TW202307668A - System for detecting abnormalities and method for detecting abnormalities, electronic device, and storage medium - Google Patents

Abstract

The present application provides a system for detecting abnormalities. The system includes a sensor, a data acquisition card, and an electronic device. The present application also provides a method for detecting abnormalities, an electronic device, and a storage medium. The method includes: acquiring data of a component during a normal operation, and extracting a plurality of principal components of the data using a principal component analysis algorithm; calculating a standard value based on the plurality of principal components; generating an envelope curve based on the standard value; obtaining data to be tested when the component is currently running, and detecting whether the component is currently running abnormally according to the data to be tested and the envelope curve. By utilizing the present application, an abnormality detection efficiency and a detection accuracy of an equipment can be improved.

Description

Abnormality detection system and abnormality detection method, electronic device and storage medium

The present application relates to the field of information analysis, and in particular to an anomaly detection system and anomaly detection method, electronic equipment and storage media.

In the forging process of production, abnormalities occasionally occur, resulting in unqualified forged products and increasing production costs. At the same time, abnormalities in the forging process are highly concealed, and it is easy for relevant personnel to fail to discover abnormalities in time.

In view of the above, it is necessary to provide an anomaly detection method, an electronic device and a storage medium, which can improve the abnormality detection efficiency and detection accuracy of the device.

The present application provides an anomaly detection method, the method comprising: obtaining information on normal operation of components, extracting multiple principal components of the information through a principal component analysis algorithm; calculating standard values according to the multiple principal components; The standard value generates an envelope curve; obtains information to be tested when the component is currently running, and detects whether the component is running abnormally according to the information to be tested and the envelope curve.

In a possible implementation manner, the acquiring the information of the components during normal operation, and extracting multiple principal components of the information through the principal component analysis algorithm includes: acquiring the first sensor during the normal operation of the forging press through the first sensor One information, and the second information obtained by the second sensor when the forging die is in normal operation; the multiple first principal components of the first information are extracted through the principal component analysis algorithm, and all the first principal components are extracted through the principal component analysis algorithm A plurality of second principal components of the second information.

In a possible implementation manner, the calculating the standard value according to the multiple principal components includes: calculating the first standard value according to the multiple first principal components; calculating the first standard value according to the multiple second principal components. Two standard values.

In a possible implementation manner, the generating the envelope curve according to the standard value includes: determining a first upper envelope curve according to the first standard value and a preset first upper deviation value; Determine the first lower envelope curve with the preset first lower deviation value; determine the first upper envelope curve and the first lower envelope curve as the first envelope curve; according to the second standard value and the preset The set second upper deviation value determines the second upper envelope curve; determines the second lower envelope curve according to the second standard value and the preset second lower deviation value; the second upper envelope curve and the second A lower envelope curve is determined as the second envelope curve.

In a possible implementation manner, the acquiring the information to be tested when the component is currently running, and detecting whether the component is running abnormally according to the information to be tested and the envelope curve includes: through the first sensor obtaining the first transient information of the forging press by a detector, and obtaining the second transient information of the forging die through the second sensor; obtaining a first statistical value based on the first transient information, and Obtaining a second statistical value based on the second transient information; determining the first detection result according to the first statistical value and the first envelope curve, and determining the first detection result according to the second statistical value and the second envelope The curve determines the second detection result; and determines whether there is an abnormality in the forging process according to the first detection result and the second detection result.

In a possible implementation manner, the determining the first detection result according to the first statistical value and the first envelope curve includes: judging whether the first statistical value is included in the first envelope curve ; When the first statistical value is included in the first envelope curve, determine that the first detection result is normal; when the first statistical value is not included in the first envelope curve, determine that the The above-mentioned first detection result is abnormal.

In a possible implementation manner, the determining whether there is an abnormality in the forging process according to the first detection result and the second detection result includes: when each of the first detection result and the second detection result detects When the results are all normal, it is determined that the forging process is normal; when at least one of the first detection result and the second detection result is abnormal, it is determined that the forging process is abnormal.

The present application also provides an abnormality detection system, the abnormality detection system includes: a sensor, used to collect information of components; an information acquisition card, communicated with the sensor, used to receive the information; electronic equipment , communicated with the information acquisition card, and is used to analyze the information to obtain an abnormality detection result; determine whether the components are abnormal according to the abnormality detection result.

The present application also provides an electronic device, the electronic device includes a processor and a memory, and the processor is configured to implement the abnormality detection method when executing a computer program stored in the memory.

