Operation and maintenance data analysis method and system of a folder gluer

Abstract

The application relates to the technical field of data processing, in particular to a paste box machine operation and maintenance data analysis method and system. The method comprises the following steps: acquiring a mutation degree value of a monitoring parameter of the paste box machine at a current moment relative to a previous moment; determining a correlation value at the current moment, and taking the product of the correlation value and the mutation degree value as an abnormality degree value at the current moment; clustering the monitoring parameters of the paste box machine at the current moment and historical moments to obtain an initial feature distance at the current moment; determining an adjustment coefficient by using the abnormality degree values at the current moment and the previous moment, and adjusting the initial feature distance by using the adjustment coefficient to obtain a target feature distance; and determining whether a belt between a main drive wheel and a driven wheel is damaged by using the target feature distance. Through the above technical scheme, operation and maintenance analysis of the belt of the paste box machine can be accurately realized.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a method and system for analyzing operation and maintenance data of a box-gluing machine. Background Technology

[0002] The carton gluing machine processes paperboard raw materials into finished carton products through stages such as paper feeding, pre-folding, bottom hooking, forming, and pressing. The main drive wheel, driven wheel, servo motor, and belt of the carton gluing machine together constitute the key transmission system of the carton gluing machine.

[0003] The main drive wheel is directly driven by a servo motor, and the driven wheel is connected to the main drive wheel by a belt, which plays a role in supporting the carton and synchronizing its movement. The belt transmits power between the main drive wheel and the driven wheel, ensuring the smooth movement of the carton on the conveyor belt. The servo motor plays a role in precise control throughout the process, while the belt ensures the stability and synchronicity of power transmission.

[0004] As the power source for the main drive wheel, the servo motor ensures that the main drive wheel rotates at a stable speed by precisely controlling its speed and torque. The servo motor is also responsible for adjusting operating parameters according to production needs. For example, the servo motor can operate in acceleration, deceleration or constant speed modes to adapt to different types of paper box processing.

[0005] When the conveyor belt of a gluing machine ages, its friction decreases, reducing the power transmission efficiency between the belt and the main drive wheel and driven wheel. This results in unstable conveying speed of the cartons, potentially causing cartons to pile up or jam. Conveyor belt aging also increases the instantaneous slip rate between the main drive wheel and driven wheel, affecting the conveying accuracy of the cartons. Therefore, it is necessary to analyze the operation and maintenance data of the gluing machine to monitor its conveyor belt. Summary of the Invention

[0006] To enable monitoring of the conveyor belt of a box-gluing machine, this application provides a method and system for analyzing operation and maintenance data of a box-gluing machine.

[0007] According to a first aspect of the embodiments of this application, a method for analyzing operation and maintenance data of a gluing machine is provided, comprising: acquiring the degree of change of monitoring parameters of the gluing machine at the current moment relative to the previous moment; the sub-monitoring parameters of the monitoring parameters include the instantaneous slip rate between the main drive wheel and the driven wheel of the gluing machine, the frequency of the carton passing through the detection point, and the operating current of the servo motor of the main drive wheel; determining the correlation value at the current moment, and determining the degree of abnormality at the current moment based on the correlation value and the degree of change; the correlation value is used to characterize the degree of negative correlation between the instantaneous slip rate and the operating current and the frequency; Cluster the monitoring parameters of the box-gluing machine at the current time and historical time to obtain the initial feature distance at the current time. The initial feature distance is equal to the minimum value among the distances from the monitoring parameters at the current time to the multiple density peak data points after clustering. The ratio of the anomaly value at the current time to the anomaly value at the previous time is used as an adjustment coefficient, and the initial feature distance is adjusted using the adjustment coefficient to obtain the target feature distance. The data points at the current time in the clustering results are re-clustered using the target feature distance, and the belt between the main drive wheel and the driven wheel is determined based on the allocation result.

[0008] In this way, the adjustment coefficient takes into account the instantaneous slip rate of the gluing machine and the degree of negative correlation between the working current and the frequency, which can more accurately adjust the initial characteristic distance and thus more accurately determine whether the belt of the gluing machine is damaged.

[0009] Optionally, the mutation degree value is determined in the following way: for the target sub-monitoring parameter among multiple sub-monitoring parameters, the change in the value of the target sub-monitoring parameter at the current time and the change in the value at the previous time are determined, and the ratio of the change to the maximum value of the target sub-monitoring parameter at historical time is taken as the relative change rate; the sum of the relative change rates of all sub-monitoring parameters is taken as the mutation degree value.

[0010] Optionally, the correlation value at the current moment is determined in the following ways: determine the first relative deviation between the instantaneous slip rate at the current moment and the historical average instantaneous slip rate, and the second relative deviation between the operating current at the current moment and the historical average operating current, and determine the third relative deviation between the frequency at the current moment and the historical average frequency; determine the first difference between the first relative deviation and the third relative deviation, and determine the second difference between the second relative deviation and the third relative deviation, and determine the average difference between the first difference and the second difference; determine the similarity values ​​of the monitoring parameters of the gluing machine at multiple historical moments before the current moment, and use the average difference as the base and the similarity value as the exponent, and use the obtained exponential calculation result as the correlation value.

