Fault diagnosis method, device and equipment for industrial robot and storage medium

Abstract

The invention discloses a fault diagnosis method, device and equipment for an industrial robot and a storage medium, and the method comprises the steps: obtaining the modeling parameters of a target industrial robot, and building a digital twin model corresponding to the target industrial robot according to the modeling parameters; collecting real-time operation data corresponding to the target industrial robot, and determining a fault diagnosis result corresponding to the target industrial robot according to the real-time operation data and the fault diagnosis model; and a simulation display image corresponding to the target industrial robot is generated according to the real-time operation data and the digital twinborn model, and the simulation display image and the fault diagnosis result are displayed. The operation data of the industrial robot are analyzed through the fault diagnosis model to determine the fault diagnosis result, and the operation condition of the industrial robot is simulated in combination with the digital twin model, so that the accuracy of fault diagnosis of the industrial robot is improved, the real-time detection of the operation condition of the industrial robot is realized, and the maintenance efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular, to a fault diagnosis method, device, equipment and storage medium for an industrial robot. Background Art

[0002] Industrial robot technology is developing in the direction of high precision and high intelligence, and the structure of industrial robots is becoming more complex, involving technologies in many fields. As a basic component of the production system, if an industrial robot has a fault, it will cause an unexpected shutdown, and the working efficiency of the entire production line will be greatly reduced, which will cause huge economic losses to the enterprise.

[0003] However, the existing technical solutions for industrial robot fault diagnosis mainly rely on the state recognition and fault diagnosis mode of the industrial robot itself, and the fault information and fault type are reflected according to the error code, which is convenient for engineers to directly locate the type and location of the fault; or through the operating state data of the industrial robot, such as abnormalities in current, voltage, speed, temperature, vibration, and noise, and comparison with the normal operating parameters in the past, to make a maintenance judgment, and it is impossible to accurately locate the fault location and fault type. Summary of the Invention

[0004] The present invention provides a fault diagnosis method, device, equipment and storage medium for an industrial robot, which analyzes the operation data of the industrial robot through a fault diagnosis model to determine a fault diagnosis result, and combines a digital twin model to simulate the operation status of the industrial robot, thereby improving the accuracy of industrial robot fault diagnosis.

[0005] According to one aspect of the present invention, there is provided a fault diagnosis method for an industrial robot, which includes:

[0006] Obtain the modeling parameters of the target industrial robot, and establish a digital twin model corresponding to the target industrial robot according to the modeling parameters;

[0007] Collect the real-time operation data corresponding to the target industrial robot, and determine the fault diagnosis result corresponding to the target industrial robot according to the real-time operation data and the fault diagnosis model, wherein the fault diagnosis model is a neural network model pre-trained;

[0008] Generate a simulation display image corresponding to the target industrial robot according to the real-time operation data and the digital twin model, and display the simulation display image and the fault diagnosis result.

[0009] According to another aspect of the present invention, there is provided a fault diagnosis device for an industrial robot, which includes:

[0010] A twin model construction module, configured to obtain the modeling parameters of a target industrial robot and establish a digital twin model corresponding to the target industrial robot according to the modeling parameters;

[0011] A diagnosis result determination module, configured to collect real-time operation data corresponding to the target industrial robot and determine a fault diagnosis result corresponding to the target industrial robot according to the real-time operation data and a fault diagnosis model, wherein the fault diagnosis model is a neural network model pre-trained;

[0012] A real-time status monitoring module, configured to generate a simulated display image corresponding to the target industrial robot according to the real-time operation data and the digital twin model, and display the simulated display image and the fault diagnosis result.

[0013] According to another aspect of the present invention, there is provided an electronic device, which includes:

[0014] At least one processor; and

[0015] A memory communicatively connected to the at least one processor; wherein,

[0016] The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the fault diagnosis method of the industrial robot according to any embodiment of the present invention.

[0017] According to another aspect of the present invention, there is provided a computer-readable storage medium storing computer instructions for causing a processor to implement the fault diagnosis method of the industrial robot according to any embodiment of the present invention when executed.

