Real-time monitoring method and system for running state of stamping equipment

Abstract

The invention relates to the field of stamping equipment, in particular to a stamping equipment operation state real-time monitoring method and system, and the method comprises the steps: obtaining operation data of stamping equipment, taking any operation data as central data, presetting a distance, calculating the distance between the central data and other data, determining the aggregation degree of each operation data in the preset distance, and determining the aggregation degree of each operation data in the preset distance; forming an aggregation degree sequence under different preset distances; performing clustering analysis on the aggregation degree sequence to obtain a plurality of clusters, and determining an optimal clustering result through an evaluation function; and operation data, collected in real time, of the stamping equipment are obtained, a corresponding clustering cluster, belonging to the optimal clustering result, of the real-time operation data is calculated, and the operation state of the stamping equipment is judged according to the clustering cluster which the real-time operation data belong to. According to the invention, by analyzing the abnormal mode in the operation data and combining the clustering comparison of the real-time data, the operation state of the stamping equipment is accurately identified, the targeted maintenance is realized, and the service life of the stamping equipment is prolonged.

Description

Technical Field

[0001] The present invention relates to the field of stamping equipment. In particular, it relates to a method and system for real-time monitoring of the operating state of stamping equipment. Background Art

[0002] A stamping equipment is a mechanical device that uses a die and stamping process to process metal or non-metal sheets into parts with specific shapes and sizes by pressure. It is applied in many fields such as automotive, aerospace, electronics, household appliances, machinery, etc. With the development of society, stamping equipment will develop towards the direction of intelligence, high precision, energy conservation and environmental protection. The stamping process includes blanking, bending, stretching, forming, etc., and can efficiently realize the processing of parts with complex shapes.

[0003] The existing Chinese patent application document with the publication number CN117150333A discloses an operation and maintenance supervision method and system for a stamping and forming equipment, which relates to the field of data processing technology. The method includes: calling equipment operation data and establishing a mapping association between the called data and product data; compensating the stamping quality with input samples and performing a mapping between the operation data of the forming equipment and an abnormal identifier; clustering the mapping results to generate N data clustering clusters; generating an initial feature constraint, obtaining compensation data through linkage abnormal compensation, establishing an operation and maintenance supervision network, analyzing the real-time monitoring data, and generating an operation and maintenance supervision result.

[0004] This application document compensates the equipment operation data through the quality characteristics of raw materials to improve the accuracy of the operation analysis of stamping equipment. Currently, the fault modes of stamping equipment are complex, including mechanical faults, electrical faults, software faults, etc. Different fault types have different impacts on the performance of stamping equipment, and different faults may exhibit similar abnormal characteristics, so it is difficult to accurately distinguish and diagnose. Summary of the Invention

[0005] To solve the problem that different faults of stamping equipment may show similarity in data and it is difficult to distinguish the operating states of different faults, the present invention provides solutions in the following aspects.

[0006] In a first aspect, a method for real-time monitoring of the operating state of a stamping device includes: obtaining the operating data and data tags of a historical stamping device, where the operating data includes: temperature, current, feeding speed, stamping speed, stamping pressure, and motor load, and the data tags are mechanical faults, electrical faults, and software faults; taking any operating data as the central data and presetting a distance, calculating the distance between the central data and other data, determining whether the other data falls within the preset distance, calculating the aggregation degree of each operating data within the preset distance, and obtaining the aggregation degree sequence of each operating data under different preset distances; calculating the similarity between the aggregation degree sequences of each operating data, clustering the aggregation degree sequences of the operating data based on the similarity to obtain multiple clustering clusters, and calculating the evaluation function of each clustering cluster to obtain the best clustering result; obtaining the operating data of the stamping device collected in real time, calculating that the real-time operating data belongs to the corresponding clustering cluster in the best clustering result, and judging the operating state of the stamping device according to the belonging clustering cluster.

[0007] The effects are as follows: By obtaining the multi-dimensional operating data of the stamping device and combining the clustering analysis of the aggregation degree sequence, the accurate monitoring and fault diagnosis of the device operating state are realized; by calculating the weighted distance and aggregation degree sequence between data points, the operating data is classified into different clustering clusters, and the best clustering result is selected through the evaluation function, which can not only effectively distinguish complex fault modes such as mechanical faults, electrical faults, and software faults, but also judge the operating state of the device in real time, reduce false alarms and missed alarms. In addition, the real-time monitoring function of the present invention can detect potential faults in advance, reduce sudden downtime, optimize production efficiency, and facilitate the timely maintenance and management of the stamping device.

