A method and system for monitoring operation of a digital printing press

Abstract

The present invention relates to the field of data processing technology, and in particular to a method and system for monitoring the operation of a digital printing press. The method comprises the following steps: collecting each operating parameter, clustering all operating parameters to obtain each cluster, and obtaining the stability of the data distribution in the cluster where each operating parameter is located; correcting the stability of the data distribution in the cluster where each operating parameter is located based on the time interval between operating parameters in the cluster where each operating parameter is located, obtaining the weighted stability of the data distribution in the cluster where each operating parameter is located; adjusting the fixed K value of each operating parameter based on the weighted stability of the data distribution in the cluster where each operating parameter is located, obtaining the adjusted K value of each operating parameter, and identifying abnormal operating parameters based on the obtained adjusted K value of each operating parameter. The present invention improves the accuracy of abnormality detection.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to an operation monitoring method and system for a digital printing press. Background Art

[0002] Digital printing presses, as core equipment in the modern printing industry, are widely used in publishing, packaging, advertising, and other fields. Their operating status directly impacts print quality. For example, insufficient pressure can lead to blurred or even incomplete images and text, unstable speed can cause uneven ink color, and temperature fluctuations can affect ink flow, resulting in poor quality of the final printed product. Therefore, it is necessary to monitor the operating status of digital printing presses. By analyzing the multidimensional data characteristics during operation, potential problems can be identified and preventive measures can be taken to avoid multiple product printing defects caused by parameter anomalies, providing strong technical support for printing companies.

[0003] Digital printing machines are prone to insufficient operating parameter accuracy and abnormal parameters due to problems such as mechanical wear, unstable electrical parameters, and aging of equipment materials. Since the local density of abnormal parameters is usually significantly different from the local density of their neighborhood data, the outlier factor of each operating parameter can be calculated using the local outlier factor algorithm (LOF) to identify abnormal points. Considering that digital printing machines contain multiple working stages in actual operation, the parameter characteristics of different stages are significantly different. For example, the operating parameters in the plate reading startup stage gradually tend to a stable state, the parameters in the printing stage tend to be balanced and regularly distributed, and the parameter stability decreases due to state transition requirements at the end of the plate changing stage. Therefore, the local data density in different working stages varies greatly. If the fixed K value in the traditional LOF algorithm is used for anomaly detection, it will not be able to accurately adapt to the data distribution characteristics of different stages, which may lead to distorted anomaly detection results and false detection or missed detection. Summary of the Invention

[0004] In order to solve the technical problem that the fixed K value in the LOF algorithm for anomaly detection cannot accurately adapt to the data distribution characteristics of different stages, which may lead to distortion of anomaly detection results and false detection or missed detection, the present invention provides an operation monitoring method and system for a digital printing press.

[0005] In a first aspect, the present invention provides a method for monitoring the operation of a digital printing press, which adopts the following technical solution:

[0006] A method for monitoring the operation of a digital printing press comprises the following steps:

[0007] Collect each operating parameter; cluster all the operating parameters to obtain each cluster; obtain the stability of the data distribution in the cluster where each operating parameter is located based on the distribution concentration of the operating parameters in the cluster where each operating parameter is located;

[0008] Obtaining each set of adjacent parameters of the time series parameter sequence of each operating parameter based on the operating parameters in the cluster where each operating parameter is located; correcting the stability of the data distribution in the cluster where each operating parameter is located based on the time interval of each set of adjacent parameters in the time series parameter sequence of each operating parameter, and obtaining the weighted stability of the data distribution in the cluster where each operating parameter is located;

[0009] According to the weighted stability of the data distribution in the cluster where each operating parameter is located, the fixed K value of each operating parameter is adjusted to obtain the adjusted K value of each operating parameter; based on the obtained adjusted K value of each operating parameter, abnormal operating parameters are identified.