The present application also provides a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the abnormality detection method is implemented.

The anomaly detection method and related equipment disclosed in the present application use a method combining the forging envelope curve and the PCA algorithm to detect whether there is an anomaly in the forging process, which improves the accuracy of anomaly detection.

In order to make the purpose, technical solution and advantages of the present application clearer, the present application will be described in detail below in conjunction with the accompanying drawings and specific embodiments.

FIG. 1 is an anomaly detection system provided by an embodiment of the present application. As shown in FIG. 1 , the anomaly detection system 1 includes: a sensor 10 , an information acquisition card 11 , and an electronic device 12 . Wherein, the sensor 10 is used to collect information of components, and is installed on the surface of the components, and the information acquisition card 11 is connected to the sensor for receiving the information, and the electronic The device 12 is communicatively connected with the information acquisition card, and is used for performing information analysis on the information to obtain abnormality detection results.

In one embodiment of the present application, the components include a forging press 13 and a forging die 14 . The sensor 10 includes at least one first sensor 101 and at least one second sensor 102, wherein the first sensor 101 is used to collect information of the forging press 13, and the second The sensor 102 is used for collecting information of the forging die 14 .

In one embodiment of the present application, the ultrasonic sensor is fixed on the surface of the forging die by pasting and mounting, without destroying the main body structure of the forging die, in view of the characteristics of high hardness of the forging die and difficulty in grooving.

In one embodiment of the present application, the information acquisition card 11 is equipped with a laser trigger device 111 for dividing the information.

In one embodiment of the present application, the electronic device 12 displays an information analysis interface, and the information analysis interface displays information analysis results for displaying the abnormality detection results. By displaying the abnormal detection results, it is convenient for relevant personnel to find problems in time.

By installing the sensor, the forging process can be monitored in real time, and information can be obtained through the sensor to warn the health status of the forging process.

Please refer to FIG. 2 , which is a schematic diagram of an electronic device 12 according to an embodiment of the present application. As shown in FIG. 2 , the electronic device 12 includes, but is not limited to, a memory 121 and at least one processor 122 . The above components may be connected through a bus bar or directly.

The electronic device 12 may be a computer, a mobile phone, a tablet computer, a personal digital assistant (Personal Digital Assistant, PDA) and other devices installed with application programs. Those skilled in the art can understand that the schematic diagram 2 is only an example of the electronic device 12, and does not constitute a limitation to the electronic device 12. For example, the electronic device 12 may also include input and output devices, network access devices, bus bars, and the like.

As shown in FIG. 3 , it is a flow chart of a preferred embodiment of the anomaly detection method of the present application. The abnormality detection method is applied in the electronic device 12 . According to different requirements, the order of the steps in the flowchart can be changed, and some steps can be omitted. In this embodiment, the abnormality detection method includes:

S11. Acquiring information on normal operation of components, and extracting multiple principal components of the information through a principal component analysis algorithm.

In one embodiment of the present application, the acquisition of the information of the normal operation of the components, and the extraction of multiple principal components of the information through the principal component analysis algorithm include: obtaining the information of the normal operation of the forging press through the first sensor The first information, and the second information obtained by the second sensor when the forging die is in normal operation; the plurality of first principal components of the first information are extracted through the principal component analysis algorithm, and extracted through the principal component analysis algorithm A plurality of second principal components of the second information. Principal components analysis (PCA) is a statistical method that converts multiple variables into a few principal components through dimensionality reduction techniques. It can reduce the dimensionality of high-dimensional information to reduce the number of predictive variables, and remove noise through dimensionality reduction.

During specific implementation,

(1) Standardize the first information to obtain first standardized information. Using the PCA algorithm to reduce the dimensionality of the first standardized information to obtain several first principal components, specifically, calculate the first covariance matrix of the first standardized information, and then calculate the first covariance The first eigenvalue and the first eigenvector of the matrix, the contribution rate of each of the first eigenvalues is obtained, and the first principal component whose contribution rate is greater than the preset first threshold is used as the plurality of first principal components Element.

(2) Standardize the second information to obtain second standardized information. Using the PCA algorithm to reduce the dimension of the second normalized information to obtain several second principal components, specifically, calculate the second covariance matrix of the second normalized information, and then calculate the second covariance The second eigenvalue and the second eigenvector of the matrix, obtaining the contribution rate of each of the second eigenvalues, and using the second principal component whose contribution rate is greater than the preset second threshold as the plurality of second principal components Element.