[0008] Preferably, the method further includes preprocessing the obtained operating data, where the preprocessing includes: aligning the operating data according to the acquisition timestamp, and sequentially performing smoothing processing using the exponential smoothing method and noise reduction processing using the filtering algorithm, and normalizing each data point of the operating data.

[0009] The effects are as follows: Through the preprocessing methods of timestamp alignment, exponential smoothing method smoothing processing, filtering noise reduction, and normalization processing, the quality and consistency of the operating data are effectively improved, an accurate time reference and a consistent data format are provided, a reliable data basis for operating monitoring is provided, and thus the accuracy and reliability of fault diagnosis and operating state monitoring are significantly enhanced.

[0010] Preferably, the calculating the distance between the central data and other data includes: According to the operation data collected at each moment, obtain the probability of the operation data appearing in each collection as the core degree of the corresponding operation data, calculate the Euclidean distance between the central data and other data, take the sum of the core degree of the central data and the core degree of other data as the adjustment factor, and obtain the distance between the central data and other data according to the product of the adjustment factor and the Euclidean distance.

[0011] Its effect is that by considering the probability of the operation data appearing in each collection, that is, the core degree, the importance of data points can be distinguished. Data points with a high core degree are more critical in the operation of the stamping equipment, and thus are given greater weights in the distance calculation. By introducing the core degree as an adjustment factor, the distance measurement becomes more accurate and can better reflect the significance of data points in the actual operation state.

[0012] Preferably, the aggregation degree sequence of each operation data includes: Taking any preset distance as the target distance, calculate the number of other operation data around the operation data at the target distance, and take the ratio of the number of other operation data to the total number of operation data as the aggregation degree of the operation data at the corresponding target distance, so as to obtain the aggregation degree sequence at each target distance.

[0013] Its effect is that by calculating the number of other data points around the operation data at the preset target distance, the local density of each data point at this distance can be quantified. The aggregation degree sequence can reveal the distribution characteristics of data points at different distance scales and help identify dense areas and sparse areas in the data set.

[0014] Preferably, the aggregation degree sequence of each operation data further includes: Obtain the maximum value and the minimum value of the Euclidean distance between the central data and other data, calculate the ratio between the maximum value and the minimum value as the aggregation degree of the central data, and obtain the aggregation degree sequence at different preset distances according to different preset distances.

[0015] Its effect is that by calculating the ratio of the maximum value to the minimum value of the Euclidean distance between the central data and other data as the aggregation degree and generating the aggregation degree sequence according to different preset distances, the dispersion degree of data points in the multi-dimensional space can be quantified, and the dense or sparse degree of the data distribution can be intuitively reflected. It not only enhances the accuracy of anomaly detection by capturing the high dispersion of anomaly data through multi-dimensional analysis, but also optimizes the effect of clustering analysis and improves the accuracy and robustness of the clustering results.

[0016] Preferably, the calculation of the similarity between the aggregation degree sequences of each operation data includes: Taking the aggregation degree sequence of the operation data collected at any time as the annotation sequence, calculating the covariance between the annotation sequence and other aggregation degree sequences, calculating the product of the standard deviation of the annotation sequence and the standard deviations of other aggregation degree sequences, and taking the ratio of the covariance to the product of the standard deviations as the similarity between the aggregation degree sequences.

[0017] Preferably, calculating the similarity between the aggregation degree sequences of each operation data further includes: Taking the aggregation degree sequence of the operation data collected at any time as the annotation sequence, and calculating the distance between the annotation sequence and other aggregation degree sequences, and using the distance to calculate the similarity between the annotation sequence and other aggregation degree sequences by using the negative exponential function.

[0018] Preferably, the evaluation function satisfies the following relational expression: ; In the formula, represents the evaluation function of the clustering cluster, represents the entropy of all clustering clusters as a whole, represents the total number of clustering clusters, represents the th entropy of the clustering cluster.

[0019] Preferably, judging the operation state of the stamping equipment according to the clustering cluster includes: Calculating the distance between the real-time collected data and the clustering centers of each clustering cluster to obtain the matching clustering cluster, and taking the label corresponding to the mode of the data in the matching clustering cluster as the operation state of the real-time collected data.

[0020] The effect is that: according to the clustering result of the real-time data, the maintenance plan can be formulated more accurately, unnecessary inspections can be avoided, and by accurately judging the operation state of the stamping equipment, the downtime caused by misjudgment can be reduced, and the efficiency of the production line can be improved.