[0011] By comparing the frequencies of the current moment and multiple historical moments prior to the current moment, and by comparing the instantaneous slip rates of the current moment and multiple historical moments prior to the current moment, the obtained correlation values ​​can be used to better distinguish between the two situations where the belt between the main drive wheel and the driven wheel is damaged and undamaged.

[0012] Optionally, the similarity value is determined as follows: multiple data points are obtained by mapping the monitoring parameters of the box-gluing machine at multiple historical moments before the current moment to a three-dimensional spatial coordinate system; the three-dimensional spatial coordinate system has the average instantaneous slip rate, average operating current, and average frequency as different dimensions; the distance between two data points in a data point pair is determined, and the normalized result of the reciprocal of the average distance of different data point pairs is used as the similarity value; the data point pair includes two different data points among the multiple data points in the three-dimensional spatial coordinate system.

[0013] In this way, the similarity value can reflect the stability of various monitoring parameters of the gluing machine over a period of time before the current moment, thus reflecting the reference value of the period of time before the current moment to the current moment. Therefore, the similarity value can help to better determine whether the belt between the main drive wheel and the driven wheel of the gluing machine is damaged.

[0014] Optionally, cluster the monitoring parameters of the box-gluing machine at the current moment and historical moments to obtain the initial feature distance at the current moment, including: mapping the monitoring parameters of the box-gluing machine at the current moment and multiple historical moments before the current moment to a three-dimensional spatial coordinate system to obtain multiple data points; the three-dimensional spatial coordinate system has the average instantaneous slip rate, average operating current, and average frequency as different dimensions; clustering the multiple data points in the three-dimensional spatial coordinate system, and taking the data points with local density values ​​greater than a preset density threshold in the clustering results as density peak data points; taking the minimum value among the distances from the data point corresponding to the current moment to the multiple density peak data points as the initial feature distance at the current moment.

[0015] In this way, the monitoring parameters of the current moment and multiple historical moments before the current moment are mapped to a three-dimensional spatial coordinate system to obtain multiple data points, and the initial feature distance of the current moment is determined. The initial feature distance can reflect the probability of belt damage between the main drive wheel and the driven wheel of the gluing box machine.

[0016] Optionally, the target feature distance can be obtained by adjusting the initial feature distance using an adjustment coefficient, including: using the adjustment coefficient as an exponent and the initial feature distance as the base for exponential operation, and using the result of the exponential operation as the target feature distance.

[0017] In this way, if the abnormality value at the current moment is greater than the abnormality value at the previous moment, it means that the belt between the main drive wheel and the driven wheel of the box gluing machine is likely to be damaged. Therefore, a target feature distance greater than or equal to the initial feature distance can be determined based on the initial feature distance, and it is more likely that the data point at the current moment will be identified as an abnormal data point.

[0018] Optionally, the data points at the current time in the clustering results can be re-assigned to clusters using the target feature distance, including: if the target feature distance is less than or equal to a preset distance threshold, the data points at the current time are assigned to the nearest cluster; if the target feature distance is greater than the preset distance threshold, the data points at the current time are treated as outlier data points.

[0019] Optionally, determining whether the belt between the main drive wheel and the driven wheel is damaged based on the allocation result includes: if the allocation result indicates that the data point at the current moment has been assigned to a cluster, determining that the belt between the main drive wheel and the driven wheel is not damaged; or, if the allocation result indicates that the data point at the current moment has not been assigned to a cluster, determining that the belt between the main drive wheel and the driven wheel is damaged.

[0020] Optionally, the method further includes: if it is determined that the belt between the main drive wheel and the driven wheel is damaged, sending a prompt message to the terminal device of the user to which the box gluing machine is bound; the prompt message is used to indicate that the belt between the main drive wheel and the driven wheel is damaged.

[0021] In this way, sending a notification message to the terminal device of the user to which the box gluing machine is bound makes it easier for the user to deal with the damaged belt between the main drive wheel and the driven wheel of the box gluing machine.

[0022] According to a second aspect of the embodiments of this application, a maintenance data analysis system for a box-gluing machine is provided, comprising: a processor and a memory, wherein the memory stores computer program instructions, and the computer program instructions, when executed by the processor, implement the steps of the maintenance data analysis method for a box-gluing machine provided in the first aspect of this application.

[0023] The technical solutions provided by the embodiments of this application may include the following beneficial effects: the adjustment coefficient at the current moment takes into account the instantaneous slip rate between the main drive wheel and the driven wheel of the gluing machine, the frequency of the carton passing through the detection point, and the change of the working current of the servo motor of the main drive wheel at different times before the current moment. The target feature distance is obtained by adjusting the initial feature distance through the adjustment coefficient. The target feature distance can better reflect the difference between the monitoring parameters of the gluing machine and the monitoring parameters at the current moment, thereby realizing a more refined identification of the damage to the belt between the main drive wheel and the driven wheel, and facilitating the treatment of the damaged belt in the early stage of belt damage between the main drive wheel and the driven wheel.

[0024] Carton gluing machines can typically adaptively control the operating current of the servo motor based on the frequency at which the carton passes through the detection point. This allows the instantaneous slip rate to change accordingly with the change in operating current. The adjustment coefficient at the current moment takes into account the degree of negative correlation between the instantaneous slip rate of the carton gluing machine and the operating current and frequency, thus avoiding misjudgments about whether the belt is damaged.