By using the principal component analysis algorithm to reduce a large amount of information into a small amount of information, the complexity of the information is reduced, and the efficiency and accuracy of information analysis can be improved.

S12. Calculate a standard value according to the plurality of principal components.

In an embodiment of the present application, the calculating the standard value according to the plurality of principal components includes: calculating the first standard value according to the plurality of first principal components; calculating and obtaining the standard value according to the plurality of second principal components second standard value.

During specific implementation, the first standard value calculated according to the plurality of first principal components includes:

Input the multiple first principal components into the pre-trained PCA model, and output the first standard value , where is the significance level, and is a distribution that obeys the first degree of freedom and the second degree of freedom.

As an optional implementation manner, the calculating and obtaining the first standard value according to the plurality of first principal components further includes:

Input the multiple first principal components into the pre-trained PCA model, and output the first standard value as .

During specific implementation, the second standard value calculated according to the plurality of second principal components includes:

The plurality of second principal components are input to the pre-trained PCA model, and the second standard value is outputted, where is the significance level, and is a distribution that obeys the first degree of freedom and the second degree of freedom.

As an optional implementation manner, the calculating and obtaining the second standard value according to the plurality of second principal components further includes:

Input the plurality of second principal components into the pre-trained PCA model, and output the second standard value as .

In one embodiment of the present application, the standard value is standard information when the component is in normal operation.

S13. Generate an envelope curve according to the standard value.

In an embodiment of the present application, the generating the envelope curve according to the standard value includes: generating a first envelope curve according to the first standard value; generating a second envelope curve according to the second standard value.

In an embodiment of the present application, the generating the first envelope curve according to the first standard value includes: setting a first upper deviation value and a first lower deviation value; The upper deviation value determines the first upper envelope curve; the first lower envelope curve is determined according to the first standard value and the first lower deviation value; the first upper envelope curve and the first lower envelope curve are determined as The first envelope curve.

During specific implementation, the sum of the first standard value and the first upper deviation value is used as the first upper vertex, and the difference between the first standard value and the first lower deviation value is used as the first lower vertex. Draw a first information graph according to the first information in the step S11 , for example, the information graph shown in FIG. 4 . Then mark the first upper vertex and the first lower vertex in the first information graph. Finding a maximum value point smaller than the first upper vertex in the first information graph, and connecting the first upper vertex and the maximum value point to obtain the first upper envelope curve. Finding a minimum value point smaller than the first lower vertex in the first information graph, and connecting the first lower vertex and the minimum value point to obtain the first lower envelope curve. The first upper envelope curve and the first lower envelope curve are determined as the first envelope curve. For example, the envelope graph shown in Figure 5.

In an embodiment of the present application, the generating the second envelope curve according to the second standard value includes: setting a second upper deviation value and a second lower deviation value; according to the second standard value and the second The upper deviation value determines the second upper envelope curve; determines the second lower envelope curve according to the second standard value and the second lower deviation value; determines the second upper envelope curve and the second lower envelope curve as The second envelope curve.

During specific implementation, the sum of the second standard value and the second upper deviation value is used as the second upper vertex, and the difference between the second standard value and the second lower deviation value is used as the second lower vertex. Draw a second information graph according to the second information in the step S11. Then mark the second upper vertex and the second lower vertex in the second information graph. Finding a maximum value point smaller than the second vertex in the second information graph, and connecting the second upper vertex and the maximum value point to obtain the second upper envelope curve. Finding a minimum value point smaller than the second lower vertex in the second information graph, and connecting the second lower vertex and the minimum value point to obtain the second lower envelope curve. The second upper envelope curve and the second lower envelope curve are determined as the second envelope curve.

By drawing the envelope curve according to the standard value of the normal information, the interval of the normal information is defined, and the normal information and the abnormal information can be directly determined through the envelope curve subsequently.

S14. Obtain the information to be tested when the component is currently running, and detect whether the component is running abnormally according to the information to be tested and the envelope curve.

In one embodiment of the present application, the acquisition of the information to be tested when the component is currently running, and detecting whether the component is currently running abnormally according to the information to be tested and the above envelope curve includes:

(1) Obtaining first transient information of the forging press through the first sensor, and obtaining second transient information of the forging die through the second sensor. During specific implementation, after the first transient information and the second transient information are obtained, the first transient information and the second transient information are segmented through the laser trigger device. Irregular information and redundant information can be removed through information segmentation to obtain useful information.