[0021] In a second aspect, a real-time monitoring system for the operation state of a stamping equipment includes: a processor and a memory, and the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the above-mentioned real-time monitoring method for the operation state of the stamping equipment is implemented.

[0022] The present invention has the following effects: 1. By acquiring multi-dimensional operation data and using the aggregation degree sequence and clustering analysis, the present invention can more accurately distinguish and diagnose different types of faults such as mechanical faults, electrical faults, and software faults. At the same time, according to the fault conditions of the historical data corresponding to each clustering cluster, a quick response can be made, and an accurate judgment on the operation state of the stamping equipment can be obtained by combining past experience.

[0023] 2. The present invention calculates the distance between the real-time operation data and the cluster center in the optimal clustering result, quickly judges the current operation state of the stamping equipment, which is beneficial to timely discover potential abnormalities, provides data support for preventive maintenance, reduces unexpected downtime, and improves production efficiency. At the same time, it helps to optimize the maintenance plan of the stamping equipment, gives priority to stamping equipment with unstable operation state or frequent abnormalities, and then guides resource allocation to ensure that key equipment is properly maintained and monitored, thereby extending the service life of the stamping equipment and reducing operating costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 is a flowchart of the method of steps S1 - S4 in a method for real-time monitoring of the operation state of a stamping equipment according to an embodiment of the present invention.

[0025] Figure 2 is a structural block diagram of a system for real-time monitoring of the operation state of a stamping equipment according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0026] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments.

[0027] The following will describe in detail the specific embodiments of the present invention in conjunction with the accompanying drawings.

[0028] Referring to Figure 1 , a method for real-time monitoring of the operation state of a stamping equipment includes steps S1 - S4, specifically as follows: S1: Obtain the operation data and data labels of the historical stamping equipment, where the operation data includes: temperature, current, feeding speed, stamping speed, stamping pressure, and motor load, and the data labels are mechanical faults, electrical faults, and software faults.

[0029] The method further includes preprocessing the obtained operation data, where the preprocessing includes: aligning the operation data according to the acquisition timestamp, and sequentially performing smoothing processing using the exponential smoothing method and noise reduction processing using the filtering algorithm, and performing normalization processing on each data point of the operation data.

[0030] It should be noted that the operation data is a multi-dimensional spatial vector. By installing sensors on the stamping equipment, such as temperature sensors, current sensors, speed sensors, and load sensors, etc., multi-dimensional data during the operation of the equipment can be collected in real time. The operation data includes but is not limited to parameters such as the collected temperature, current, feed speed, and motor load. To ensure data quality, the operation data of the stamping equipment collected is filtered and denoised to remove high-frequency interference, outliers are removed by statistical methods to avoid misjudgment, and normalization processing is performed to eliminate the differences in dimensions and orders of magnitude of different parameters.

[0031] The data tags respectively correspond to data features: Mechanical failures are usually related to the components of the equipment, such as: dies, sliders, guide rails, etc. Mechanical failures may manifest as: abnormal temperature, unstable feed speed, and unstable motor load, etc.; Electrical failures are usually related to the electrical system of the equipment, such as: circuits, motors, and controllers, etc. Electrical failures may manifest as: abnormal current, abnormal voltage, frequency change, and electromagnetic interference, etc.; Software failures are usually related to the control system software of the equipment, such as: programming errors, compatibility issues, etc. Software failures may manifest as: the stamping speed and pressure data not matching the actual operation data.

[0032] During actual operation, the normal operation state of the stamping equipment usually corresponds to the dense area of the data, while the abnormal state may be manifested as isolated or sparse distributed data points. By analyzing the density or sparsity of the data points around each data point, it can be judged whether the data point is an abnormal point, so as to effectively distinguish the normal operation state and the potential failure state, providing a basis for fault diagnosis. The specific implementation steps are as follows: S2: Taking any operation data as the central data and presetting a distance, calculate the distance between the central data and other data, judge whether other data fall within the preset distance, calculate the aggregation degree of each operation data within the preset distance, and obtain the aggregation degree sequence of each operation data under different preset distances.

[0033] Among them, the Euclidean distance between the central data and other data is used as the calculation method of the distance. Specifically, the distance satisfies the following relational expression: ; In the formula, represents the distance between the operation data collected for the th time and the operation data collected for the th time, represents the spatial vector of the operation data collected for the th time, represents the spatial vector of the operation data collected for the th time, The norm represents the Euclidean distance between the operation data collected twice.