[0025] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0026] Figure 1 This is a flowchart illustrating a method for analyzing maintenance data of a box-gluing machine according to an exemplary embodiment; Figure 2 This is a schematic diagram illustrating the structure of an operation and maintenance data analysis system for a box-gluing machine according to an exemplary embodiment. Detailed Implementation

[0027] First, a brief introduction to the application scenario of this application embodiment will be given. In the application scenario of this application, the belt in the gluing machine is used to convey the cardboard raw materials during the processing. Damage to the belt will directly affect the processing of the cardboard raw materials by the processing parts in the gluing machine, so that damage to the conveyor belt will directly affect the quality of the paper boxes produced by the gluing machine. Therefore, it is necessary to analyze the operation and maintenance data of the gluing machine to monitor the conveyor belt, thereby ensuring the quality of the paper boxes produced by the gluing machine.

[0028] To address the aforementioned technical problems, embodiments of this application provide a method and system for analyzing maintenance data of a carton gluing machine. Figure 1 This is a flowchart illustrating a maintenance data analysis method for a box-gluing machine according to an exemplary embodiment, such as... Figure 1 As shown, the method includes the following steps.

[0029] In step S101, the degree of change of the monitoring parameters of the box gluing machine at the current time relative to the previous time is obtained.

[0030] The sub-monitoring parameters include the instantaneous slip rate between the main drive wheel and the driven wheel of the gluing machine, the frequency of the carton passing through the detection point, and the operating current of the servo motor of the main drive wheel; the frequency of the carton passing through the detection point can be collected by setting photoelectric sensors at the detection point of the gluing machine.

[0031] A current sensor can be used to obtain the operating current of the servo motor; an angular velocity sensor can be used to obtain the rotational speed of the main drive wheel and the driven wheel. Based on the rotational speed of the main drive wheel and the driven wheel, the instantaneous slip rate between the main drive wheel and the driven wheel of the gluing machine can be determined.

[0032] When there are multiple driven wheels, the instantaneous slip ratio can be determined based on the average rotational speed of the multiple driven wheels. The instantaneous slip ratio can be equal to the ratio between the difference in rotational speed between the main drive wheel and the driven wheels of the gluing machine and the rotational speed of the main drive wheel.

[0033] The degree of change of the monitoring parameter at the current moment relative to the previous moment is used to characterize the degree of change of the monitoring parameter at the current moment relative to the monitoring parameter at the previous moment; the higher the degree of change of the monitoring parameter at the current moment relative to the previous moment, the greater the probability that the conveyor belt of the gluing machine will be damaged at the current moment.

[0034] In one embodiment, the mutation degree value is determined as follows: for a target sub-monitoring parameter among multiple sub-monitoring parameters, the change in the value of the target sub-monitoring parameter at the current time compared to its value at the previous time is determined, and the ratio of the change to the maximum value of the target sub-monitoring parameter at a historical time is taken as the relative change rate; the sum of the relative change rates of all sub-monitoring parameters is taken as the mutation degree value.

[0035] For example, the mutation level value is determined in the following way: ,in, Let be the mutation level value of the gluing machine at time t, norm be the normalization function, and n be the number of sub-monitoring parameters. as well as The values ​​of the a-th sub-monitoring parameter of the box gluing machine are taken at time t and time t-1, respectively. Let be the maximum value of the a-th sub-monitoring parameter of the box gluing machine at any historical moment. To take the absolute value.

[0036] The normalization function `norm` is used to normalize variables to the range of 0 to 1. For example, normalization functions can be min-max normalization, logarithmic transformation, arctangent function, and sigmoid function.

[0037] In the formula for calculating the degree of mutation, using the maximum value of the a-th sub-monitoring parameter of the gluing machine at a historical moment as the denominator can normalize the sub-monitoring parameter, ensuring that the normalized value is between 0 and 1. By comparing the differences of the same sub-monitoring parameter of the box gluing machine at adjacent moments, it can be seen that the degree of change of the same sub-monitoring parameter of the box gluing machine at the current moment is higher.

[0038] The three sub-monitoring parameters of the gluing machine are the instantaneous slip rate between the main drive wheel and the driven wheel, the frequency of the paper box passing through the detection point, and the operating current frequency of the servo motor of the main drive wheel. Compared with monitoring the belt of the gluing machine only by the frequency of the paper box output by the gluing machine, the instantaneous slip rate and operating current of the gluing machine can also reflect the working status of the gluing machine. Therefore, it is possible to better monitor the operating status of the gluing machine.

[0039] Meanwhile, the working status of other components besides the belt can affect or reflect the working status of the belt. Therefore, considering the working status of other components connected to or in contact with the belt can better monitor the belt of the gluing machine.

[0040] In this way, by comparing the difference between the sub-monitoring parameter at the current moment and the same sub-monitoring parameter at the previous moment, the obtained mutation degree value can better characterize the degree of mutation in the monitoring parameters of the gluing box machine.

[0041] In step S102, the correlation value at the current moment is determined, and the product of the correlation value and the mutation degree value is used as the abnormality degree value at the current moment.