[0018] The technical solution of the embodiment of the present invention is to obtain the modeling parameters of a target industrial robot, establish a digital twin model corresponding to the target industrial robot according to the modeling parameters, collect real-time operation data corresponding to the target industrial robot, determine a fault diagnosis result corresponding to the target industrial robot according to the real-time operation data and a fault diagnosis model, generate a simulated display image corresponding to the target industrial robot according to the real-time operation data and the digital twin model, and display the simulated display image and the fault diagnosis result. Based on the above technical solution, the fault diagnosis result is determined by analyzing the operation data of the industrial robot through the fault diagnosis model, and the operation status of the industrial robot is simulated in combination with the digital twin model, thereby improving the accuracy of the fault diagnosis of the industrial robot and realizing the real-time detection of the operation status of the industrial robot, and improving the maintenance efficiency.

[0019] It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present invention, nor is it used to limit the scope of the present invention. Other features of the present invention will become readily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the accompanying drawings required for the description of the embodiments. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0021] Figure 1 is a schematic flowchart of a fault diagnosis method for an industrial robot provided by an embodiment of the present invention;

[0022] Figure 2 is a schematic diagram of a fault diagnosis method for an industrial robot provided by an embodiment of the present invention;

[0023] Figure 3 is a flowchart of a data storage solution provided by an embodiment of the present invention;

[0024] Figure 4 is a structural block diagram of a fault diagnosis device for an industrial robot provided by an embodiment of the present invention;

[0025] Figure 5 is a schematic structural diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0026] In order to enable those skilled in the art to better understand the solutions of the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the scope of protection of the present invention.

[0027] It should be noted that the terms "first", "second", etc. in the description, claims and above-mentioned drawings of the present invention are used to distinguish similar objects, and do not necessarily have to be used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments of the present invention described here can be implemented in an order other than those illustrated or described here. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device comprising a series of steps or units does not have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0028] Embodiment 1

[0029] Figure 1 is a schematic flowchart of a fault diagnosis method for an industrial robot provided by an embodiment of the present invention. This embodiment is applicable to the situation of performing fault diagnosis based on the real-time operation data of the industrial robot and monitoring the operation state of the industrial robot in real time in combination with the digital twin mode. This method can be executed by a fault diagnosis device of the industrial robot. The fault diagnosis device of the industrial robot can be implemented in the form of hardware and / or software. The fault diagnosis device of the industrial robot can be configured in an electronic device, and the electronic device can be a terminal device or a server. As Figure 1 shown, the method includes:

[0030] S110. Obtain the modeling parameters of the target industrial robot, and establish a digital twin model corresponding to the target industrial robot according to the modeling parameters.

[0031] Among them, the target industrial robot can be an industrial robot that needs to perform real-time status monitoring and fault diagnosis. For example, the target industrial robot can be a programmable robot suitable for industrial production such as an assembly robot or a handling robot. The modeling parameters can be parameters used to establish a digital twin model of the target industrial robot, and can include physical attribute parameters, environmental parameters, dynamic operation parameters, model parameters, etc. The digital twin model can be understood as a real-time mapping of the target industrial robot in the virtual space, and is used to achieve two-way synchronization between the physical entity and the virtual model through data interaction.

[0032] Specifically, it can be achieved by importing the URDF / STEP file of the industrial robot, extracting parameters such as geometric dimensions, joint degrees of freedom, and motion ranges, and reading kinematic parameters in the PLC, such as DH parameters and joint limits, as well as dynamic parameters of the industrial robot, such as mass and inertia matrix, through OPC UA / Modbus TCP. Then, a digital twin model corresponding to the target industrial robot is established based on the collected parameters. Exemplarily, the parameters required for industrial robot modeling include physical property parameters, environmental parameters, dynamic operation parameters, and model parameters; physical property parameters can include geometric parameters, such as joint link dimensions, joint spacing, and link shapes; material parameters, such as material type, material density, and elastic modulus; mass and inertia parameters, such as the mass of each rod, the position of the center of mass, and the inertia tensor; environmental parameters can include the external environment, such as temperature and humidity; interaction constraints, such as ground friction coefficient and contact stiffness; dynamic operation parameters include energy consumption and vibration frequency; model parameters include the direction of gravity, time step, controller type, error range between the physical twin model and the entity, and data synchronization delay. Then, three-dimensional modeling is carried out according to the same scale through CAD auxiliary software such as Solidworks and Proe based on the collected modeling parameters, and texture carving and rendering processing are performed in combination with the actual industrial robot material, color, and environmental factors to ensure a high degree of similarity of the industrial robot virtual physical twin model.

[0033] S120. Collect the real-time operation data corresponding to the target industrial robot, and determine the fault diagnosis result corresponding to the target industrial robot according to the real-time operation data and the fault diagnosis model.