[0034] Furthermore, upon further analysis, there are duplicate data in the operation data. Duplication indicates that the data appears more frequently during the operation of the stamping equipment. The operation state corresponding to the data with a higher occurrence frequency is more likely to be normal because abnormal data has greater randomness and a lower likelihood of being the same. Therefore, the more times the data appears repeatedly, the greater the likelihood that the point is a core point. Thus, when calculating the distance between different operation data, the core degree of the data can be introduced into the calculation.

[0035] In addition, in another embodiment, it further includes: Based on the operation data collected at each moment, obtain the probability of occurrence of the operation data collected each time as the core degree of the corresponding operation data, calculate the Euclidean distance between the central data and other data, use the sum of the core degree of the central data and the core degree of other data as a modulation factor, and obtain the distance between the central data and other data according to the product of the modulation factor and the Euclidean distance.

[0036] Specifically, the distance satisfies the following relational expression: ; In the formula, represents the th time of collecting operation data and the th time of collecting operation data, represents the core degree of the th time of collecting operation data, represents the core degree of the th time of collecting operation data, represents the spatial vector of the th time of collecting operation data, represents the spatial vector of the th time of collecting operation data, represents The norm represents the Euclidean distance between the operation data collected twice.

[0037] That is to say, the greater the core degree of the collected operation data, the larger the influence range of the data. Then, being far from the operation data with a large core degree indicates a greater probability that it is an abnormal point. In fault diagnosis, the weighted distance can help more accurately locate the cause of the fault because the data points with a high core degree are more likely to be associated with the normal operation state.

[0038] Furthermore, the preset distance is a key parameter for defining the neighborhood range. By setting multiple different preset distances, the distribution characteristics of data points can be analyzed at multiple scales. The preset distance can be adjusted according to the distribution range of the data and the operating characteristics of the device.

[0039] The aggregation degree sequence of each operating data includes: Taking any preset distance as the target distance, calculate the number of other operating data around the operating data at the target distance, and take the ratio of the number of other operating data to the total number of operating data as the aggregation degree of the operating data at the corresponding target distance, so as to obtain the aggregation degree sequence at each target distance.

[0040] Furthermore, the aggregation degree sequence reflects the characteristics of a certain operating data among all operating data and how it is distributed among all data. For example, if a certain operating data is at the center point of all data, the obtained aggregation degree sequence should increase steadily, that is, it shows that the aggregation degree will not suddenly increase a lot or suddenly increase very little. If a certain operating data is at the edge point or an abnormal point, the growth of its obtained aggregation degree sequence is not stable, because as the distance increases, the increase of other data points within this distance range is sudden and unstable.

[0041] In addition, in another embodiment, it further includes: Obtain the maximum and minimum Euclidean distances between the central data and other data, calculate the ratio between the maximum and minimum values as the aggregation degree of the central data, and obtain the aggregation degree sequences at different preset distances according to different preset distances.

[0042] Specifically, the aggregation degree satisfies the following relational expression: ; In the formula, represents the aggregation degree of the operating data collected for the th time, represents the maximum value of the Euclidean distance between the spatial vector of the operating data collected for the th time and the spatial vector of the operating data collected for the th time, represents the minimum value of the Euclidean distance between the spatial vector of the operating data collected for the th time and the spatial vector of the operating data collected for the th time.

[0043] That is to say, within the same preset distance range, the larger the ratio between the maximum value of the Euclidean distance and the minimum value of the Euclidean distance, the more dispersed the data distribution in the preset distance range, and the smaller the aggregation degree. On the contrary, the smaller the ratio, the more concentrated the data distribution in the preset distance range, and the larger the aggregation degree.

[0044] It should be noted that by analyzing the similarity of the distribution characteristics of data points at different scales, data points with similar operating states are grouped into one category, so as to achieve accurate classification of the operating states of stamping equipment. Through the classification results between different data points, the operating states of the data points corresponding to the classification results can be determined, which is conducive to distinguishing the causes of faults. The specific steps are as follows: S3: Calculate the similarity between the aggregation degree sequences of each operating data, cluster the aggregation degree sequences of the operating data based on the similarity, obtain multiple clustering clusters, and calculate the evaluation function of each clustering cluster to obtain the best clustering result.