(2) Obtaining a first statistical value based on the first transient information, and obtaining a second statistical value based on the second transient information. Wherein, the first statistical value is a first T2 value, and the second statistical value is a second T2 value. During specific implementation, the PCA algorithm is used to calculate the first transient information to obtain the first T2 value, and the PCA algorithm is used to calculate the second transient information to obtain the second Statistics T2.

(3) Determine the first detection result according to the first statistical value and the first envelope curve, and determine the second detection result according to the second statistical value and the second envelope curve. During specific implementation, it is judged whether the first T2 value is included in the first envelope curve, and when the first T2 value is included in the first envelope curve, it means that the first transient information is normal Information, determine that the first detection result is normal. When the first T2 value is not included in the first envelope curve, it indicates that the first transient information is abnormal information, and the first detection result is determined to be abnormal. Then judge whether the second T2 value is included in the second envelope curve, when the second T2 value is included in the second envelope curve, it means that the second transient information is normal information, and determine The second detection result is normal. When the second T2 value is not included in the second envelope curve, it indicates that the second transient information is abnormal information, and the second detection result is determined to be abnormal.

(4) Determine whether there is an abnormality in the forging process according to the first detection result and the second detection result. When each of the first detection result and the second detection result is normal, it is determined that the forging process is normal. When at least one of the first detection result and the second detection result is abnormal, it is determined that the forging process is abnormal, that is, when the first abnormal detection result is abnormal, and the When the second abnormality detection result is no abnormality, it is determined that there is an abnormality in the forging process; when the first abnormality detection result is no abnormality, and the second abnormality detection result is abnormal, it is determined that there is an abnormality in the forging process , when the first abnormality detection result is abnormal and the second abnormality detection result is abnormal, it is determined that the forging process is abnormal. And the detection results are displayed on the information analysis interface. When there is an abnormality in the forging process, an alarm message is generated to remind the staff to check it in time.

By continuously collecting and analyzing information and conducting production verification, the abnormal rate of products can be reduced. At the same time, the combination of forging envelope curve and PCA algorithm is used to detect whether there is anomaly in the forging process, which can improve the accuracy of anomaly detection.

As an optional implementation, the method further includes: collecting the accuracy rate of abnormal alarms, and when the accuracy rate is less than a preset threshold, reacquiring information about the normal operation of the components, and performing step S11- Step S14.

Please continue to refer to FIG. 2 , in this embodiment, the memory 121 may be an internal memory of the electronic device 12 , that is, a memory built in the electronic device 12 . In other embodiments, the memory 121 may also be an external memory of the electronic device 12 , that is, a memory externally connected to the electronic device 12 .

In some embodiments, the memory 121 is used to store program codes and various information, and realize high-speed and automatic access to programs or information during the operation of the electronic device 12 .

The memory 121 may include random access memory, and may also include non-volatile memory, such as hard disk, memory, plug-in hard disk, smart memory card (Smart Media Card, SMC), secure digital ( Secure Digital (SD) card, flash memory card (Flash Card), at least one disk memory device, flash memory device, or other volatile solid-state memory device.

In one embodiment, the processor 122 may be a central processing unit (Central Processing Unit, CPU), and may also be other general-purpose processors, digital signal processors (Digital Signal Processor, DSP), application-specific integrated circuits (Application Specific Integrated Circuit, ASIC), Field-Programmable Gate Array (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor, or the processor may be any other conventional processor and the like.

If the program codes and various information in the memory 121 are implemented in the form of software function units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the processes in the methods of the above-mentioned embodiments, such as the anomaly detection method, which can also be completed by instructing related hardware through computer programs, and the computer programs can be stored in a computer-readable In the storage medium, when the computer program is executed by the processor, the steps of the above-mentioned various method embodiments can be realized. Wherein, the computer program includes computer program code, and the computer program code may be in the form of original code, object code, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM, Read- Only Memory), etc.

It can be understood that the module division described above is a logical function division, and there may be another division method in actual implementation. In addition, each functional module in each embodiment of the present application may be integrated into the same processing unit, or each module may exist separately physically, or two or more modules may be integrated into the same unit. The above-mentioned integrated modules can be implemented in the form of hardware, or in the form of hardware plus software function modules.

Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application without limitation. Although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be Make modifications or equivalent replacements without departing from the spirit and scope of the technical solutions of the present application.