[0042] The correlation value is used to characterize the degree of negative correlation between the instantaneous slip rate and the operating current and the frequency at which the carton passes the detection point; the control module of the gluing machine can usually adaptively adjust the operating current of the servo motor according to the frequency of the carton output by the gluing machine.

[0043] For example, when controlling the gluing machine to work at a constant speed, if the actual output frequency of the paper box is lower than the predetermined frequency, the control module can increase the working current of the servo motor to ensure the transmission power of the paper box and compensate for the transmission speed of the paper box carried by the belt.

[0044] When the belt of the gluing machine is damaged, the friction coefficient of the belt decreases, which reduces the transmission force of the belt on the paper boxes being transported. The servo motor may require a larger compensation amount to achieve the same level of compensation for the transmission speed of the paper boxes being transported by the belt, in order to maintain the stability of the transmission speed of the paper boxes.

[0045] In the event of belt damage in a gluing machine, in order to achieve the same level of compensation for the conveying speed of the cartons transported by the belt, as the compensation amount of the servo motor increases and the friction coefficient of the belt decreases, resulting in a decrease in the conveying force of the belt on the transported cartons, the instantaneous slip rate between the main drive wheel and the driven wheel of the gluing machine may increase more significantly. This causes the instantaneous slip rate and the operating current to exhibit a greater negative correlation with the frequency.

[0046] It is evident that, compared to the case where the belt of the gluing machine is not aged, the instantaneous slip rate and operating current show a stronger negative correlation with frequency when the belt of the gluing machine is aged. The correlation value can distinguish between the two cases of belt damage and belt undamaged.

[0047] In one embodiment, the correlation value at the current moment is determined as follows: determining the first relative deviation between the instantaneous slip rate at the current moment and the historical average instantaneous slip rate, and the second relative deviation between the operating current at the current moment and the historical average operating current, and determining the third relative deviation between the frequency at the current moment and the historical average frequency; determining the first difference between the first relative deviation and the third relative deviation, and determining the second difference between the second relative deviation and the third relative deviation, and determining the average difference between the first difference and the second difference; determining the similarity values ​​of the monitoring parameters of the gluing machine at multiple historical moments prior to the current moment, using the average difference as the base and the similarity value as the exponent, and using the obtained exponential calculation result as the correlation value.

[0048] For example, the correlation value at the current moment can be determined in the following way: Where P is the correlation value at the current time, and norm is the normalization function; , as well as The instantaneous slip rate, operating current, and frequency of the gluing machine at the current moment are, in order. , as well as The values ​​are, in order, the average instantaneous slip rate, average operating current, and average frequency at multiple historical moments prior to the current moment; C is the similarity value of the monitoring parameters of the gluing machine at multiple historical moments prior to the current moment; the similarity value is used to characterize the similarity of the monitoring parameters between the current moment and multiple historical moments prior to the current moment.

[0049] Since the hyperbolic tangent function takes values ​​between -1 and 1, the normalization process in the correlation value at the current moment can be achieved by using the sum of 0.5 times and 0.5 of the result of the hyperbolic tangent function tanh on the object to be normalized.

[0050] The gluing machine can monitor multiple historical moments before the current moment, such as 10 historical moments adjacent to the current moment. By observing the performance of different sub-monitoring parameters of the gluing machine at multiple historical moments before the current moment, the degree of negative correlation between instantaneous slip rate and operating current and frequency can be determined.

[0051] In the formula for calculating the correlation value at the current moment, The units for both the numerator and denominator are instantaneous slip rate. The units for both the numerator and denominator in the formula are operating current. The units of the numerator and denominator are both frequency, which can eliminate the influence of units after calculation and facilitate the determination of the degree of negative correlation between instantaneous slip rate and operating current and frequency.

[0052] The similarity values ​​of the monitoring parameters of the box gluing machine at multiple historical moments before the current moment are used to characterize the similarity of the monitoring parameters of the box gluing machine at multiple historical moments before the current moment.

[0053] The higher the similarity value, the more stable the monitoring parameters of the gluing machine were over the previous historical time. The more likely the belt of the gluing machine is to be in good condition at the current moment. Therefore, by considering the similarity value, it is possible to better determine whether the belt of the gluing machine is damaged.

[0054] If the frequency at which the gluing machine outputs paper boxes at the current moment is greater than the average frequency at multiple historical moments prior to the current moment, it indicates that the frequency at which the gluing machine outputs paper boxes at the current moment is showing an upward trend compared to the average frequency.

[0055] Conversely, if the frequency at which the gluing machine outputs paper boxes at the current moment is less than the average frequency of the gluing machine at several previous historical moments, it indicates that the frequency of the gluing machine at the current moment is decreasing compared to the average frequency.

[0056] If the instantaneous slip rate of the gluing machine at the current moment is greater than the average instantaneous slip rate of the gluing machine at multiple historical moments before the current moment, it indicates that the instantaneous slip rate of the gluing machine at the current moment shows an upward trend compared to the average instantaneous slip rate.

[0057] If the frequency of the gluing machine at the current moment shows an upward trend compared to the average frequency, and the instantaneous slip rate of the gluing machine at the current moment shows an upward trend compared to the average instantaneous slip rate, it indicates that the belt of the gluing machine is less affected by the transmission effect of the carton at the current moment; or, the belt of the gluing machine is not affected by the transmission effect of the carton at the current moment, and a smaller correlation value can be determined after normalization.