[0034] Among them, the fault diagnosis model is a neural network model pre-trained, and the fault diagnosis model is a convolutional neural network, including four convolutional layers, four pooling layers, and a fully connected layer. The real-time operation data can be understood as the data generated by the target industrial robot during operation. The fault diagnosis result can be the current fault state of the target industrial robot, which can include whether a fault occurs, and in the case of a fault, the fault diagnosis result can include the fault type.

[0035] Specifically, sensors can be installed at key parts such as robot joints, motors, and controllers to collect the real-time operation data associated with the target industrial robot. After preprocessing the data, the preprocessed real-time data is input into the fault diagnosis model, and the model judges the current state and fault type of the robot according to the input data, and outputs the fault diagnosis result, including the fault type, location, severity, and recommended handling measures.

[0036] On the basis of the above technical solution, collecting the real-time operation data corresponding to the target industrial machine includes: collecting the vibration acceleration data corresponding to the target industrial robot through a three-axis vibration sensor arranged at the vibration data collection position of the target industrial robot.

[0037] Among them, the vibration data collection position can be the position for deploying the vibration sensor, and the vibration data collection position includes at least one of the RV reducer, the joint bearing, and the surface of the robotic arm rod. The vibration acceleration data is an index for representing the vibration intensity. The vibration acceleration is proportional to the vibration amplitude and proportional to the square of the vibration frequency.

[0038] Specifically, by deploying a three-axis vibration sensor at the key parts of the target industrial robot, the vibration acceleration data is collected in real time to monitor the running state of the robot and diagnose potential faults, such as wear of the transmission system, mechanical looseness, unbalanced load, etc. For example, the three-axis vibration sensor can be powered by DC 9V, the measuring range is ±16g for vibration acceleration, the response frequency is 6kHz, the resolution is 0.488mg / LSB, the maximum sampling frequency is 26.667kHz, the vibration measurement directions are the X, Y, and Z axes respectively, the size is 26mm * 68mm * 24mm (barrel diameter * height * opposite side), and the installation method is magnetic attraction.

[0039] On the basis of the above technical solution, collecting the real-time operation data corresponding to the target industrial machine includes: collecting the temperature data corresponding to the target industrial robot through a temperature sensor arranged at the temperature data collection position of the target industrial robot.

[0040] Among them, the temperature data collection position can be the position for deploying the temperature sensor. The temperature data collection position includes at least one of the outside of the RV reducer, the motor winding housing, and the encoder top cover.

[0041] Specifically, the temperature acquisition device selects the Keysight data acquisition system DAQ970A to directly measure temperature. This instrument integrates precise measurement functions and flexible signal connection functions. There are three module slots at the back of the instrument, which can perform any combination of data acquisition or switching modules, and has USB data recording and data acquisition functions. The three module slots of the Keysight data acquisition system DAQ970A cooperate with three acquisition card modules. Each acquisition card module can provide 60 channels (120 single-ended channels). The reading rate in a single channel exceeds 5,000 readings per second, and the scanning rate is as high as 450 channels per second. When in use, connect the thermocouple wire to the acquisition card module. First, pull out the acquisition card module from the slot, use a flat-head screwdriver to push the tongue on the acquisition card module forward, lift the outer shell to separate it from the module, then pass the thermocouple wire through the cable protective sleeve, connect it to the terminal of the acquisition card module and fix it, reattach the outer shell of the acquisition card module, and install the acquisition card module into the temperature measurement device in the correct direction.

[0042] On the basis of the above technical solution, before collecting the real-time operation data corresponding to the target industrial machine, it further includes: obtaining the historical fault data corresponding to the target industrial robot, and performing fault simulation according to the digital twin model to generate standard fault data corresponding to the fault type; determining a training data set based on the historical fault data and the standard fault data, and training the to-be-trained fault diagnosis model according to the training data set to obtain the fault diagnosis model.

[0043] Among them, the historical fault data can be the fault data generated by the target industrial robot during historical operation. It should be noted that the historical fault data can not be limited to the target industrial robot, that is, the fault data of other robots of the same type as the target industrial robot can be obtained, thereby ensuring the quantity of sample data. The fault type can be understood as the preset fault type, which can include RV reducer faults such as broken teeth, pitting corrosion of tooth surface, tooth surface wear, overheating, bearing damage, loosening of components; joint motor faults such as damage to the driver, damage to the rotor or stator, damage to the winding group; damage to components on the encoder circuit board; loosening and breakage of the transmission belt; spring breakage of the balance cylinder, hydraulic oil / gas leakage, seal failure; wear of rolling elements of the joint bearing, damage to the support body; deformation or distortion of the mechanical arm rod structure.