[0045] The similarity includes: Taking the aggregation degree sequence of the operating data collected at any time as the annotation sequence, calculate the covariance between the annotation sequence and other aggregation degree sequences, calculate the product of the standard deviation of the annotation sequence and the standard deviation of other aggregation degree sequences, and take the ratio of the covariance to the product of the standard deviations as the similarity between the aggregation degree sequences.

[0046] Specifically, the similarity satisfies the following relational expression: ; Wherein, represents the similarity between the aggregation degree sequence of the operating data collected for the th time and the aggregation degree sequence of the operating data collected for the th time, represents the covariance between the aggregation degree sequence of the operating data collected for the th time and the corresponding aggregation degree sequence of the operating data collected for the th time, represents the standard deviation of the aggregation degree sequence of the operating data collected for the th time, represents the standard deviation of the aggregation degree sequence of the operating data collected for the th time.

[0047] That is to say, by dividing the covariance by the product of the standard deviations of the two sequences, we obtain a dimensionless measure, which makes the value of the similarity within the range of to evaluate the similarity between data points, so as to help determine which data points should be assigned to the same clustering cluster.

[0048] In addition, in another embodiment, it also includes: Taking the aggregation degree sequence of the operating data collected at any time as the annotation sequence, calculate the distance between the annotation sequence and other aggregation degree sequences, and take the The similarity between the labeled sequence and other aggregation degree sequences is calculated using the negative exponential function.

[0049] Specifically, the similarity satisfies the following relational expression: ; In the formula, represents the similarity between the aggregation degree sequence of the operation data collected for the th time and the aggregation degree sequence of the operation data collected for the th time, represents the exponential function with the natural number as the base, represents the distance between the aggregation degree sequence of the operation data collected for the th time and the aggregation degree sequence of the operation data collected for the th time.

[0050] That is to say, in the similarity calculation, the negative exponential function can convert a non - negative DTW distance into a similarity value between . The greater the distance, the smaller the similarity; the smaller the distance, the closer the similarity is to 1.

[0051] Specifically, the evaluation function satisfies the following relational expression: ; In the formula, represents the evaluation function of the clustering cluster, represents the entropy of the th clustering cluster, represents the entropy of all clustering clusters as a whole.

[0052] That is to say, the greater the evaluation function, the better the clustering effect. represents the entropy of all clustering clusters as a whole. By calculating the entropy of all clustering clusters as a whole, the degree of chaos of this clustering cluster is reflected. The more consistent the running states of the operation data in the clustering cluster, the smaller the degree of chaos of this clustering.

[0053] Among them, the entropy of the clustering cluster satisfies the following relational expression: ; In the formula, represents the entropy of the th clustering cluster, represents the probability that the th data belongs to the th clustering cluster, represents the logarithmic function with as the base.

[0054] The entropy of all clustering clusters as a whole respectively satisfies the following relational expressions: ; In the formula, represents the entropy of all clustering clusters as a whole, represents the probability that the th data appears in all clustering clusters, represents the logarithmic function with as the base.

[0055] S4: Obtain the operation data of the stamping equipment collected in real time, calculate that the real-time operation data belongs to the corresponding clustering cluster in the optimal clustering result, and judge the operation state of the stamping equipment according to the clustering cluster to which it belongs.

[0056] Calculate the distances between the data collected in real time and the clustering centers of each clustering cluster to obtain the matching clustering clusters, and use the label corresponding to the mode of the data in the matching clustering clusters as the operation state of the data collected in real time.

[0057] It should be noted that by calculating the distances between the real-time collected data and the clustering centers of each clustering center, the position of the real-time data point in the multi-dimensional feature space can be determined, and then it can be judged which clustering cluster is the closest. The real-time data point is assigned to the clustering cluster with the closest distance, indicating that the characteristics of the real-time data point are most similar to the central characteristics of the clustering cluster. Using the mode operation state of the matching clustering cluster as the operation state of the real-time data point, based on the experience of historical regulation, there is more data basis for judging the abnormal situation of the operation state of the stamping equipment, and guiding adjustments are made.

[0058] The present invention also provides a real-time monitoring system for the operation state of a stamping equipment. As Figure 2 shown, the system includes a processor and a memory, and the memory stores computer program instructions. When the computer program instructions are executed by the processor, a real-time monitoring method for the operation state of a stamping equipment according to the first aspect of the present invention is implemented.

[0059] The system also includes a communication bus and a communication interface and other components well known to those skilled in the art. Their settings and functions are known in the art, so they will not be elaborated here.