[0058] If the frequency at which the gluing machine outputs paper boxes varies significantly at different times, while the instantaneous slip rate of the gluing machine remains relatively stable at different times, it indicates that the gluing machine has not been affected as a whole. Furthermore, since the belt's impact on the gluing machine is usually as a whole, it suggests that the probability or extent of belt damage to the gluing machine is low.

[0059] By comparing the frequency at which the gluing machine outputs paper boxes at the current moment with the average frequency at multiple historical moments prior to the current moment, and by comparing the instantaneous slip rate at the current moment with the average instantaneous slip rate at multiple historical moments prior to the current moment, the obtained correlation value can help determine whether the belt of the gluing machine is damaged.

[0060] In one embodiment, the similarity value is determined as follows: multiple data points are obtained by mapping the monitoring parameters of the box-gluing machine at multiple historical moments prior to the current moment to a three-dimensional spatial coordinate system; the three-dimensional spatial coordinate system has the average instantaneous slip rate, average operating current, and average frequency as different dimensions; the distance between two data points in a data point pair is determined, and the normalized result of the reciprocal of the average distance of different data point pairs is used as the similarity value; the data point pair includes two different data points among the multiple data points in the three-dimensional spatial coordinate system.

[0061] For example, the past 10 historical moments of the box gluing machine before the current moment can be considered as multiple historical moments. These 10 historical moments can correspond to 10 data points in a three-dimensional coordinate system. The distance between any two data points in the three-dimensional coordinate system can be obtained in 45 combinations. Different combinations correspond to different two data points among the 10 data points. The average of the 45 distances corresponding to these 45 combinations can be used as the average distance between any two data points.

[0062] The average distance between any two data points reflects the dispersion of the monitoring parameters of the gluing machine at multiple historical moments before the current moment; the reciprocal of the average distance between any two data points reflects the stability of the monitoring parameters of the gluing machine at multiple historical moments before the current moment.

[0063] The higher the stability of the monitoring parameters of the gluing machine over multiple historical moments prior to the current moment, the higher the reference value of the monitoring parameters over multiple historical moments prior to the current moment. When the monitoring parameters of the gluing machine at the current moment fluctuate significantly compared to the monitoring parameters at the previous moment, it is more likely that the belt of the gluing machine may be damaged at the current moment, affecting its ability to transport paper boxes.

[0064] For example, if the monitoring parameters of the gluing machine are relatively stable between 12:00:00 and 12:02:00, but the monitoring parameters of the gluing machine at 12:02:01 fluctuate more than those at 12:02:00, it indicates that the conveyor belt of the gluing machine is likely to be affected or abnormal at 12:02:01. Therefore, it is convenient to identify the abnormality of the monitoring parameters of the gluing machine at an early stage when there is an abnormality in the conveyor belt of the gluing machine.

[0065] The lower the stability of the monitoring parameters of the gluing machine over multiple historical moments prior to the current moment, the lower the reference value of the monitoring parameters over multiple historical moments prior to the current moment. When the monitoring parameters of the gluing machine at the current moment fluctuate significantly compared to the monitoring parameters at the previous moment, the probability that the belt of the gluing machine is actually damaged at the current moment rather than being affected by other components is smaller. Therefore, a smaller correlation value can be determined.

[0066] In step S103, the monitoring parameters of the gluing machine at the current time and at historical times are clustered to obtain the initial feature distance at the current time.

[0067] The initial feature distance is equal to the minimum distance among the current monitoring parameter to the multiple density peak data points after clustering.

[0068] Clustering of monitoring parameters of the box gluing machine at the current and historical times can be achieved using density peak clustering. Density peak clustering is a density-based clustering algorithm. The core idea of ​​density peak clustering is to use the density peak data points with higher density as cluster centers. Density peak data points are data points with local density greater than a preset density threshold among all data points participating in the clustering.

[0069] Based on the minimum distance from the data point participating in the clustering to the density peak data point, the corresponding cluster is assigned to the data point participating in the clustering, thereby achieving clustering of all data points participating in the clustering.

[0070] The preset density threshold can be, for example, the local density corresponding to the data points ranked in the top 10% to top 15% of the local density; the local density of a data point can be determined based on the number and distance of other data points in the neighborhood of the data point.

[0071] For example, when there are 6 density peak data points, the minimum distance among the 6 density peak data points from the data point corresponding to the monitoring parameter at the current moment can be used as the initial feature distance at the current moment.

[0072] Clustering the monitoring parameters of the gluing machine at the current moment and at historical moments yields the initial feature distance at the current moment. The initial feature distance can initially reflect whether the monitoring parameters of the gluing machine deviate from the normal range at the current moment, providing data support for subsequent determination of whether there are any abnormalities in the gluing machine.

[0073] In one embodiment, clustering the monitoring parameters of the box-gluing machine at the current moment and at historical moments to obtain the initial feature distance at the current moment includes: mapping the monitoring parameters of the box-gluing machine at the current moment and at multiple historical moments before the current moment to a three-dimensional spatial coordinate system to obtain multiple data points; the three-dimensional spatial coordinate system has the average instantaneous slip rate, average operating current, and average frequency as different dimensions; clustering the multiple data points in the three-dimensional spatial coordinate system, and taking the data points in the clustering results whose local density values ​​are greater than a preset density threshold as density peak data points; taking the minimum value among the distances from the data point corresponding to the current moment to the multiple density peak data points as the initial feature distance at the current moment.