[0044] Specifically, historical failure data associated with the target industrial robot is determined through robot maintenance records, sensor historical data, and maintenance logs. It can also be retrieved from the cloud server for the historical failure data corresponding to the target industrial robot. After obtaining the historical failure data, duplicate, missing, or noisy data is removed, and the data format is unified. Fault type tags are added to the data based on the maintenance records, such as gear wear, bearing failure, etc. For example, data with an increased multiple frequency of the gear meshing frequency in the vibration spectrum is labeled as "gear wear". Additionally, corresponding fault data is generated according to the simulated typical faults in the digital twin model, such as gear crack propagation and bearing raceway spalling. For example, by modifying the gear stiffness parameter, the change in meshing stiffness caused by gear wear is simulated to generate vibration time-domain and frequency-domain data. Subsequently, the historical failure data and the simulated fault data are mixed in a certain proportion, such as 7:3, to balance the weights of the real data and the simulated data. Operations such as adding noise and time-domain stretching are performed on the simulated fault data to improve the robustness of the model. Finally, after dividing the final sample data, the fault diagnosis model to be trained is trained to obtain the fault diagnosis model.

[0045] Based on the above technical solution, determining the fault diagnosis result corresponding to the target industrial robot according to the real-time operation data and the fault diagnosis model includes: preprocessing the real-time operation data to obtain the preprocessed real-time operation data, and extracting the time-domain features corresponding to the preprocessed real-time operation data; segmenting the time-domain features based on the sliding window processing parameters, and inputting the segmented time-domain features into the fault diagnosis model, and taking the output of the fault diagnosis model as the fault diagnosis result.

[0046] Among them, the time-domain features can be the feature data obtained by performing time-domain analysis on the preprocessed real-time operation data. The sliding window processing parameters can be the parameters preset for performing sliding window processing, which can include the window length and the step size, etc.

[0047] Specifically, data during the operation of the industrial robot is collected, including state-related data such as vibration signals, current, speed, temperature, etc. The collected data is preprocessed, which can be to remove irrelevant or redundant information such as noise and misleading features, and then time-domain feature extraction such as RMS, MAV, KUR, etc. is performed. After that, a sliding window operation is performed on the extracted features, and the extracted data is combined with the fault labels of various types of the industrial robot and input into the fault diagnosis model based on machine learning or deep learning, and finally the fault diagnosis result of the industrial robot is obtained. It should be noted that the simulated data is the twin data simulated by each component of the industrial robot. The generated twin data is compared with the actual collected data, and the fault diagnosis model corresponding to the digital twin model is used for training.

[0048] Exemplarily, the sliding window operation is used to divide the continuous sensor time series data into data segments of fixed length so as to be input into the fault diagnosis model for feature extraction and pattern recognition. The window length represents the number of continuous data points contained in a single window. For example, the window length is 30, which means that each window covers data at 30 sampling moments. The step size represents the number of data points each time the window slides. For example, the step size is 16, which means that there is an overlap of 14 data points between adjacent windows. The input dimension represents the single window data converted into the model input format after preprocessing. For example, from the continuously collected sensor data, the data segments are intercepted according to the window length and step size. If the original data sequence is X = [x1, x2, ..., xN], the subsequence generated by the sliding window is W1 = [x1, x2, ..., x30], W2 = [x17, x18, ..., x46]. It should be noted that if the length of the end of the data is less than the window length, zero padding or repeated end value padding is used to ensure that the timestamp of each window is aligned with the physical event.

[0049] On the basis of the above technical solution, after collecting the real-time operation data corresponding to the target industrial machine, it also includes: sending the real-time operation data to a cloud server for storage; when the cloud server receives a data access request, it determines the user authority information corresponding to the data access request, and determines the feedback information corresponding to the data access request based on the user authority information.

[0050] The cloud server may be a remote server for storing robot operation data. The data access request may be a request message sent by a maintenance person when accessing the cloud server. The user rights information may be understood as the data viewing rights corresponding to the maintenance person. The feedback information may be information returned to the user, and may include device operation data corresponding to the data access request, or prompt information for prompting the user that the rights are insufficient.

