each risk fault handled, and the score will be recorded in the maintenance performance score record of the corresponding risk fault in the candidate's maintenance case.

[0091] It should be noted that building a maintenance personnel database and combining it with risk and fault early warning to screen personnel for task dispatch can accurately identify the most suitable candidates to handle the current fault from a large pool of maintenance personnel. By comprehensively considering multi-dimensional data such as the number of cases handled, average maintenance time, and performance scores, as well as the current task load, the selection evaluation value is calculated to determine the dispatched personnel, which improves the utilization efficiency of maintenance resources and ensures that faults can be handled in a timely and effective manner. The maintenance performance scoring and record updates after the task is completed provide continuous data support for the assessment of maintenance personnel's capabilities.

[0092] A printing fault data intelligent analysis system includes: a fault parameter correlation analysis module, a risk assessment and data acquisition module, an evaluation and early warning module, and a maintenance personnel dispatch module, specifically:

[0093] Fault Parameter Correlation Analysis Module: Constructs a fault parameter library and analyzes the correlation between each risk fault in the fault parameter library and each fault parameter in the fault parameter library. Based on this, it analyzes the pre-press risk assessment parameters corresponding to each risk fault.

[0094] Risk assessment and data acquisition module: Analyze the risk assessment value of risk faults based on the pre-press risk assessment parameters corresponding to the risk faults; construct the in-press risk assessment parameter set corresponding to the risk faults; and match the collection frequency of the corresponding in-press risk assessment parameter set according to the risk fault risk assessment value.

[0095] Evaluation and early warning module: Set a sliding time window, calculate the evaluation value of the printing-in-the-world risk assessment parameters within the sliding time window, update the collection frequency of the printing-in-the-world risk assessment parameters according to the evaluation value; analyze the evaluation value of each printing-in-the-world risk assessment parameter for risk faults to obtain the fault value, and determine whether to issue a risk fault early warning based on the fault value;

[0096] Maintenance personnel dispatch module: Build a maintenance personnel database, analyze the maintenance personnel's evaluation scores based on their maintenance cases and the risk faults corresponding to the current risk fault warnings, and select the maintenance personnel with the highest evaluation scores as task dispatch personnel.

[0097] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method of intelligent analysis of print failure data, characterized by: Comprising the following steps: Step one: build pre-press fault parameter library and risk fault library, build risk fault and fault parameter scatter plot, analyze the correlation between each risk fault and each fault parameter in the fault parameter library according to the scatter plot, obtain the fault parameter information value, and determine the pre-press wind evaluation parameter corresponding to each risk fault based on the fault parameter information value; The specific process of analyzing the pre-press wind evaluation parameter corresponding to each risk fault is: For each risk fault in the risk fault library, a risk fault and fault parameter rectangular coordinate system is established, in which the risk fault degree is the ordinate and the fault parameter is the abscissa; Then, from the data recorded in the process log of the printing production, the one-to-one corresponding data points of the fault parameter and the risk fault are extracted; These data points are labeled one by one in the established rectangular coordinate system to obtain the analysis scatter plot; Divide the risk of failure into m intervals, denoted as C1, C2, ..., Cn. m Simultaneously, the fault parameters are divided into n intervals, denoted as A1, A2, ..., A n Count the number of data points that fall within the intersection area of ​​each risk failure severity interval and failure parameter interval, and mark them as follows: Where i represents the label of the risk level interval and j represents the label of the fault parameter interval; using the formula: To obtain the total number of data points in all intersecting regions. The joint probability of being in the risk fault level interval Ci and the fault parameter interval Aj The marginal probability of being in the risk failure level range Ci ; and the marginal probability of being in the fault parameter range Aj ; The above , , and Mutual information formula is used to get mutual information value MI: , Compare the fault parameter information value corresponding to each fault parameter and risk fault with the corresponding threshold value, if the fault parameter information value is greater than the corresponding threshold value, mark the fault parameter corresponding to the fault parameter information value as the pre-press wind evaluation parameter of the risk fault; At the same time, according to the scatter plot corresponding to each fault parameter and risk fault, mark the positive and negative correlation of the fault parameter and risk fault; Step two: according to the correlation between the pre-press wind evaluation parameter corresponding to the risk fault and the risk fault, analyze the wind evaluation value of the risk fault; Build the pre-press wind evaluation parameter set corresponding to the risk fault, and match the collection frequency of the corresponding pre-press wind evaluation parameter set according to the wind evaluation value of the risk fault; Step three: in the printing work, the value of the pre-press wind evaluation parameter is collected according to the collection frequency of the pre-press wind evaluation parameter set, a sliding time window is set, the pre-press evaluation value of the pre-press wind evaluation parameter in the sliding time window is calculated, and the collection frequency of the pre-press wind evaluation parameter is updated according to the pre-press evaluation value; The pre-press evaluation value of each pre-press wind evaluation parameter of the risk fault is analyzed to obtain the fault evaluation value, and whether to perform risk fault warning is determined according to the fault evaluation value.