[0010] The innovation of the present invention lies in first evaluating the stability of the data distribution in the cluster where the operating parameter is located by analyzing the distance deviation and standard deviation between the operating parameter and the cluster center in the cluster where the parameter is located; further weighting the stability by considering the continuous time series characteristics of the working stage, obtaining the weighted stability of the data distribution in the cluster where each operating parameter is located, and reducing the stability of the clusters whose operating parameters are not in the same working stage; further adjusting the fixed K value of each operating parameter according to the weighted stability of the data distribution in the cluster where each operating parameter is located to perform anomaly detection, so that the K value of the operating parameter when the equipment is most likely to be in a stable operating stage is larger, and the K value of the operating parameter when the equipment is in an unstable stage such as the plate reading start-up, end and plate changing stage is smaller. By dynamically adjusting the K value, it is possible to adapt to the status of the current operating stage of the equipment in real time, thereby improving the accuracy of anomaly identification.

[0011] Preferably, clustering all operating parameters to obtain each cluster includes:

[0012] An iterative self-organizing clustering algorithm is used to cluster all the collected operating parameters to obtain each cluster and the cluster center of each cluster.

[0013] Preferably, obtaining the stability of data distribution in the cluster where each operating parameter is located includes:

[0014] ;

[0015] Where, Represents the stability of the data distribution in the cluster where the i-th operating parameter is located; Represents the Euclidean distance between the i-th operating parameter and the cluster center of the cluster it belongs to; represents the mean of the Euclidean distances between all operating parameters and their cluster centers in the cluster where the i-th operating parameter is located; || represents the absolute value symbol; Represents the maximum value of the Euclidean distances between all operating parameters and their cluster centers in the cluster where the i-th operating parameter is located; Represents the standard deviation of the Euclidean distances between all operating parameters and their cluster centers in the cluster where the i-th operating parameter is located.

[0016] Preferably, the step of obtaining each group of adjacent parameters of the timing parameter sequence of each operating parameter includes:

[0017] According to the operating parameters in the cluster where the i-th operating parameter is located, a time series parameter sequence of the i-th operating parameter is obtained; and two adjacent operating parameters of the time series parameter sequence of the i-th operating parameter are recorded as a group of adjacent parameters of the time series parameter sequence of the i-th operating parameter.

[0018] Subsequently, the stability is corrected according to the temporal continuity of the parameters in the temporal parameter sequence of each operating parameter to obtain the weighted stability of the data distribution in the cluster where each operating parameter is located.

[0019] Preferably, obtaining the weighted stability of data distribution in the cluster where each operating parameter is located includes:

[0020] ;

[0021] Where, Represents the weighted stability of the data distribution in the cluster where the i-th operating parameter is located; Represents the stability of the data distribution in the cluster where the i-th operating parameter is located; Represents the number of operating parameters in the cluster where the i-th operating parameter is located; The time interval between the jth group of adjacent parameters of the timing parameter sequence representing the i-th operating parameter.

[0022] The stability of clusters whose operating parameters are not in the same working stage is reduced, and the accuracy of stability is improved.

[0023] Preferably, obtaining the adjusted K value of each operating parameter includes:

[0024] ;

[0025] Where, represents the K value after adjustment of the i-th operating parameter; Represents a fixed K value; Represents the weighted stability of the data distribution in the cluster where the i-th operating parameter is located; The slope mean of all groups of adjacent parameters representing the time series parameter sequence of the i-th operating parameter; Represents the slope between the first and last operating parameters in the timing parameter sequence of the i-th operating parameter; exp() represents an exponential function with a natural constant as the base; || represents the absolute value symbol.

[0026] It can adapt to the status of the current operation stage of the equipment in real time, thereby improving the accuracy of anomaly identification.

[0027] Preferably, the step of identifying abnormal operating parameters based on obtaining the adjusted K value of each operating parameter includes:

[0028] The adjusted K value of each operating parameter is used as the outlier factor of each operating parameter. If the outlier factor of any operating parameter is greater than 1, the operating parameter is recorded as an abnormal operating parameter.