[0074] By mapping the monitoring parameters of the gluing machine at the current moment and at historical moments to a three-dimensional spatial coordinate system, the instantaneous slip rate between the main drive wheel and the driven wheel of the gluing machine at different moments, the frequency of the carton passing through the detection point, and the working current of the servo motor of the main drive wheel can be visualized. Through clustering, high-density areas in the monitoring parameters of the gluing machine at different moments can be identified, that is, the more normal data range in multiple moments.

[0075] If the initial feature distance at the current moment increases significantly, it indicates that the difference between the monitoring parameter at the current moment and the most similar monitoring parameter among multiple moments is also large. The probability that the monitoring parameter of the gluing machine at the current moment is abnormal is high, and the belt of the gluing machine is more likely to be damaged.

[0076] Conversely, if the initial feature distance at the current moment is small, it indicates that there are other monitoring parameters similar to the monitoring parameters at the current moment within multiple moments. The probability that the monitoring parameters of the gluing machine at the current moment are abnormal is lower, and the belt of the gluing machine is more likely to be in an intact state.

[0077] In this way, the monitoring parameters of the current moment and multiple historical moments before the current moment are mapped to a three-dimensional spatial coordinate system to obtain multiple data points, thereby determining the initial feature distance at the current moment. The initial feature distance can reflect the probability of the belt of the gluing machine being damaged at the current moment to a certain extent, so as to more accurately determine whether the belt of the gluing machine is damaged.

[0078] In step S104, the ratio of the anomaly level value at the current moment to the anomaly level value at the previous moment is used as an adjustment coefficient, and the initial feature distance is adjusted using the adjustment coefficient to obtain the target feature distance.

[0079] Since the abnormality level is determined based on the ambient temperature and instantaneous slip rate of the gluing machine, if the abnormality level at the current moment is greater than or equal to the abnormality level at the previous moment, it means that the abnormality level of the ambient temperature and instantaneous slip rate of the gluing machine at the current moment is greater than or equal to the abnormality level at the previous moment. If the ratio of the abnormality level at the current moment to the abnormality level at the previous moment is greater than or equal to 1, then an adjustment coefficient greater than or equal to 1 can be determined.

[0080] Conversely, if the abnormality value at the current moment is less than the abnormality value at the previous moment, it means that the abnormality of the ambient temperature and instantaneous slip rate of the gluing machine at the current moment is lower than that at the previous moment. If the ratio of the abnormality value at the current moment to the abnormality value at the previous moment is less than 1, then an adjustment coefficient less than 1 can be determined.

[0081] In one embodiment, adjusting the initial feature distance using an adjustment coefficient to obtain the target feature distance includes: using the adjustment coefficient as an exponent and the initial feature distance as the base for exponential operation, and using the result of the exponential operation as the target feature distance.

[0082] Since the initial feature distance is used to characterize the difference between the monitoring parameters at the current moment and the monitoring parameters with a higher degree of aggregation in multiple historical moments before the current moment; the larger the initial feature distance corresponding to the current moment, the more isolated the monitoring parameters at the current moment are compared with the monitoring parameters at other moments, and the higher the probability that the belt of the box gluing machine will be damaged at the current moment.

[0083] The adjustment coefficient is equal to the ratio of the abnormality value at the current moment to the abnormality value at the previous moment. The abnormality value is determined based on the degree of change of the monitoring parameter at the current moment relative to the previous moment, and the correlation between the instantaneous slip rate and frequency of the gluing machine at the current moment.

[0084] In addition to monitoring the instantaneous slip rate between the main drive wheel and the driven wheel of the gluing machine, the frequency of the carton passing through the detection point, and the working current of the servo motor of the main drive wheel, the degree of negative correlation between the instantaneous slip rate and the working current and the frequency is also considered. Therefore, the abnormality value can more accurately reflect the probability or degree of belt damage of the gluing machine, so that the adjustment coefficient can better reflect whether the belt of the gluing machine is damaged.

[0085] The adjustment coefficient is used as the exponential term to perform an exponential operation on the initial feature distance. If the ratio of the abnormality value at the current moment to the abnormality value at the previous moment is greater than or equal to 1, then a target feature distance greater than or equal to the initial feature distance can be determined based on the initial feature distance. A larger target feature distance can make the data point at the current moment more likely to be identified as an isolated data point, which is convenient for discovering shallower damage in the belt of the gluing machine, or for detecting belt damage in the gluing machine at an early stage.

[0086] If the ratio of the anomaly level value at the current moment to the anomaly level value at the previous moment is less than 1, then a target feature distance smaller than the initial feature distance can be determined based on the initial feature distance. A smaller target feature distance makes it more likely that the data point corresponding to the current moment will be assigned to a normal cluster, avoiding the identification of belts that are not actually damaged as damaged belts, and improving the accuracy of identifying whether the belts of the gluing machine are damaged.