[0051] Specifically, the real-time operating status raw data and operating history raw data collected by the key components of the industrial robot are saved in the cloud server. The cloud server has various data access management and access permission functions. When there is a need to use the relevant data of the industrial robot, it is necessary to apply for permission and obtain approval before the data can be accessed, viewed or downloaded; data storage is the effective storage of the operating status data read by the robot server of different key components of the industrial robot in industrial scenarios and the operating status-related data collected by the peripheral data acquisition sensors; data maintenance is the scientific naming, classification and other maintenance operations of the stored data by the management personnel, and data clouding is to upload the maintained data to the cloud, so that people in need can access and download it through the Internet.

[0052] Exemplarily, the real-time operation data is transmitted to the cloud server for storage via a communication protocol. Then, when a data access request is received, the cloud server performs the following operations: parse the user identity information in the data access request; determine the user's access privilege level according to the predefined privilege classification rules; generate feedback information based on the access privilege level, where the feedback information includes the data range, data format, or operation privilege allowed for access, and the level privileges can be set according to different roles, such as internal enterprise personnel, external enterprise personnel, scientific research personnel, and top management personnel, etc.; in terms of access management, according to the different backgrounds or requirements of the personnel accessing the data, the management personnel formulate corresponding role classifications according to the actual situation, so as to grant corresponding privileges; data storage is the storage of the massive data of industrial robots of different manufacturers, different models, different working conditions, different fault locations, and different fault types collected; data maintenance involves a lot of industrial robot data of different manufacturers, different models, different working conditions, different fault locations, and different fault types in the system, which belongs to the privacy of the enterprise, and it is necessary to ensure the accuracy of the information, otherwise it will affect the reliability, applicability, and authenticity of using the data for subsequent analysis and research. Therefore, it is necessary to regularly maintain the data to ensure the integrity of all data in the system.

[0053] S130. Generate a simulation display image corresponding to the target industrial robot based on the real-time operation data and the digital twin model, and display the simulation display image and the fault diagnosis result.

[0054] Among them, the simulation display image can be a virtual image used to display the current operation state of the industrial robot, and this virtual image corresponds to the real-time operation data of the industrial robot.

[0055] Specifically, based on the real-time operation data of the target industrial robot, generate a three-dimensional simulation display image corresponding to the target industrial robot through the digital twin model, dynamically map the operation state of the key components of the industrial robot through the simulation display image, and synchronously display the three-dimensional simulation display image and the fault diagnosis result in the visualization interface, where the fault location is highlighted in the image, and the fault details are associated and displayed. It can be in the three-dimensional simulation image, and the health state of the components is distinguished by color coding, green for normal, yellow for warning, and red for fault; and a floating window pops up after clicking on the fault mark, displaying the fault type, such as "tooth surface wear of RV reducer", the confidence level, such as 92%, and the maintenance suggestion, such as "stop immediately and replace the gear"; and generate a diagnostic report and support PDF / Excel format export, and the report content includes the fault time axis, impact analysis, and maintenance log.

[0056] The technical solution of the embodiment of the present invention is to obtain the modeling parameters of the target industrial robot, establish a digital twin model corresponding to the target industrial robot according to the modeling parameters, collect the real-time operation data corresponding to the target industrial robot, determine the fault diagnosis result corresponding to the target industrial robot according to the real-time operation data and the fault diagnosis model, generate a simulation display image corresponding to the target industrial robot according to the real-time operation data and the digital twin model, and display the simulation display image and the fault diagnosis result. Based on the above technical solution, the fault diagnosis result is determined by analyzing the operation data of the industrial robot through the fault diagnosis model, and the operation status of the industrial robot is simulated in combination with the digital twin model, thereby improving the accuracy of the fault diagnosis of the industrial robot and realizing the real-time detection of the operation status of the industrial robot, and improving the maintenance efficiency.