[0029] Improved anomaly detection accuracy.

[0030] Preferably, the collecting of each operating parameter includes:

[0031] The preset sampling time is 1 second / time. A pressure sensor is installed on the printing cylinder of the digital printing press, a speed sensor is installed on the nozzle, and a temperature sensor is installed on the ink roller. During the operation of the digital printing press, the pressure data, speed data and temperature data at each moment are collected as each operating parameter.

[0032] Preferably, the step of obtaining the timing parameter sequence of the i-th operating parameter includes:

[0033] According to the order of sampling time from front to back, the operating parameters in the cluster where the i-th operating parameter is located are sorted to obtain the time series parameter sequence of the i-th operating parameter.

[0034] In a second aspect, the present invention provides an operation monitoring system for a digital printing press, which adopts the following technical solution:

[0035] An operation monitoring system for a digital printing press includes: a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the operation monitoring method for the digital printing press is implemented.

[0036] By adopting the above technical solution, the above-mentioned operation monitoring method of a digital printing machine is generated into a computer program and stored in a memory so as to be loaded and executed by a processor, thereby making a terminal device based on the memory and the processor for easy use.

[0037] The present invention has the following technical effects: the present invention first analyzes the distance deviation and standard deviation between the operating parameter and the cluster center in the cluster where it is located to evaluate the stability of the data distribution in the cluster where it is located; further considers the continuous time series characteristics of the working stage to weight the stability, obtains weighted stability, and reduces the stability of the clusters whose operating parameters are not in the same working stage; further adjusts the fixed K value of each operating parameter according to the weighted stability of the data distribution in the cluster where each operating parameter is located to perform anomaly detection, so that the K value of the operating parameter when the equipment is most likely to be in a stable operating stage is larger, and the K value of the operating parameter when the equipment is in an unstable stage such as the plate reading start-up, end and plate changing stage is smaller, which can adapt to the status of the current operating stage of the equipment in real time, thereby improving the accuracy of anomaly identification. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 This is a flow chart of a method for monitoring the operation of a digital printing press according to an embodiment of the present invention. DETAILED DESCRIPTION

[0039] The technical solutions in the embodiments of the present invention will be described clearly and completely below with reference to the accompanying drawings in the embodiments of the present invention.

[0040] The embodiment of the present invention discloses a method for monitoring the operation of a digital printing machine. Figure 1 , including steps S1 to S4:

[0041] S1: Collect each operating parameter.

[0042] In an embodiment of the present invention, the preset sampling time is 1 second / time, a pressure sensor is installed on the printing cylinder of the digital printing press, a speed sensor is installed on the nozzle, and a temperature sensor is installed on the ink roller. During the operation of the digital printing press, pressure data, speed data and temperature data at each moment are collected, and the pressure data, speed data and temperature data at each moment are used as each operating parameter.

[0043] It should be noted that, considering the influence of unit dimension in multidimensional data analysis, the pressure data, velocity data and temperature data collected in real time at all times are normalized to the maximum and minimum values ​​to eliminate the influence of dimension.

[0044] S2: Cluster all operating parameters to obtain each cluster, and obtain the stability of the data distribution in the cluster where each operating parameter is located; according to the time interval between the operating parameters in the cluster where each operating parameter is located, correct the stability of the data distribution in the cluster where each operating parameter is located, and obtain the weighted stability of the data distribution in the cluster where each operating parameter is located.

