Abnormality detection method, device, and equipment for a stamping apparatus, and a medium

Abstract

The application discloses an abnormality detection method, device and equipment of a stamping equipment and a medium. The abnormality detection method of the stamping equipment comprises the following steps: acquiring first measurement data of the stamping equipment in real time through a sensor; acquiring second measurement data of the stamping equipment as reference data by a tonnage measuring device; performing a data feature extraction operation on the first measurement data to obtain effective data corresponding to the first measurement data; inputting the reference data as a training set into a convolutional neural network model to perform training, obtaining an equipment detection model, and calibrating the effective data based on the equipment detection model; and detecting whether the stamping equipment is abnormal based on the error data between the effective data and the reference data based on the equipment detection model. The application can quickly judge the fault of the stamping equipment or the die, thereby reducing the repair cost of the stamping equipment or the production cost of the die, and further improving the efficiency of the stamping equipment operation.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of equipment detection, and particularly relates to an abnormality detection method, device, equipment and medium for a stamping equipment. BACKGROUND

[0002] Currently, the pain point in the stamping field is that it is impossible to quickly distinguish the faults of stamping equipment or dies in the stamping production process, resulting in low fault detection efficiency in the stamping production process. The industry currently has a relatively high requirement for the accuracy of abnormality detection of stamping equipment. Therefore, the long-term working state of the stamping equipment needs to be monitored in the stamping production process, so as to avoid the loss caused by the interruption of production due to the abnormality of the stamping equipment. SUMMARY

[0003] The application provides an abnormality detection method, device, equipment and medium for a stamping equipment, and aims to solve the technical problems that the abnormality detection method for the stamping equipment is not fast enough in detecting faults, and the detection efficiency is low because historical detection data cannot be fed back in time as a reference.

[0004] The first aspect of the application provides an abnormality detection method for a stamping equipment, which comprises the following steps:

[0005] obtaining first measurement data of the stamping equipment in real time through a sensor;

[0006] obtaining second measurement data of the stamping equipment through a tonnage measuring device as reference data;

[0007] performing a data feature extraction operation on the first measurement data to obtain effective data corresponding to the first measurement data;

[0008] inputting the reference data as a training set into a convolutional neural network model for training to obtain an equipment detection model, and calibrating the effective data based on the equipment detection model;

[0009] detecting whether the stamping equipment is abnormal based on the equipment detection model and error data between the effective data and the reference data.

[0010] In a possible implementation, the step of obtaining first measurement data of the stamping equipment in real time through a sensor comprises the following steps:

[0011] obtaining pressure data of at least one position on the stamping equipment in real time through the sensor as the first measurement data, wherein the at least one position comprises a connecting rod position, two side positions and a slider position.

[0012] In a possible implementation, the performing a data feature extraction operation on the first measurement data to obtain effective data corresponding to the first measurement data comprises:

[0013] performing a feature transformation on the first measurement data obtained in real time by the sensor to obtain frequency domain data;

[0014] filtering high-frequency signals in the frequency domain data;

[0015] obtaining effective frequency domain data by performing a windowing operation on the filtered frequency domain data;

[0016] performing an inverse feature transformation on the effective frequency domain data to obtain the effective data.

[0017] In a possible implementation, the method further comprises:

[0018] after obtaining the effective frequency domain data, determining whether a curve corresponding to the effective frequency domain data conforms to a Gaussian distribution curve, comprising:

[0019] if it is determined that the curve corresponding to the effective frequency domain data conforms to the Gaussian distribution curve, performing an inverse feature transformation on the effective frequency domain data to obtain the effective data; or

[0020] if it is determined that the curve corresponding to the effective frequency domain data does not conform to the Gaussian distribution curve, continuously obtaining updated first measurement data until the curve corresponding to the effective frequency domain data obtained based on the updated first measurement data conforms to the Gaussian distribution curve.

[0021] In a possible implementation, the inputting the reference data as a training set into a convolutional neural network model to train the convolutional neural network model to obtain a device detection model, and calibrating the effective data based on the device detection model comprises:

[0022] obtaining a mapping relationship between the effective data and reference data of the stamping device;

[0023] inputting the effective data and the mapping relationship into the convolutional neural network model to train the convolutional neural network model to establish the device detection model.