2. The method of claim 1, wherein: In step one, the specific process of building the fault parameter library and the fault parameter library is: Collect the raw material data, printing environment data and printing image data in the current printing production, and organize and build the pre-press fault parameter library; The raw material data includes: paper thickness, ink fluidity, plate wear resistance; The printing environment data includes: workshop temperature and humidity, cleanliness, air pressure; The printing image data includes: color number, dot distribution density, image fineness; Comb all printing faults caused by pre-press raw material data, printing environment data and printing image data, and organize and build the risk fault library.

3. The method of claim 2, wherein: In step two, the specific process of analyzing the wind evaluation value of the risk fault according to the pre-press wind evaluation parameter corresponding to the risk fault is: For each risk failure, the values of the pre-printing wind evaluation parameters corresponding to the risk failure are obtained, the pre-printing wind evaluation parameters positively correlated with the risk failure are marked as , r = 1, 2, …, P; wherein P is the total number of the pre-printing wind evaluation parameters positively correlated with the risk failure; the pre-printing wind evaluation parameters negatively correlated with the risk failure are marked as , a = 1, 2, …, K; wherein K is the total number of the pre-printing wind evaluation parameters negatively correlated with the risk failure; after normalization processing of all the values of the pre-printing wind evaluation parameters, the formula is used to obtain the wind evaluation value FP, wherein is a preset weight coefficient of the pre-printing wind evaluation parameters positively correlated with the risk failure; is a preset weight coefficient of the pre-printing wind evaluation parameters negatively correlated with the risk failure.

4. The method of claim 3, wherein: In step two, the specific process of building the pre-press wind evaluation parameter set corresponding to the risk fault and matching the collection frequency of the corresponding pre-press wind evaluation parameter set according to the wind evaluation value of the risk fault is: Sort all risk faults in the risk fault library according to their corresponding wind evaluation value from large to small to obtain a risk fault sorting sequence, and send the risk fault sorting sequence to the monitoring personnel terminal; Each risk fault corresponds to a corresponding pre-press wind evaluation parameter set; According to the current printing material risk fault order sequence, each risk fault is matched with the corresponding set of printing wind evaluation parameters of all risk faults in sequence, and the set of printing wind evaluation parameters corresponding to each risk fault is output; For each risk fault, a plurality of wind evaluation value intervals are preset, and each wind evaluation value interval corresponds to a printing wind evaluation parameter collection frequency; the wind evaluation value of the risk fault is matched with all the preset wind evaluation value intervals, and the corresponding printing wind evaluation parameter collection frequency is output, and the printing wind evaluation parameter collection frequency is taken as the collection frequency of each wind evaluation parameter in the printing wind evaluation parameter set of the risk fault.