[0045] It should be noted that a digital printing machine will go through multiple different working stages during actual operation, and its working stages can be divided into a start printing stage, a stable printing stage and a plate changing stage. The operating parameters collected in the stable stage need to remain stable (printing pressure, nozzle operating speed and ink roller temperature data) to maintain consistent printing quality. The operating parameters collected in the start printing stage and the plate changing stage are usually in dynamic change (such as acceleration at the start of printing, equipment adjustment when changing the plate), which is significantly different from the parameters in the stable printing stage. Therefore, the present invention first clusters all the collected operating parameters, and quantifies the stability of the data distribution in the cluster where each operating parameter is located through the distribution concentration of the operating parameters in the cluster where each operating parameter is located. Specifically, if the distribution of the operating parameters in the cluster where any operating parameter is located is more concentrated, it indicates that the similarity of the operating parameters in the cluster cluster is higher, which further reflects that the stability of the data distribution in the cluster cluster where the operating parameter is located is stronger.

[0046] In the embodiment of the present invention, an iterative self-organizing clustering algorithm is used to cluster all collected operating parameters to obtain each cluster and the cluster center of each cluster;

[0047] Get the stability of the data distribution in the cluster where each running parameter is located:

[0048] ;

[0049] Where, Represents the stability of the data distribution in the cluster where the i-th operating parameter is located; Represents the Euclidean distance between the i-th operating parameter and the cluster center of the cluster it belongs to; represents the mean of the Euclidean distances between all operating parameters and their cluster centers in the cluster where the i-th operating parameter is located; || represents the absolute value symbol; Represents the maximum value of the Euclidean distances between all operating parameters and their cluster centers in the cluster where the i-th operating parameter is located; Represents the standard deviation of the Euclidean distance between all operating parameters and their cluster centers in the cluster where the i-th operating parameter is located;

[0050] The smaller the value of , the more concentrated the distribution of the operating parameters in the cluster where the i-th operating parameter is located, which means that the data distribution in the cluster where the i-th operating parameter is located is more stable; The smaller the value of , the more consistent the Euclidean distances between all operating parameters and their cluster centers in the cluster where the i-th operating parameter is located. The higher the confidence level of the value of

[0051] The smaller the value of , the The closer to 1, the The smaller the value of , the greater the stability of the data distribution in the cluster where the i-th operating parameter is located; conversely, The larger the value of The closer to 0, the The less credible the value of The closer the value of is to 1, the less stable the data distribution in the cluster where the i-th operating parameter is located.

[0052] It should be noted that in the stable printing stage, each operating parameter needs to maintain a stable equilibrium state to ensure the best printing quality, that is, the higher the parameter stability of the cluster where any operating parameter is located, the greater the probability that the cluster belongs to the stable stage. However, clustering is clustering of the operating parameters collected in all stages. There may be situations where the operating parameters of other working stages are similar to those of the stable printing stage. In addition, when clustering the operating parameters, the time continuity between the operating parameters is not considered. It is very likely that the operating parameters in the stable printing stage will be clustered into one category with the operating parameters in the start printing stage and the plate changing stage. If the time interval between the operating parameters in the cluster is large, it means that the operating parameters in the cluster are not in the same working stage. Therefore, the present invention corrects the stability of the data distribution in the cluster where each operating parameter is located according to the time interval between the operating parameters in the cluster.

[0053] In the embodiment of the present invention, the operating parameters in the cluster where the i-th operating parameter is located are sorted according to the order of sampling time from front to back to obtain the time series parameter sequence of the i-th operating parameter; two adjacent operating parameters in the time series parameter sequence of the i-th operating parameter are recorded as a group of adjacent parameters of the time series parameter sequence of the i-th operating parameter;

[0054] Get the weighted stability of the data distribution in the cluster where each operating parameter belongs:

[0055] ;

[0056] Where, Represents the weighted stability of the data distribution in the cluster where the i-th operating parameter is located; Represents the stability of the data distribution in the cluster where the i-th operating parameter is located; Represents the number of operating parameters in the cluster where the i-th operating parameter is located; The time interval between the jth group of adjacent parameters of the timing parameter sequence representing the i-th operating parameter;

[0057] Reflects the degree of dispersion of the operating parameters in the cluster where the i-th operating parameter is located in time distribution. The larger the value, the more dispersed the operating parameters in the cluster are in time distribution. The longer the time interval between adjacent operating parameters is, the greater the probability that the operating parameters in the cluster are in different working stages of the digital printing press, thus reducing the stability of the data distribution in the cluster where the i-th operating parameter is located.