[0024] In a possible implementation, after the device detection model of the stamping device is established, the method further comprises:

[0025] obtaining historical error data between the effective data and reference data of the stamping device based on the device detection model; adjusting parameters of the device detection model and updating the device detection model according to the device detection model, the effective data, the reference data, the mapping relationship, and the historical error data.

[0026] In a possible implementation, the detecting whether the stamping device is abnormal based on the device detection model according to error data between the effective data and the reference data comprises:

[0027] inputting the real-time acquired effective data of the stamping device into the device detection model to obtain actual error data;

[0028] if the actual error data is greater than the calibrated error data, determining that the stamping device is abnormal, or if the actual error data is less than or equal to the calibrated error data, determining that the stamping device is normal.

[0029] The application further provides a stamping device abnormality detection device, which comprises:

[0030] a first acquisition unit configured to acquire first measurement data of a stamping device in real time through a sensor;

[0031] a second acquisition unit configured to acquire second measurement data of the stamping device through a tonnage measuring device as reference data;

[0032] an extraction unit configured to perform a data feature extraction operation on the first measurement data to obtain effective data corresponding to the first measurement data;

[0033] a calibration unit configured to input the reference data as a training set into a convolutional neural network model to obtain a device detection model, and calibrate the effective data based on the device detection model;

[0034] a detection unit configured to detect whether the stamping device is abnormal based on the device detection model according to error data between the effective data and the reference data.

[0035] The application further provides an electronic device, which comprises a processor and a memory, and the processor is configured to execute a computer program stored in the memory to implement the stamping device abnormality detection method.

[0036] The application further provides a computer readable storage medium, which stores at least one instruction, and the at least one instruction is executed by a processor to implement the stamping device abnormality detection method.

[0037] This application enables rapid diagnosis of stamping equipment or mold malfunctions, facilitating quicker and more accurate detection of these issues. This reduces repair costs and production costs, thereby improving operational efficiency. Furthermore, it provides real-time monitoring data feedback, allowing for real-time data review and analysis of calibration curves to identify and diagnose stamping equipment or mold malfunctions online. This prevents improper operation or other faults from causing molds to fail to meet standards during mass production, thus avoiding significant waste of manpower and resources and substantial production losses. Ultimately, this reduces losses during factory production and significantly improves the operational efficiency of stamping equipment. Attached Figure Description

[0038] Figure 1 This is a flowchart of an anomaly detection method for a stamping equipment provided in an embodiment of this application.

[0039] Figure 2 This is a diagram of data collected by the sensor provided in the embodiments of this application.

[0040] Figure 3 This is a signal waveform diagram of the sensor provided in the embodiments of this application.

[0041] Figure 4 This is a framework diagram of an anomaly detection device for a stamping equipment provided in an embodiment of this application.

[0042] Figure 5 This is a schematic diagram of the structure of an electronic device for implementing an anomaly detection method for stamping equipment, as provided in an embodiment of this application.

[0043] Explanation of main component symbols

[0044] first acquisition unit 41 second acquisition unit 42 extraction unit 43 calibration unit 44 detection unit 45 electronic device 1 memory 11 processor 12 input / output device 13

[0045] The following detailed description, in conjunction with the accompanying drawings, will further illustrate the present invention. Detailed Implementation

[0046] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0047] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0048] In addition, the terms "first", "second", etc. are used only for descriptive purposes and are not to be construed as indicating or implying relative importance or an ordered ranking of the indicated technical features. Thus, features defined with "first", "second" can explicitly or implicitly include one or more of the features. In the description of the present application, the meaning of "a plurality of" is two or more, unless otherwise explicitly and specifically limited.

[0049] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used in the description of the specification herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. As used herein, the term "and / or" includes any and all combinations of one or more of the associated listed items. There are generally two technical problems in the abnormality detection method of the stamping equipment, the first is that the stamping equipment detects faults not fast enough, mainly because the sensor on the stamping equipment collects data not sensitive enough, the second is that the stamping equipment cannot feedback historical detection data as a reference in time, resulting in low detection efficiency. The embodiments of the present application solve the technical problems of the abnormality detection method of the stamping equipment that the detection fault is not fast enough and the historical detection data cannot be fed back in time as a reference, resulting in low detection efficiency.