[0060] It should be pointed out that for those of ordinary skill in the art, without departing from the concept of the present invention, several deformations and improvements can still be made, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the present invention patent shall be subject to the appended claims.

Claims

1. A method for real-time monitoring of the operating status of a stamping equipment, characterized in that: include: Acquire historical stamping equipment operation data and data tags, wherein the operation data includes: temperature, current, feed speed, stamping speed, stamping pressure and motor load, and the data tags are mechanical failure, electrical failure and software failure; Taking any running data as the central data and presetting the distance, calculating the distance between the central data and other data, judging whether other data falls within the preset distance, calculating the aggregation degree of each running data within the preset distance, and obtaining the aggregation degree sequence of each running data under different preset distances; Calculating the similarity between the clustering degree sequences of each running data, clustering the clustering degree sequences of the running data based on the similarity to obtain multiple clustering clusters, and calculating the evaluation function of each clustering cluster to obtain the best clustering result; Acquire the real-time operation data of the stamping equipment, calculate the clustering cluster corresponding to the best clustering result to which the real-time operation data belongs, and judge the operation status of the stamping equipment according to the clustering cluster to which it belongs.

2. A method for real-time monitoring of the operating status of a stamping equipment according to claim 1, characterized in that: The method also includes preprocessing the acquired operation data, wherein the preprocessing includes: aligning the operation data according to the acquisition timestamp, and using the exponential smoothing method to perform smoothing processing and the filtering algorithm to perform noise reduction processing in sequence, and normalizing each data point of the operation data.

3. A method for real-time monitoring of the operating status of a stamping equipment according to claim 1, characterized in that: The distance between the computing center data and other data includes: According to the operation data collected at each moment, the probability of occurrence of the operation data collected each time is obtained as the core degree of the corresponding operation data, the Euclidean distance between the central data and other data is calculated, and the sum of the core degree of the central data and the core degree of other data is used as an adjustment factor. According to the product of the adjustment factor and the Euclidean distance, the distance between the central data and other data is obtained.

4. A method for real-time monitoring of the operating status of a stamping equipment according to claim 1, characterized in that: The aggregation degree sequence of each operation data includes: Taking any preset distance as the target distance, the number of other running data around the running data at the target distance is calculated, and the ratio of the number of other running data to the total number of running data is taken as the aggregation degree of the running data at the corresponding target distance, and the aggregation degree sequence at each target distance is obtained.

5. A method for real-time monitoring of the operating status of a stamping equipment according to claim 1, characterized in that: The aggregation degree sequence of each operation data further includes: The maximum and minimum values ​​of the Euclidean distance between the central data and other data are obtained, and the ratio between the maximum and minimum values ​​is calculated as the aggregation degree of the central data. According to different preset distances, the aggregation degree sequence under different preset distances is obtained.

6. A method for real-time monitoring of the operating status of a stamping equipment according to claim 1, characterized in that: The calculating the similarity between the aggregation degree sequences of each running data includes: The aggregation sequence of any collected running data is used as the annotation sequence, the covariance between the annotation sequence and other aggregation sequences is calculated, the product between the standard deviation of the annotation sequence and the standard deviation of other aggregation sequences is calculated, and the ratio of the product between the covariance and the standard deviation is used as the similarity between the aggregation sequences.

7. A method for real-time monitoring of the operating status of a stamping equipment according to claim 1, characterized in that: The calculating of the similarity between the aggregation degree sequences of each running data further includes: Take the aggregation sequence of any collected running data as the annotation sequence and calculate the relationship between the annotation sequence and other aggregation sequences. Distance, will The distance is calculated using a negative exponential function to obtain the similarity between the labeled sequence and other clustering sequences.

8. A method for real-time monitoring of the operating status of a stamping equipment according to claim 1, characterized in that: The evaluation function satisfies the following relationship: ； In the formula, represents the evaluation function of the clustering cluster, represents the overall entropy of all clusters, Represents the total number of clusters. Indicates The entropy of the clusters.

9. A method for real-time monitoring of the operating status of a stamping equipment according to claim 1, characterized in that: The step of judging the operating state of the stamping equipment according to the cluster to which it belongs includes: The distance between the real-time collected data and the cluster center of each cluster cluster is calculated to obtain the matching cluster cluster, and the label corresponding to the majority of the data in the matching cluster cluster is used as the running status of the real-time collected data.

10. A real-time monitoring system for the operation status of a stamping equipment, 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, a method for real-time monitoring of the operating status of a stamping equipment according to any one of claims 1 to 9 is implemented.

Images