5. The method of claim 4, wherein: In step three, a sliding time window is set, the printing evaluation value of the printing wind evaluation parameter in the sliding time window is calculated, and the collection frequency of the printing wind evaluation parameter is updated according to the printing evaluation value. The specific process is as follows: In the printing work, for each risk fault, the values of each printing wind evaluation parameter in the printing wind evaluation parameter set are collected according to the corresponding collection frequency of the printing wind evaluation parameter set of the risk fault; The values of each printing wind evaluation parameter corresponding to each collection time are obtained; For each printing wind evaluation parameter, a sliding time window is set, the mean, standard deviation and range of the printing wind evaluation parameter in the sliding window are calculated, and then the three are assigned different weight coefficients. The mean, standard deviation and range of the printing wind evaluation parameter are multiplied by the corresponding weight coefficients and then accumulated to obtain the printing evaluation value; A plurality of printing evaluation value sampling frequency updating intervals are preset, and the printing evaluation values in the current sliding window and the previous sliding window are continuously compared; Once it is found that the printing evaluation value in the current sliding window jumps to the sampling frequency updating interval, the sampling frequency updating operation is triggered; Each printing evaluation value sampling frequency updating interval corresponds to a printing wind evaluation parameter collection updating frequency, and when the sampling frequency updating operation is triggered, the collection updating frequency corresponding to the sampling frequency updating interval in which the printing evaluation value in the current sliding window is located is taken as the collection frequency of the printing wind evaluation parameter in the next period of time. The process of continuously calculating the printing evaluation value in each sliding time window, judging the interval jump and updating the collection frequency again if the interval jump occurs again is as follows:

6. The method of claim 5, wherein: In step three, the printing evaluation value of each printing wind evaluation parameter of the risk fault is analyzed to obtain the fault evaluation value, and the specific process of judging whether to perform risk fault warning according to the fault evaluation value is as follows: For each printing wind evaluation parameter, a plurality of printing wind evaluation value risk intervals are set, a preset risk value corresponding to each risk interval is set, the printing wind evaluation value corresponding to the current moment of the printing wind evaluation parameter is matched with all the printing wind evaluation value risk intervals, and the corresponding risk value is output ; g = 1, 2, …, H; g is the index of the printing wind evaluation parameter, and H is the total number of the printing wind evaluation parameters The risk evaluation corresponding to each of the printing and wind evaluation parameters corresponding to the risk of failure is substituted into a preset formula model: , to obtain a fault evaluation value GP, wherein is a preset weight coefficient corresponding to the risk evaluation of each printing and wind evaluation parameter. A two-dimensional rectangular coordinate system is established with the fault evaluation value as the vertical coordinate and time as the horizontal coordinate. The value of the risk fault fault evaluation value is calculated in real time and substituted into the two-dimensional rectangular coordinate system to obtain a series of fault evaluation data points. The adjacent fault evaluation data points are sequentially connected by a curve to obtain a fault evaluation change curve. A fault evaluation threshold line is preset, and the area surrounded by the part of the fault evaluation change curve that exceeds the fault evaluation threshold line and the fault evaluation threshold line is marked as a wind evaluation area. All wind evaluation areas from the starting time to the current time are accumulated to obtain the instantaneous risk accumulation. If the instantaneous risk accumulation is greater than or equal to the corresponding threshold, the risk fault warning is triggered. If the risk failure early warning is triggered, the maintenance personnel is dispatched.

7. The method of claim 6, wherein: The specific process of the maintenance personnel dispatching is as follows: A maintenance personnel database is constructed to record personal information, maintenance cases and maintenance scheduling of the maintenance personnel; the personal information includes name, ID card, personnel image and contact information; The maintenance cases store the names of the risk failures that the maintenance personnel can maintain and the number, average maintenance time and maintenance performance score of the processing of each risk failure; When the risk failure early warning occurs, all the risk failures corresponding to the early warning are obtained, the personnel in the current maintenance scheduling are screened from the maintenance personnel database, the maintenance cases of the personnel are extracted, and the maintenance personnel who have processed all the early warning risk failures are screened as the preliminary selected personnel; The selection evaluation value of the maintenance personnel is analyzed, and the maintenance personnel corresponding to the maximum selection evaluation value is selected as the task dispatching personnel to perform the early warning failure operation and maintenance.