[0050] Based on the above technical problems, the embodiments of the present application provide an abnormality detection method of a stamping equipment, which can quickly judge the fault of the stamping equipment or the die, facilitate to quickly and accurately find the fault of the stamping equipment or the die in time, thereby reducing the repair cost of the stamping equipment or the production cost of the die, and further improving the efficiency of the stamping equipment operation. At the same time, the monitoring data can be fed back in time, facilitating to view the monitoring data in real time, analyze the calibration curve according to the monitoring data, and judge and find the fault of the stamping equipment or the die in real time, so as to avoid that the die production does not meet the standard in the batch production operation process of the stamping equipment due to improper operation or other faults, thereby causing the factory to waste a large amount of manpower and material resources cost and huge production cost, thereby reducing the loss in the production operation process of the factory, and greatly improving the operation efficiency of the stamping equipment.

[0051] Figure 1 An implementation flowchart of an abnormality detection method of a stamping equipment provided by the embodiments of the present application is shown. The order of steps in the flowchart can be changed according to different needs, and some steps can be omitted.

[0052] S101: Real-time acquisition of first measurement data of the stamping equipment by the sensor.

[0053] In an embodiment of the present application, the stamping device is mainly applied to a production die workshop, and is mainly used for blanking, punching, forming, drawing, trimming, fine punching, shaping, riveting, extrusion and the like of a plate through a die. One or more stamping devices can be deployed in the production die workshop, and the stamping devices can be connected to an industrial computer, an upper computer and the like. The stamping device includes, but is not limited to, a punch, and is not particularly limited.

[0054] It should be noted that the first measurement data of the stamping device is acquired in real time by the sensor. The first measurement data is mainly the value of the impact force in the product stamping forming process, which is detected and recorded in real time by the sensor. The product stamping forming process includes a descending process and an ascending process. The impact force gradually increases in the descending process, and the impact force gradually decreases in the ascending process. Correspondingly, the first measurement data gradually increases and then gradually decreases in the product stamping forming process. The sensor can be a new type of high-sensitivity ultrasonic sensor. The ultrasonic sensor has the advantages of small size, fast response, wide measurement frequency range, high linearity, and no need for external power supply. At the same time, it is convenient to paste on the structure of the punch, the stamping die, the motor and the like, so as to facilitate the sensor to accurately detect and record the specific value of the impact force in the product stamping forming process.

[0055] Specifically, the pressure data of at least one position of the stamping device is acquired in real time by the sensor as the first measurement data in the entire stamping process. The at least one position includes a connecting rod position, two side positions and a slider position.

[0056] It should be noted that the sensor can be pasted on at least one of the connecting rod position, the two side positions and the slider position of the stamping device. In this way, the sensor can collect and record the pressure data of at least one of the connecting rod position, the two side positions and the slider position of the stamping device. For example, as shown in Figure 2 The first sensor is pasted on the connecting rod position of the stamping device, the second sensor is pasted on the two side positions of the stamping device, and the third sensor is pasted on the slider position of the stamping device. At each moment, the first sensor, the second sensor and the third sensor record the pressure data at the positions, thereby improving the accuracy of collecting the pressure data.

[0057] S102: Acquire the second measurement data of the stamping device by the tonnage measuring device as the reference data.

[0058] In the embodiment of the present application, the tonnage measuring device is placed directly below the stamping equipment driving assembly, and the stamping equipment driving assembly generates impact force on the tonnage measuring device during the descending and ascending processes. Thus, the tonnage measuring device driving assembly can sense and record the pressure data generated by the impact force in real time during the descending and ascending processes. The pressure data generated by the impact force at each moment during the descending and ascending processes of the tonnage measuring device can reflect the stress of the product during the forming process, and thus the pressure data collected by the tonnage measuring device can be used as reference data for comparison with the pressure data collected by the sensor. Thus, the abnormality of the punch or the die can be reminded to the factory operator.

[0059] It should be noted that the tonnage measuring device includes but is not limited to a tonnage meter, and is not specifically limited.

[0060] S103: performing a data feature extraction operation on the first measurement data to obtain effective data corresponding to the first measurement data.

[0061] In the embodiment of the present application, a plurality of sensors collect and record pressure data at a plurality of positions of the stamping equipment at each moment, as shown in FIG. 1. Figure 3 As shown in FIG. 1, the pressure data at three positions of the stamping equipment collected and recorded by the three sensors are presented in the form of signal waveforms. When the pressure data at the three positions of the stamping equipment are normal, the pressure data are presented in the form of Normal. When the pressure data at the three positions of the stamping equipment are abnormal, the pressure data are presented in the form of AbNormal.