[0057] Embodiment 2

[0058] Figure 2 It is a schematic diagram of a fault diagnosis method for an industrial robot provided by an embodiment of the present invention. In this embodiment, the technical solution of the fault diagnosis method for an industrial robot is further optimized on the basis of the above technical solution. As Figure 2 shown, the method includes:

[0059] Data acquisition of key components of industrial robots and use of peripheral sensors: Specifically, the key components of industrial robots mainly include RV reducers, joint motors, encoders, drive belts, balance cylinders, joint bearings, and robotic arm members; among them, the data collected by RV reducers, joint motors, encoders, balance cylinders, joint bearings, etc. are mainly the current data, voltage data, joint angle data, angular velocity data, and acceleration information of each component during operation in the industrial environment. The drive belt mainly collects the tightness of the belt after long-term operation, generally using the force information and acoustic information collected by the sound sensor. Since the belt is inside the industrial robot, manual measurement by on-site staff is required during regular shutdown and maintenance of the industrial robot during measurement; in addition, the temperature sensors provided by the present invention are used to collect the temperature information of RV reducers, joint motors, encoders, etc. during equipment operation in the industrial environment. During use, the temperature measurement wire ends are arranged on the side wall of the RV reducer, the side wall of the motor winding, the bottom of the motor, and the outer wall of the encoder. The vibration sensor mainly collects the vibration information of RV reducers, joint bearings, robotic arm members, etc. from the outside. The optical sensor is used to collect the repeatability accuracy information of each joint position and the end of the robotic arm member after long-term operation. During use, the optical sensor is installed at a suitable position of the industrial robot in the industrial environment;

[0060] Usage of the digital twin model of industrial robots: Specifically, the digital twin model of industrial robots is a virtual physical twin model established in proportion using 3D modeling software. Its purpose is to monitor the operation process of industrial robots in industrial scenarios. In actual industrial production, due to the requirement of unmanned factory workshops, it is impossible to observe and monitor robots on-site. Through an equi-proportional and highly realistic digital twin model and efficient signal communication, real-time monitoring of industrial robots can be achieved, avoiding inspections and repairs after any problems occur. Moreover, the virtual physical twin model can simulate the fault simulation data of different key components, different working conditions, and different fault types of industrial robots, and conduct fault diagnosis research on industrial robots in combination with the fault diagnosis model.

[0061] Usage of the data processing and calculation part: Specifically, the edge computing module of the data processing and calculation part is used to deploy a fault diagnosis algorithm model based on machine learning or deep learning. The input data of the model is the preprocessed original operation history data or real-time operation status data or simulated fault data of industrial robots. Collect the data during the operation of industrial robots, including state-related data such as vibration signals, current, speed, and temperature. Preprocess the collected data, then extract time-domain features such as RMS, MAV, and KUR. After that, perform sliding window operation processing on the extracted features, and input the extracted data combined with various types of fault labels of industrial robots into the fault diagnosis model based on machine learning or deep learning to finally obtain the fault diagnosis result of industrial robots. The simulated data is the twin data simulated by each component of industrial robots. Compare the generated twin data with the actual collected data and use it to train the corresponding fault diagnosis model of the digital twin model. It can also be storing the real-time operation status original data and operation history original data of industrial robots in the cloud server, reading them from the fault management module when needed, preprocessing them and then inputting them into the fault diagnosis model. The fault diagnosis model of industrial robots outputs the results, and the diagnosis results are compared and the real-time status of industrial robots is monitored.

[0062] Industrial robot data management: Specifically, the original real-time operation status data and operation history original data collected by each key component of the industrial robot are saved to the cloud server. The cloud server has various data access management and access permission functions. When it is necessary to use the industrial robot-related data, a permission application needs to be made, and data access, viewing, or downloading can only be carried out after being approved by the management personnel; data storage is to effectively store the operation status data read by the robot servo in the industrial scenario and the operation status-related data collected by the peripheral data acquisition sensors of different key components of the industrial robot; data maintenance is for the management personnel to perform scientific naming, classification, and other maintenance operations on the stored data, and data cloud uploading is to upload the maintained data to the cloud to facilitate personnel in need to access and download it through the Internet.

[0063] It should be noted that data storage is the storage of the massive data of industrial robots from different manufacturers, different models, different working conditions, different fault locations, and different fault types collected; a MySQL database is deployed in the cloud to store industrial robot-related data. Through the read-write separation technology, master / slave (Master / Slave) replication technology, and distributed storage technology of MySQL, faults caused by massive data transmission are avoided, and a Redis database is used to store data caches. The data storage scheme process is as Figure 3 shown.

[0064] The Master realizes the function of writing data, the Slave realizes the function of reading data, the Master and Slave databases synchronize data using the binlog file, and the MyCAT distributed database intermediate component is referenced to enhance the scalability and flexibility of the system and improve the database performance; a caching mechanism is introduced to handle data that is stored for a long time and has a large number of read times, and to avoid the inconsistency between the MySQL data and the cached data.