8. The method of claim 7, wherein: The specific process of analyzing the selection evaluation value of the maintenance personnel and selecting the maintenance personnel corresponding to the maximum selection evaluation value as the task dispatching personnel to perform the early warning failure operation and maintenance is as follows: For the preliminary selected personnel, the number SLz of the processing of each early warning risk failure, the average maintenance time WTz and the maintenance performance score JXz in the maintenance cases of the personnel are comprehensively analyzed, wherein z is the label of the risk failure, z=1, 2, …, D; D is the total number of risk failures; the processing number of risk failures SLz, the average repair time WTz, and the repair performance score JXz are normalized, and the formula is used to obtain the maintenance evaluation value WP, wherein , is the minimum value of the average repair time of all risk failures , and b is a loop variable when summing the processing number of risk failures, b = 1, 2, …, D; 、 、 , respectively, are the weight coefficients of the processing number of risk failures, the average repair time, and the repair performance score. The number of the risk failures that have not been processed by the preliminary selected personnel is obtained, the task load RF is obtained by multiplying each risk failure that has not been processed by the average maintenance time of the corresponding risk failure and then adding them up; For each primary candidate, the corresponding evaluation value WP of the primary candidate is normalized with the task load RF to obtain an evaluation value XPZ by using a formula: , wherein d1 and d2 are preset weight coefficients. The preliminary selected personnel with the maximum selection evaluation value is selected from all the preliminary selected personnel as the task dispatching personnel to process the risk failure generated by the early warning; After the task dispatching personnel successfully solves the risk failure, maintenance performance scoring is carried out for each risk failure processed by the personnel, and the obtained score is recorded in the maintenance performance score record of the corresponding risk failure in the maintenance case of the personnel.

9. A system for intelligent analysis of printing failure data, applied to the method for intelligent analysis of printing failure data according to any one of claims 1-8, characterized in that: It includes: A fault parameter correlation analysis module: a prepress fault parameter library and a risk failure library are constructed, a scatter plot of risk failure and fault parameter is constructed, the correlation between each risk failure in the fault parameter library and each fault parameter in the fault parameter library is analyzed based on the scatter plot, a fault parameter information value is obtained, and a prepress wind evaluation parameter corresponding to each risk failure is determined based on the fault parameter information value; The specific process of analyzing the prepress wind evaluation parameter corresponding to each risk failure is as follows: For each risk failure in the risk failure library, a rectangular coordinate system of risk failure and fault parameter is established, wherein the risk failure degree is the ordinate and the fault parameter is the abscissa; then the one-to-one data points of the fault parameter and the risk failure are extracted from the data recorded in the process log of the printing production; These data points are labeled one by one in the constructed rectangular coordinate system to obtain an analysis scatter plot; The risk failure degree is divided into m intervals and denoted as C1, C2, …, Cm. m The failure parameter is divided into n intervals and denoted as A1, A2, …, An. n The number of data points falling into each intersection area of the risk failure degree interval and the failure parameter interval is counted and marked as where i represents the index of the risk failure degree interval and j represents the index of the failure parameter interval; the total number of data points in all intersection areas is obtained by using the formula: The joint probability of being in the risk failure degree interval Ci and in the failure parameter interval Aj The marginal probability of being in the risk failure degree interval Ci ;​ and the edge probability of being in the fault parameter interval Aj ; The above , , and Mutual information formula: , get the information value MI; The fault parameter information value corresponding to each fault parameter and risk failure is compared with the corresponding threshold value, if the fault parameter information value is greater than the corresponding threshold value, the fault parameter corresponding to the fault parameter information value is marked as the prepress wind evaluation parameter of the risk failure; At the same time, according to the scatter plot corresponding to each fault parameter and risk failure, the positive and negative correlation of the fault parameter and the risk failure is marked. The wind evaluation and data collection module: according to the pre-press wind evaluation parameter corresponding to the risk fault, the wind evaluation value of the risk fault is analyzed; the corresponding set of wind evaluation parameters in the middle of printing is constructed, and the collection frequency of the corresponding set of wind evaluation parameters in the middle of printing is matched according to the wind evaluation value of the risk fault; The evaluation and early warning module: set a sliding time window, calculate the middle of printing evaluation value of the middle of printing wind evaluation parameter in the sliding time window, and update the collection frequency of the middle of printing wind evaluation parameter according to the middle of printing evaluation value; The middle of printing evaluation value of each middle of printing wind evaluation parameter of the risk fault is analyzed, the fault evaluation value is obtained, and whether the risk fault early warning is judged according to the fault evaluation value; The maintenance personnel scheduling module: a maintenance personnel database is constructed, the selection evaluation value of the maintenance personnel is analyzed according to the maintenance case of the maintenance personnel and the current risk fault corresponding to the risk fault, and the maintenance personnel corresponding to the maximum selection evaluation value is selected as the task dispatch personnel.

Images