[0062] It should be noted that in actual application, the pressure data obtained by the sensor often contains a large amount of redundant data types. In order to improve the efficiency and accuracy of subsequent data analysis, effective pressure data need to be extracted from a large amount of pressure data as effective data.

[0063] Specifically, performing feature transformation on the first measurement data obtained by the sensor in real time to obtain frequency domain data;

[0064] Filtering high-frequency signals in the frequency domain data;

[0065] Performing a windowing operation on the filtered frequency domain data to obtain effective frequency domain data;

[0066] Performing inverse feature transformation on the effective frequency domain data to obtain effective data.

[0067] It should be noted that the pressure signal obtained by the sensor is first filtered by a filter to filter out high-frequency interference time domain signal, then Fourier transform is performed to convert the time domain signal into a frequency domain signal, further remove high-frequency interference signals in the frequency domain signal, and finally inverse Fourier transform is performed to convert the frequency domain signal filtered out of the high-frequency interference signal into a time domain signal. At this time, the time domain signal is an effective signal.

[0068] Further specifically, the method further comprises:

[0069] After obtaining the effective frequency domain data, it is judged whether the curve corresponding to the effective frequency domain data conforms to a Gaussian distribution curve, comprising:

[0070] If it is judged that the curve corresponding to the effective frequency domain data conforms to the Gaussian distribution curve, inverse feature transformation is performed on the effective frequency domain data to obtain the effective data; or

[0071] If it is judged that the curve corresponding to the effective frequency domain data does not conform to the Gaussian distribution curve, the updated first measurement data is continuously acquired until the curve corresponding to the effective frequency domain data obtained based on the updated first measurement data conforms to the Gaussian distribution curve.

[0072] It should be noted that when the curve corresponding to the effective frequency domain data conforms to the Gaussian distribution curve, the frequency domain data is abnormal data, and then the effective frequency domain data is further subjected to inverse feature transformation to obtain the effective data.

[0073] When the curve corresponding to the effective frequency domain data does not conform to the Gaussian distribution curve, the pressure data of the stamping equipment is continuously collected, and the above data processing process is repeated until the curve corresponding to the frequency domain data conforms to the Gaussian distribution curve.

[0074] S104: input the reference data as a training set into a convolutional neural network model for training to obtain an equipment detection model, and calibrate the effective data based on the equipment detection model.

[0075] In the embodiment of the present application, the convolutional neural network model comprises an input layer, a first convolutional layer, a first down-sampling layer, a second convolutional layer, a second down-sampling layer, a full connection layer and an output layer. The input of the first convolutional layer is connected with the input layer, and the output of the first convolutional layer is connected with the input of the first down-sampling layer; the output of the first down-sampling layer is connected with the input of the second convolutional layer; the output of the second convolutional layer is connected with the input of the second down-sampling; the output of the second down-sampling layer is connected with the input of the output layer through the full connection layer. The training set with the reference data as the training sample is input into the input layer of the convolutional neural network model, and the convolutional neural network model is iterated and converged automatically until the error between the output value and the target value meets the expectation, so as to establish the device detection model, and the effective data is calibrated according to the device detection model.

[0076] Specifically, a mapping relationship between the effective data of the stamping device and the reference data is obtained;

[0077] The effective data and the mapping relationship are input into the convolutional neural network model for training, and the device detection model of the stamping device is established.

[0078] Optionally, after the device detection model of the stamping device is established, the method further comprises:

[0079] Based on the device detection model, historical error data between the effective data of the stamping device and the reference data is obtained;

[0080] According to the device detection model, the effective data, the reference data, the mapping relationship and the historical error data, the parameters of the device detection model are adjusted, and the device detection model is updated.

[0081] S105: Based on the device detection model, whether the stamping device is abnormal is detected according to the error data between the effective data and the reference data.

[0082] Specifically, the effective data of the stamping device acquired in real time is input into the device detection model to obtain actual error data;

[0083] If the actual error data is greater than the calibrated error data, it is determined that the stamping device is abnormal, or if the actual error data is less than or equal to the calibrated error data, it is determined that the stamping device is normal.

[0084] Figure 4 A framework diagram of a stamping device abnormality detection device provided by an embodiment of the present application is shown, only the parts related to the embodiments of the present application are shown for the convenience of description, and the details are as follows:

[0085] The first acquisition unit 41 is configured to acquire first measurement data of the stamping equipment in real time through the sensor.