[0058] On the contrary, the smaller the value is, the more concentrated the temporal distribution of the operating parameters in the cluster is, and the smaller the time interval between adjacent operating parameters is, which means that the operating parameters in the cluster are in the same working stage of the digital printing machine, indicating that the stability of the data distribution in the cluster where the i-th operating parameter is located is more accurate.

[0059] S3: According to the weighted stability of the data distribution in the cluster where each operating parameter is located, the fixed K value of each operating parameter is adjusted to obtain the adjusted K value of each operating parameter.

[0060] It should be noted that the greater the weighted stability of the data distribution in the cluster where any operating parameter is located, the more likely the operating parameters in the cluster where the operating parameter is located are to be in a unified working stage. Taking into account the data in the initial plate reading operation stage interval of this process, a relatively stable cluster relative to this stage will also be evaluated. However, in fact, the stage is not the stage where the operating parameters maintain balance (normal printing stage). Therefore, it is necessary to further consider the trend intensity of the data point changes in the cluster where the current data point is located, so as to distinguish the clusters with relatively stable parameters in the non-printing stage and ensure that the adaptively adjusted K value meets the scene characteristics.

[0061] In the embodiment of the present invention, a fixed K value is preset. In other embodiments, the implementer may preset The value of .

[0062] Get the K value after each operating parameter is adjusted;

[0063] ;

[0064] Where, represents the K value after adjustment of the i-th operating parameter; Represents a fixed K value; Represents the weighted stability of the data distribution in the cluster where the i-th operating parameter is located; The slope mean of all groups of adjacent parameters representing the time series parameter sequence of the i-th operating parameter; represents the slope between the first and last operating parameters in the time series parameter sequence of the i-th operating parameter; exp() represents an exponential function with a natural constant as the base; || represents the absolute value symbol; The purpose is to The value of is controlled between 0.5-1.5;

[0065] The smaller the value of , the smaller the variation between all adjacent parameters of the time series parameter sequence of the i-th operating parameter is. In other words, the operating parameter changes slightly over a period of time. It is believed that the equipment is likely to be in a stable operating stage. In this case, the K value should be increased. This will make it easier for the algorithm to identify operating parameters that slightly deviate from the stable state as abnormal points in this stage.

[0066] The larger the value of , the more stable the data distribution in the cluster where the i-th operating parameter is located. In this case, the K value should be increased, so that the algorithm can more easily identify operating parameters that slightly deviate from the stable state as outliers at this stage. On the contrary, The smaller the value of , the more likely the data in the cluster where the i-th operating parameter is located is to be in the plate reading start-up and plate changing stage. Considering that the operating parameters fluctuate greatly in the unstable stage, reducing the K value can improve the algorithm's tolerance to data fluctuations, prevent normal fluctuation data from being misjudged as anomalies, and thus reduce false detections.

[0067] S4: Identify abnormal operating parameters based on the K value after obtaining each operating parameter adjustment.

[0068] In an embodiment of the present invention, the adjusted K value of each operating parameter is used as the outlier factor of each operating parameter. If the outlier factor of any operating parameter is greater than 1, the operating parameter is recorded as an abnormal operating parameter. At this time, an early warning should be given in time to prompt the operator to pay attention to the equipment operating status and check whether there are equipment failures, parameter offsets and other problems.

[0069] An embodiment of the present invention further discloses an operation monitoring system for a digital printing press, including a processor and a memory. The memory stores computer program instructions. When the computer program instructions are executed by the processor, an operation monitoring method for a digital printing press according to the present invention is implemented.