[0087] In this way, by using the adjustment coefficient as an exponential term to perform an exponential operation on the initial feature distance, the obtained target feature distance can be used to more accurately determine whether the belt of the gluing machine is damaged.

[0088] In step S105, the data points at the current time in the clustering results are re-assigned using the target feature distance, and the belt of the gluing machine is determined to be damaged based on the assignment results.

[0089] In the density peak clustering algorithm, if the distance between a data point and all other nearby data points is greater than a preset distance threshold, the data point can be considered an abnormal data point; if the distance between a data point and all other nearby data points is less than or equal to the preset distance threshold, the data point can be considered a normal data point.

[0090] When directly using the initial feature distance as the data point for the current moment to assign clusters, the initial feature distance is difficult to accurately reflect the probability of the belt of the gluing machine being damaged at the current moment. It may misidentify the belt that is actually damaged as an undamaged belt, or vice versa.

[0091] After adjusting the initial feature distance, the target feature distance is obtained. The target feature distance can more accurately reflect whether the belt of the gluing machine is damaged at the current moment. When the data points at the current moment in the clustering results are re-assigned using the target feature distance, a more accurate assignment result can be obtained. Therefore, based on the assignment result, it is possible to more accurately determine whether the belt of the gluing machine is damaged.

[0092] In one embodiment, the data points at the current moment in the clustering results are reassigned to clusters using the target feature distance, including: if the target feature distance is less than or equal to a preset distance threshold, the data points at the current moment are assigned to the nearest cluster; if the target feature distance is greater than the preset distance threshold, the data points at the current moment are treated as abnormal data points.

[0093] The preset distance threshold can be determined based on the average distance between the data points participating in the clustering. For example, the preset distance threshold can be equal to the average distance between the data points participating in the clustering, or it can be another distance specified by the user in advance based on historical clustering data.

[0094] If the target feature distance is less than or equal to the preset distance threshold, it indicates that after adjusting the initial feature distance, there is a certain degree of similarity between the data point at the current moment and the data point at the historical moment participating in the clustering. The probability or degree of damage to the belt of the gluing machine at the current moment is low, and the data point at the current moment can be assigned to the nearest cluster.

[0095] If the target feature distance is greater than the preset distance threshold, it indicates that after adjusting the initial feature distance, the similarity between the data point at the current moment and the data point at the historical moment participating in the clustering is low, the data point at the current moment has high isolation, and the performance of the monitoring data of the gluing machine at the current moment is more consistent with the characteristics of abnormal data points. Therefore, it indicates that the probability or degree of damage to the belt of the gluing machine at the current moment is high, and the data point at the current moment can be regarded as an abnormal data point.

[0096] In this way, by using the target feature distance obtained after adjusting the initial feature distance, it is possible to determine whether the data point at the current moment should be assigned to the nearest cluster, so as to determine whether the belt of the gluing machine is damaged at the current moment.

[0097] In one embodiment, determining whether the belt between the main drive wheel and the driven wheel is damaged based on the allocation result includes: if the allocation result indicates that the data point at the current moment has been assigned to a cluster, determining that the belt between the main drive wheel and the driven wheel is not damaged; or if the allocation result indicates that the data point at the current moment has not been assigned to a cluster, determining that the belt between the main drive wheel and the driven wheel is damaged.

[0098] The allocation results can include being assigned to a cluster and not being assigned to a cluster. If the allocation result indicates that the data point at the current moment is assigned to a cluster, it means that the belt between the main drive wheel and the driven wheel is not damaged, and the conveying capacity of the carton gluing machine belt to the carton is not affected. It can be determined that the belt between the main drive wheel and the driven wheel is not damaged, so that the belt can continue to be monitored for damage.

[0099] If the data point at the current moment is not assigned to a cluster, it indicates that the belt of the gluing machine is damaged, and the conveying capacity of the gluing machine belt to the paper box is affected. It can be determined that the belt between the main drive wheel and the driven wheel is damaged, so that the user can deal with the damaged belt.

[0100] In one embodiment, if it is determined that the belt of the gluing machine is damaged, a notification message can be sent to the terminal device of the bound user, the notification message being used to indicate that the belt of the gluing machine is damaged.

[0101] The bound user can be a person, robot or other device that monitors the condition of the gluing machine. The system can send alert messages to the terminal device of the bound user so that the user can know in time about the damage to the belt of the gluing machine.

[0102] To facilitate users in handling damaged belts, the location information of the damaged gluing machine can also be sent to the terminal device of the bound user.

[0103] In response to receiving a prompt message, the terminal device can output a prompt message to indicate that the belt of the gluing machine is damaged. The prompt message can be at least one of the following: audio prompt, vibration prompt, and light prompt.

[0104] Figure 2 This is a schematic diagram illustrating the structure of a maintenance data analysis system 1000 for a box-gluing machine according to an exemplary embodiment. (Refer to...) Figure 2 The operation and maintenance data analysis system 1000 of the box gluing machine includes a processor 1100 and a memory 1200. The memory 1200 stores computer program instructions. When the computer program instructions are executed by the processor 1100, they implement all or part of the steps of the operation and maintenance data analysis method of the box gluing machine in this application.

[0105] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and embodiments are to be considered exemplary only.

[0106] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope.