[0086] In an embodiment of the present application, the stamping equipment is mainly applied to a production die workshop and is mainly used for blanking, punching, forming, drawing, trimming, fine blanking, shaping, riveting, extrusion and the like of a plate through a die. One or more stamping equipment can be deployed in the production die workshop, and the stamping equipment can be connected to an industrial computer, an upper computer and the like. The stamping equipment includes but is not limited to a punch, and the specific type is not limited.

[0087] It should be noted that the first measurement data of the stamping equipment is acquired in real time through the sensor. The first measurement data is mainly the value of the impact force in the product stamping forming process, which is detected and recorded in real time by the sensor. The product stamping forming process includes a descending process and an ascending process. The impact force gradually increases in the descending process, and the impact force gradually decreases in the ascending process. Correspondingly, the first measurement data gradually increases and then gradually decreases in the product stamping forming process. The sensor can be a new type of high-sensitivity ultrasonic sensor. The ultrasonic sensor has the advantages of small size, fast response, wide measurement frequency range, high linearity and no need for external power supply. At the same time, it is convenient to paste on the structure of the punch, the stamping die and the motor, so as to facilitate the sensor to accurately detect and record the specific value of the impact force in the product stamping forming process.

[0088] The second acquisition unit 42 is configured to acquire second measurement data of the stamping equipment through a tonnage measuring device as reference data.

[0089] In an embodiment of the present application, the tonnage measuring device is arranged directly below the driving assembly of the stamping equipment. The driving assembly of the stamping equipment generates an impact force on the tonnage measuring device in the descending and ascending processes. Therefore, the tonnage measuring device generates pressure data in real time in response to the impact force in the descending and ascending processes. The pressure data generated by the tonnage measuring device in each moment in the descending and ascending processes of the driving assembly of the stamping equipment truly reflects the stress of the product in the forming process. Therefore, the pressure data collected by the tonnage measuring device can be used as reference data and compared with the pressure data collected by the sensor. Thus, it is convenient to remind the factory operator of the abnormal punch or die.

[0090] It should be noted that the tonnage measuring device includes but is not limited to a tonnage meter, and the specific type is not limited.

[0091] The extraction unit 43 is configured to perform a data feature extraction operation on the first measurement data to obtain effective data corresponding to the first measurement data.

[0092] In an embodiment of the present application, a plurality of sensors collect and record pressure data at a plurality of positions of the stamping equipment at each moment, such asFigure 3 As shown, the pressure data of the three positions of the stamping device collected and recorded by the three sensors at each moment is presented in the form of a signal waveform, and the pressure data of the three positions of the stamping device is presented as Normal if no abnormality occurs, and is presented as AbNormal if abnormality occurs.

[0093] It should be noted that in actual application, the pressure data obtained by the sensor often contains a large amount of redundant data types, and in order to improve the efficiency and accuracy of subsequent data analysis, it is necessary to extract effective pressure data from a large amount of pressure data as effective data.

[0094] The calibration unit 44 is configured to input the reference data as a training set into the convolutional neural network model for training to obtain a device detection model, and calibrate the effective data based on the device detection model.

[0095] In the embodiment of the present application, the convolutional neural network model includes an input layer, a first convolutional layer, a first down-sampling layer, a second convolutional layer, a second down-sampling layer, a full connection layer and an output layer. The input of the first convolutional layer is connected with the input layer, and the output of the first convolutional layer is connected with the input of the first down-sampling layer; the output of the first down-sampling layer is connected with the input of the second convolutional layer; the output of the second convolutional layer is connected with the input of the second down-sampling; the output of the second down-sampling layer is connected with the input of the output layer through the full connection layer. The reference data is input into the input layer of the convolutional neural network model as a training set of training samples, and is iterated and converged automatically through the convolutional neural network model until the error between the output value and the target value meets the expectation, so as to establish a device detection model, and calibrate the effective data according to the device detection model.

[0096] The detection unit 45 is configured to detect whether the stamping device is abnormal based on the error data between the effective data and the reference data according to the device detection model.

[0097] Specifically, the effective data of the stamping device obtained in real time is input into the device detection model to obtain actual error data;

[0098] If the actual error data is greater than the calibrated error data, it is determined that the stamping device is abnormal, or if the actual error data is less than or equal to the calibrated error data, it is determined that the stamping device is normal.