Claims

1. A method for monitoring the operation of a digital printing press, characterized in that: Including steps: Collect each operating parameter; cluster all operating parameters to obtain each cluster; According to the distribution concentration of the operating parameters in the cluster where each operating parameter is located, the stability of the data distribution in the cluster where each operating parameter is located is obtained, including: , Represents the stability of the data distribution in the cluster where the i-th operating parameter is located, Represents the Euclidean distance between the i-th operating parameter and the cluster center of the cluster it belongs to; 、 Respectively represent the mean and standard deviation of the Euclidean distance between all operating parameters and their cluster centers in the cluster where the i-th operating parameter is located; || represents the absolute value symbol; Represents the maximum value of the Euclidean distances between all operating parameters and their cluster centers in the cluster where the i-th operating parameter is located; Obtaining each set of adjacent parameters of the time series parameter sequence of each operating parameter based on the operating parameters in the cluster where each operating parameter is located; and correcting the stability of the data distribution in the cluster where each operating parameter is located based on the time interval between each set of adjacent parameters in the time series parameter sequence of each operating parameter to obtain the weighted stability of the data distribution in the cluster where each operating parameter is located, including: , Represents the weighted stability of the data distribution in the cluster where the i-th operating parameter is located, Represents the number of operating parameters in the cluster where the i-th operating parameter is located, The time interval between the jth group of adjacent parameters of the timing parameter sequence representing the i-th operating parameter; According to the weighted stability of the data distribution in the cluster where each operating parameter is located, the fixed K value of each operating parameter is adjusted to obtain the adjusted K value of each operating parameter, including: , represents the K value after the adjustment of the i-th operating parameter, represents a fixed K value, represents the slope mean of all groups of adjacent parameters of the time series parameter sequence of the i-th operating parameter, The slope between the first and last operating parameters in the time series parameter sequence representing the i-th operating parameter is represented. exp() represents an exponential function with a natural constant as the base. Abnormal operating parameters are identified based on the K value after adjusting each operating parameter.

2. The operation monitoring method of a digital printing press according to claim 1, characterized in that: The clustering of all operating parameters is performed to obtain each cluster, including: An iterative self-organizing clustering algorithm is used to cluster all the collected operating parameters to obtain each cluster and the cluster center of each cluster.

3. The operation monitoring method of a digital printing press according to claim 1, characterized in that: The step of obtaining each group of adjacent parameters of the timing parameter sequence of each operating parameter includes: According to the operating parameters in the cluster where the i-th operating parameter is located, a time series parameter sequence of the i-th operating parameter is obtained; and two adjacent operating parameters of the time series parameter sequence of the i-th operating parameter are recorded as a group of adjacent parameters of the time series parameter sequence of the i-th operating parameter.

4. The operation monitoring method of a digital printing press according to claim 1, characterized in that: The step of identifying abnormal operating parameters based on obtaining the adjusted K value of each operating parameter includes: The adjusted K value of each operating parameter is used as the outlier factor of each operating parameter. If the outlier factor of any operating parameter is greater than 1, the operating parameter is recorded as an abnormal operating parameter.

5. The operation monitoring method of a digital printing press according to claim 1, characterized in that: The collecting of each operating parameter includes: The preset sampling time is 1 second / time. A pressure sensor is installed on the printing cylinder of the digital printing press, a speed sensor is installed on the nozzle, and a temperature sensor is installed on the ink roller. During the operation of the digital printing press, the pressure data, speed data and temperature data at each moment are collected as each operating parameter.

6. The operation monitoring method of a digital printing press according to claim 3, characterized in that: The step of obtaining the timing parameter sequence of the i-th operating parameter includes: According to the order of sampling time from front to back, the operating parameters in the cluster where the i-th operating parameter is located are sorted to obtain the time series parameter sequence of the i-th operating parameter.

7. An operation monitoring system for a digital printing press, characterized in that: include: A processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, an operation monitoring method of a digital printing press according to any one of claims 1 to 6 is implemented.

Images