[0065] When writing data, the data user sends a request to the write application server at the client. The application server responds to the request, writes the data into the MySQL database and synchronizes the data between the master and slave, and at the same time stores it in the Redis cache server.

[0066] During the data reading process, first query whether there is corresponding data in the Redis cache. If there is, change the expiration time of the cached data and send the data back to the data user access client. If there is no corresponding data, then query in the MySQL database. If the data is not found in the MySQL, send no data to the data user access client. If the data is found in the MySQL, synchronize the data to the Redis cache server and send the data to the data user access client.

[0067] Design a Web application based on the Django framework, call each module, manage and respond to the requests of data users accessing the client. Write the front-end page using HTML, CSS, and JavaScript, and use the Highcharts tool to implement the display of various charts. Users access the system through a browser in the form of a Web interface to achieve the interaction between users and the system, and complete the coding of all functions of the system; Optionally, design the login page, data display page, and data information visualization page of the Web front-end to facilitate data users to view, read, and download; The data maintenance involves the industrial robot data of many different manufacturers' robots, different models, different working conditions, different fault locations, and different fault types in the system, which belongs to the privacy of the enterprise, and it is necessary to ensure the accuracy of the information. Otherwise, it will affect the reliability, applicability, and authenticity of subsequent data analysis and research. Therefore, it is necessary to regularly maintain the data to ensure the integrity of all data in the system.

[0068] In the industrial robot part of the present invention, it includes key components of the industrial robot such as RV reducers, joint motors, encoders, transmission belts, balance cylinders, joint bearings, and peripheral data acquisition sensors, and collects the original real-time operating state data of each component in the industrial environment; The industrial robot digital twin part is used to virtually map a high-fidelity industrial robot system and simulate and generate different types of fault data of each component of the industrial robot in the industrial environment; By deploying a fault diagnosis model based on machine learning or deep learning algorithms, through the preprocessed real-time operating state data or historical data, diagnose different components, different working conditions, and different fault types of the industrial robot, and manage, store, and maintain the original real-time operating state data and operating history original data generated by the industrial robot based on the cloud server. Analyze the operating data of the industrial robot through the fault diagnosis model to determine the fault diagnosis result, and combine the digital twin model to simulate the operating condition of the industrial robot, thereby improving the accuracy of the fault diagnosis of the industrial robot and realizing the real-time detection of the operating condition of the industrial robot, and improving the maintenance efficiency.

[0069] Embodiment III

[0070] Figure 4 It is a structural schematic diagram of a fault diagnosis device for an industrial robot provided by an embodiment of the present invention. As Figure 4 shown, the device includes: a twin model construction module 410, a diagnosis result determination module 420, and a real-time status monitoring module 430.

[0071] The twin model construction module 410 is used to obtain the physical attributes of the target industrial robot and establish a digital twin model corresponding to the target industrial robot according to the physical attributes;

[0072] A diagnostic result determination module 420, configured to collect real-time operation data corresponding to the target industrial robot, and determine a fault diagnosis result corresponding to the target industrial robot according to the real-time operation data and a fault diagnosis model, wherein the fault diagnosis model is a neural network model pre-trained;

[0073] A real-time status monitoring module 430, configured to generate a simulated display image corresponding to the target industrial robot according to the real-time operation data and the digital twin model, and display the simulated display image and the fault diagnosis result.

[0074] Based on the above technical solution, the diagnostic result determination module is configured to collect vibration acceleration data corresponding to the target industrial robot through a triaxial vibration sensor disposed at a vibration data acquisition position of the target industrial robot, wherein the vibration data acquisition position includes at least one of an RV reducer, a joint bearing, and a surface of a robotic arm rod.

[0075] Based on the above technical solution, the diagnostic result determination module is configured to collect temperature data corresponding to the target industrial robot through a temperature sensor disposed at a temperature data acquisition position of the target industrial robot, wherein the temperature data acquisition position includes at least one of the outside of the RV reducer, the motor winding housing, and the encoder top cover.

[0076] Based on the above technical solution, the diagnostic result determination module is configured to obtain historical fault data corresponding to the target industrial robot, perform fault simulation according to the digital twin model, and generate standard fault data corresponding to the fault type; determine a training data set based on the historical fault data and the standard fault data, and train the to-be-trained fault diagnosis model according to the training data set to obtain the fault diagnosis model.