[0099] Figure 5 is a structural schematic diagram of an electronic device 1 provided by a preferred embodiment of the present application to implement a kind of stamping equipment abnormality detection method. As shown in Figure 5As shown, the electronic device 1 comprises a memory 11, a processor 12 and an input / output device 13.

[0100] The electronic device 1 is a device capable of automatically performing numerical calculation and / or information processing according to pre-set or stored instructions, and its hardware includes but is not limited to microprocessors, application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), digital signal processors (DSPs), embedded devices, etc.

[0101] The electronic device 1 can be any electronic product that can interact with users, such as personal computers, tablet computers, smart phones, personal digital assistants (PDAs), game consoles, interactive Internet protocol televisions (IPTVs), smart wearable devices, etc. The electronic device 1 can be a server, which includes but is not limited to a single network server, a server group composed of multiple network servers, or a cloud composed of a large number of hosts or network servers based on cloud computing. Cloud computing is a kind of distributed computing, which is composed of a super virtual computer formed by a group of loosely coupled computers. The network in which the electronic device 1 is located includes but is not limited to the Internet, a wide area network, a metropolitan area network, a local area network, a virtual private network (VPN), etc.

[0102] The memory 11 is used to store the program and various data of the abnormality detection method of a stamping device, and to realize high-speed and automatic access to the program or data during the operation of the electronic device 1. The memory 11 can be an external storage device and / or an internal storage device of the electronic device 1. Further, the memory 11 can be a circuit without a physical form in an integrated circuit, such as a RAM (Random-Access Memory), a FIFO (First In First Out), etc., or the memory 11 can also be a storage device with a physical form, such as a memory stick, a TF card, etc.

[0103] The processor 12 can be a central processing unit (CPU). The CPU is a super large scale integrated circuit, and is an operation core and a control unit of the electronic device 1. The processor 12 can execute an operating system and installed various application programs, program codes and the like of the electronic device 1, for example, executes an operating system and installed various application programs, program codes in each module or unit in the abnormality detection apparatus of a stamping device, to implement the abnormality detection method of the stamping device.

[0104] The input and output device 13 is mainly used to realize the input and output functions of the electronic device 1, such as receiving and transmitting input digital or character information, or displaying information input by the user or information provided to the user and various menus of the electronic device 1.

[0105] The modules / units integrated in the electronic device 1 are stored in a computer readable storage medium if they are implemented in the form of software function units and sold or used as independent products. Based on such understanding, all or part of the processes in the above-mentioned embodiment methods can also be instructed by a computer program to related hardware to complete. The computer program can be stored in a computer readable storage medium. When the computer program is executed by the processor, the steps of the above-mentioned various method embodiments can be implemented. The computer program includes computer program code, which can be in the form of source code, object code, executable files or some intermediate forms. The computer readable medium can include any entity or device, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal and software distribution medium, etc. that can carry the computer program code. It should be noted that the content included in the computer readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction, for example, in some jurisdictions, according to legislation and patent practice, the computer readable medium does not include electrical carrier signals and telecommunication signals.

[0106] In several embodiments provided in the present application, it should be understood that the disclosed system, device and method can be implemented in other ways. For example, the above-described device embodiments are only schematic, for example, the division of the modules is only a logical function division, and there can be another division manner in actual implementation.

[0107] The modules described as separate components may or may not be physically separate, and the components shown as modules may or may not be physical units, i.e., may be located in one place, or may be distributed to multiple network units. Part or all of the modules can be selected according to actual needs to achieve the purpose of the embodiment.

[0108] In addition, each functional module in each embodiment of the present application can be integrated in one processing unit, or each unit can be physically present alone, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of hardware plus software functional modules.

[0109] It is obvious for those skilled in the art that the present application is not limited to the details of the above exemplary embodiments, and can be implemented in other specific forms without departing from the spirit or essential characteristics of the present application. Therefore, the embodiments should be regarded as exemplary and non-limiting, and the scope of the present application is defined by the appended claims rather than the above description, and all changes falling within the meaning and scope of the equivalent elements of the claims are intended to be included in the present application. Any reference signs in the claims should not be regarded as limiting the claims. In addition, it is obvious that the word "comprise" does not exclude other units or steps, and the singular does not exclude the plural. The plurality of modules or devices stated in the system claims can also be implemented by one module or device through software or hardware. The words first, second, etc. are used to indicate names, not any specific order.

[0110] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application and not to limit it, and 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 modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present application.