1: Anomaly detection system 10: Sensor 101: The first sensor 102: Second sensor 11: Information acquisition card 111:Laser trigger device 12: Electronic equipment 121: Memory 122: Processor 123: communication bus 13: Forging press 14: Forging die S11~S14: Steps

In order to more clearly illustrate the technical solutions in the embodiments of the present application or in the prior art, the accompanying drawings that need to be used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the accompanying drawings in the following description are only It is an embodiment of the present application, and those skilled in the art can also obtain other drawings according to the provided drawings without creative work.

FIG. 1 is a schematic structural diagram of an anomaly detection system provided by an embodiment of the present application.

FIG. 2 is a schematic structural diagram of an electronic device for an abnormality detection method provided by an embodiment of the present application.

FIG. 3 is a flow chart of an abnormality detection method provided by an embodiment of the present application.

FIG. 4 is an exemplary information graph provided by an embodiment of the present application.

Fig. 5 is an exemplary envelope curve diagram provided by an embodiment of the present application.

S11~S14: Steps

Claims (10)

An anomaly detection method, wherein the anomaly detection method includes: Obtain the information of the normal operation of the components, and extract multiple principal components of the information through the principal component analysis algorithm; calculating standard values based on the plurality of principal components; generating an envelope curve according to said standard value; Obtaining the information to be tested when the component is currently running, and detecting whether the component is running abnormally according to the information to be tested and the envelope curve. The abnormality detection method as described in claim item 1, wherein said acquisition of information of components during normal operation, and extracting multiple principal components of said information through a principal component analysis algorithm include: Obtain the first information when the forging press is in normal operation through the first sensor, and obtain the second information when the forging die is in normal operation through the second sensor; A plurality of first principal components of the first information are extracted through a principal component analysis algorithm, and a plurality of second principal components of the second information are extracted through a principal component analysis algorithm. The anomaly detection method according to claim 2, wherein said calculating standard values according to said plurality of principal components includes: calculating a first standard value according to the plurality of first principal components; A second standard value is obtained by calculating the plurality of second principal components. The anomaly detection method according to claim 3, wherein said generating an envelope curve according to said standard value comprises: determining a first upper envelope curve according to the first standard value and a preset first upper deviation value; determining a first lower envelope curve according to the first standard value and a preset first lower deviation value; determining the first upper envelope curve and the first lower envelope curve as the first envelope curve; determining a second upper envelope curve according to the second standard value and a preset second upper deviation value; determining a second lower envelope curve according to the second standard value and a preset second lower deviation value; The second upper envelope curve and the second lower envelope curve are determined as the second envelope curve. The abnormality detection method as described in claim item 4, wherein said obtaining the information to be tested when the component is currently running, and detecting whether the component is currently running abnormally according to the information to be tested and the envelope curve includes: acquiring first transient information of the forging press through the first sensor, and acquiring second transient information of the forging die through the second sensor; obtaining a first statistical value based on the first transient information, and obtaining a second statistical value based on the second transient information; determining the first detection result according to the first statistical value and the first envelope curve, and determining the second detection result according to the second statistical value and the second envelope curve; Determine whether there is an abnormality in the forging process according to the first detection result and the second detection result. The anomaly detection method according to claim 5, wherein said determining the first detection result according to the first statistical value and the first envelope curve comprises: judging whether the first statistical value is included in the first envelope curve; When the first statistical value is included in the first envelope curve, determining that the first detection result is normal; When the first statistical value is not included in the first envelope curve, it is determined that the first detection result is abnormal. The abnormality detection method according to claim 6, wherein said determining whether there is an abnormality in the forging process according to the first detection result and the second detection result includes: When each of the first detection result and the second detection result is normal, it is determined that the forging process is normal; When at least one of the first detection result and the second detection result is abnormal, it is determined that the forging process is abnormal. An anomaly detection system, wherein the anomaly detection system includes: Sensors are used to collect information about components; an information acquisition card, communicatively connected to the sensor, for receiving the information; The electronic device is communicatively connected with the information acquisition card, and is used to execute the abnormality detection method as described in the request items 1 to 7, perform information analysis on the information, obtain the abnormality detection result, and determine the abnormality detection result according to the abnormality detection result. Check whether the above components are abnormal. An electronic device, wherein the electronic device includes a processor and a memory, and the processor is used to execute a computer program stored in the memory to implement the abnormality detection method described in any one of claim 1 to claim 7 . A computer-readable storage medium, wherein the computer-readable storage medium stores at least one instruction, and when the at least one instruction is executed by a processor, the abnormality detection as described in any one of request item 1 to request item 7 is realized method.

Images