Claims

1. A method for analyzing operation and maintenance data of a box-gluing machine, characterized in that, include: Obtain the degree of change in the monitoring parameters of the box gluing machine at the current moment relative to the previous moment; The sub-monitoring parameters include the instantaneous slip rate between the main drive wheel and the driven wheel of the gluing machine, the frequency at which the carton passes through the detection point, and the operating current of the servo motor of the main drive wheel; Determining the correlation value at the current moment includes: determining the first relative deviation between the instantaneous slip rate at the current moment and the historical average instantaneous slip rate, the second relative deviation between the operating current at the current moment and the historical average operating current, and the third relative deviation between the frequency at the current moment and the historical average frequency; determining the first difference between the first and third relative deviations, the second difference between the second and third relative deviations, and the average difference between the first and second differences; determining the similarity value of the monitoring parameters of the gluing machine at multiple historical moments before the current moment, including: mapping the monitoring parameters of the gluing machine at multiple historical moments before the current moment to a three-dimensional spatial coordinate system to obtain multiple data points, wherein the three-dimensional spatial coordinate system has the average instantaneous slip rate, average operating current, and average frequency as different dimensions, determining the distance between two data points in a data point pair, and using the normalized result of the reciprocal of the average distance of different data point pairs as the similarity value, wherein the data point pair includes two different data points among the multiple data points in the three-dimensional spatial coordinate system; using the average difference as the base and the similarity value as the exponent to obtain the exponential calculation result as the correlation value; The degree of anomaly at the current moment is determined based on the correlation value and the degree of abrupt change value; the correlation value is used to characterize the degree of negative correlation between the instantaneous slip rate and the operating current and the frequency. Cluster the monitoring parameters of the box gluing machine at the current time and historical time to obtain the initial feature distance at the current time; the initial feature distance is equal to the minimum value among the distances from the monitoring parameters at the current time to multiple density peak data points after clustering; The ratio of the anomaly level value at the current moment to the anomaly level value at the previous moment is used as an adjustment coefficient; the initial feature distance is adjusted using the adjustment coefficient to obtain the target feature distance; The data points at the current moment in the clustering results are re-assigned using the target feature distance, and the belt between the main drive wheel and the driven wheel is determined based on the assignment results.

2. The method for analyzing operation and maintenance data of a box-gluing machine according to claim 1, characterized in that, The mutation level value is determined in the following way: For the target sub-monitoring parameter among multiple sub-monitoring parameters, the change in the value of the target sub-monitoring parameter at the current time compared to its value at the previous time is determined, and the ratio of the change to the maximum value of the target sub-monitoring parameter at historical time is taken as the relative rate of change; the sum of the relative rates of change of all sub-monitoring parameters is taken as the degree of mutation value.

3. The method for analyzing operation and maintenance data of a box-gluing machine according to claim 1, characterized in that, Clustering of monitoring parameters of the gluing machine at the current and historical times yields the initial feature distance at the current time, including: The monitoring parameters of the box gluing machine at the current moment and at multiple historical moments before the current moment are mapped to a three-dimensional spatial coordinate system to obtain multiple data points; the three-dimensional spatial coordinate system has the average instantaneous slip rate, average operating current and average frequency as different dimensions; Clustering is performed on multiple data points in a three-dimensional spatial coordinate system, and data points whose local density values ​​are greater than a preset density threshold are taken as density peak data points in the clustering results. The minimum distance among the multiple density peak data points corresponding to the current data point is used as the initial feature distance for the current time.

4. The method for analyzing operation and maintenance data of a box-gluing machine according to claim 1, characterized in that, The initial feature distance is adjusted using adjustment coefficients to obtain the target feature distance, including: The adjustment coefficient is used as the exponent, and the initial feature distance is used as the base for the exponential operation. The result of the exponential operation is used as the target feature distance.

5. The method for analyzing operation and maintenance data of a box-gluing machine according to claim 1, characterized in that, The data points at the current time step in the clustering results are re-assigned to clusters using the target feature distance, including: If the distance to the target feature is less than or equal to a preset distance threshold, the data point at the current moment will be assigned to the nearest cluster. If the distance to the target feature is greater than a preset distance threshold, the data point at the current moment will be regarded as an abnormal data point.

6. The method for analyzing operation and maintenance data of a box-gluing machine according to claim 1, characterized in that, Based on the distribution results, determine whether the belt between the main drive pulley and the driven pulley is damaged, including: If the allocation result indicates that the data point at the current moment has been assigned to a cluster, it is determined that the belt between the main drive wheel and the driven wheel is not damaged; or, if the allocation result indicates that the data point at the current moment has not been assigned to a cluster, it is determined that the belt between the main drive wheel and the driven wheel is damaged.

7. The method for analyzing operation and maintenance data of a box-gluing machine according to claim 6, characterized in that, The method further includes: If it is determined that the belt between the main drive wheel and the driven wheel is damaged, a prompt message is sent to the terminal device of the user to which the box gluing machine is bound; the prompt message is used to indicate that the belt between the main drive wheel and the driven wheel is damaged.

8. A maintenance data analysis system for a box-gluing machine, characterized in that, include: A processor and a memory, wherein the memory stores computer program instructions, which, when executed by the processor, implement the operation and maintenance data analysis method for a box-gluing machine according to any one of claims 1-7.

Images