Claims

1. An abnormality detection method of a press apparatus, characterized by, The abnormality detection method of the stamping equipment comprises: Real-time acquisition of first measurement data of the stamping equipment by a sensor; Acquisition of second measurement data of the stamping equipment by a tonnage measuring device as reference data; Data feature extraction operation is performed on the first measurement data to obtain effective data corresponding to the first measurement data; The reference data is input into a convolutional neural network model as a training set for training to obtain a first equipment detection model, and the effective data is calibrated based on the first equipment detection model, including: obtaining a mapping relationship between the effective data and the reference data of the stamping equipment; the effective data and the mapping relationship are input into the convolutional neural network model for training to establish a second equipment detection model; Based on the second equipment detection model, whether the stamping equipment is abnormal is detected according to error data between the effective data and the reference data, including: inputting the real-time acquired effective data of the stamping equipment into the second equipment detection model to obtain actual error data; if the actual error data is greater than the calibrated error data, it is determined that the stamping equipment is abnormal, or if the actual error data is less than or equal to the calibrated error data, it is determined that the stamping equipment is normal.

2. The abnormality detection method of a press apparatus according to Claim 1, characterized by, The real-time acquisition of first measurement data of the stamping equipment by the sensor comprises: Real-time acquisition of pressure data of at least one position on the stamping equipment by the sensor as the first measurement data, the at least one position including a connecting rod position, two side positions and a slider position.

3. The abnormality detection method of a press apparatus according to Claim 2, characterized in that, The data feature extraction operation performed on the first measurement data to obtain effective data corresponding to the first measurement data comprises: Performing feature transformation on the first measurement data acquired by the sensor in real time to obtain frequency domain data; Filtering high-frequency signals in the frequency domain data; Performing windowing operation on the filtered frequency domain data to obtain effective frequency domain data; Performing inverse feature transformation on the effective frequency domain data to obtain effective data.

4. The abnormality detection method of a press apparatus according to Claim 3, characterized in that, The method further comprises: After obtaining the effective frequency domain data, it is judged whether the curve corresponding to the effective frequency domain data conforms to a Gaussian distribution curve, comprising: If it is judged that the curve corresponding to the effective frequency domain data conforms to the Gaussian distribution curve, inverse feature transformation is performed on the effective frequency domain data to obtain the effective data; or If it is judged that the curve corresponding to the effective frequency domain data does not conform to the Gaussian distribution curve, continue to acquire updated first measurement data until the curve corresponding to the effective frequency domain data obtained based on the updated first measurement data conforms to the Gaussian distribution curve.

5. The abnormality detection method of a press apparatus according to Claim 4, characterized in that, After establishing the second equipment detection model of the stamping equipment, the method further comprises: Based on the second equipment detection model, historical error data between the effective data and the reference data of the stamping equipment is obtained; According to the second equipment detection model, the effective data, the reference data, the mapping relationship and the historical error data, the parameters of the second equipment detection model are adjusted, and the second equipment detection model is updated.

6. An abnormality detection device of a press apparatus characterized by comprising: The abnormality detection device comprises: The first acquisition unit is configured to acquire first measurement data of the stamping device in real time through a sensor; The second acquisition unit is configured to acquire second measurement data of the stamping device through a tonnage measuring device as reference data; The extraction unit is configured to perform a data feature extraction operation on the first measurement data to obtain effective data corresponding to the first measurement data; The calibration unit is configured to input the reference data as a training set into a convolutional neural network model for training to obtain a first device detection model, and calibrate the effective data based on the first device detection model, including: acquiring a mapping relationship between the effective data and the reference data of the stamping device; inputting the effective data and the mapping relationship into the convolutional neural network model for training to establish a second device detection model; The detection unit is configured to detect whether the stamping device is abnormal based on the second device detection model according to error data between the effective data and the reference data, including: inputting the effective data of the stamping device acquired in real time into the second device detection model to obtain actual error data; if the actual error data is greater than calibrated error data, determining that the stamping device is abnormal, or if the actual error data is less than or equal to the calibrated error data, determining that the stamping device is normal.

7. An electronic device, comprising: The electronic device includes a processor and a memory, and the processor is configured to execute a computer program stored in the memory to implement the abnormality detection method of the stamping device according to any one of claims 1-5.

8. A computer-readable storage medium, characterized in that, The computer readable storage medium stores at least one instruction, and the at least one instruction is executed by the processor to implement the abnormality detection method of the stamping device according to any one of claims 1-5.

Images