[0077] Based on the above technical solution, the diagnostic result determination module is configured to preprocess the real-time operation data to obtain preprocessed real-time operation data, extract time-domain features corresponding to the preprocessed real-time operation data; segment the time-domain features based on sliding window processing parameters, and input the segmented time-domain features into the fault diagnosis model, and use the output of the fault diagnosis model as the fault diagnosis result.

[0078] Based on the above technical solution, the device further includes a data management module, which is configured to, after collecting the real-time operation data corresponding to the target industrial machine, send the real-time operation data to a cloud server for storage; when receiving a data access request, the cloud server determines the user permission information corresponding to the data access request, and determines the feedback information corresponding to the data access request according to the user permission information.

[0079] Based on the above technical solution, the fault diagnosis model is a convolutional neural network; the fault diagnosis model includes four convolutional layers, four pooling layers and a fully connected layer.

[0080] The technical solution of the embodiment of the present invention obtains the modeling parameters of the target industrial robot, establishes a digital twin model corresponding to the target industrial robot according to the modeling parameters, collects the real-time operation data corresponding to the target industrial machine, and determines the fault diagnosis result corresponding to the target industrial robot according to the real-time operation data and the fault diagnosis model, generates a simulation display image corresponding to the target industrial robot according to the real-time operation data and the digital twin model, and displays the simulation display image and the fault diagnosis result. Based on the above technical solution, the fault diagnosis result is determined by analyzing the operation data of the industrial robot through the fault diagnosis model, and the operation status of the industrial robot is simulated in combination with the digital twin model, thereby improving the accuracy of the fault diagnosis of the industrial robot and realizing the real-time detection of the operation status of the industrial robot, and improving the maintenance efficiency.

[0081] The fault diagnosis device of the industrial robot provided by the embodiment of the present invention can execute the fault diagnosis method of the industrial robot provided by any embodiment of the present invention, and has the corresponding functional modules and beneficial effects for executing the method.

[0082] Embodiment IV

[0083] Figure 5 FIG. shows a schematic structural diagram of an electronic device 10 that can be used to implement an embodiment of the present invention. The electronic device is intended to represent various forms of digital computers, such as, a laptop computer, a desktop computer, a workbench, a personal digital assistant, a server, a blade server, a mainframe computer, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as, a personal digital processor, a cellular phone, a smart phone, a wearable device (such as a helmet, glasses, a watch, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described herein and / or claimed.

[0084] As Figure 5As shown, the electronic device 10 includes at least one processor 11 and a memory communicatively connected to the at least one processor 11, such as read-only memory (ROM) 12, random access memory (RAM) 13, etc. Among them, the memory stores a computer program executable by the at least one processor. The processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 into the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, ROM 12, and RAM 13 are connected to each other via a bus 14. The input / output (I / O) interface 15 is also connected to the bus 14.

[0085] Multiple components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a magnetic disk, an optical disc, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0086] The processor 11 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the processor 11 include but are not limited to a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as the fault diagnosis method of an industrial robot.

[0087] In some embodiments, the fault diagnosis method of an industrial robot can be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as the storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed onto the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the fault diagnosis method of the industrial robot described above can be executed. Alternatively, in other embodiments, the processor 11 can be configured to execute the fault diagnosis method of the industrial robot by any other appropriate means (such as by means of firmware).

[0088] The various embodiments of the systems and techniques described above in this specification can be implemented in digital electronic circuitry, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on a chip (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include: being implemented in one or more computer programs that are executable and / or interpretable on a programmable system including at least one programmable processor, which may be a special-purpose or general-purpose programmable processor that receives data and instructions from, and transmits data and instructions to, a storage system, at least one input device, and at least one output device.

[0089] The computer program for implementing the methods of the present invention can be written in any combination of one or more programming languages. These computer programs can be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the computer programs, when executed by the processor, cause the functions / operations specified in the flowchart and / or block diagram to be implemented. The computer program can be executed entirely on the machine, partly on the machine, as a stand-alone software package partly on the machine and partly on a remote machine or entirely on the remote machine or server.

[0090] In the context of the present invention, a computer-readable storage medium can be a tangible medium that can contain or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. The computer-readable storage medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. Alternatively, the computer-readable storage medium can be a machine-readable signal medium. More specific examples of the machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0091] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or a trackball) through which the user can provide input to the electronic device. Other kinds of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, speech input, or tactile input).

[0092] The systems and techniques described herein can be implemented in a computing system including backend components (e.g., as a data server), or a computing system including middleware components (e.g., an application server), or a computing system including frontend components (e.g., a user computer having a graphical user interface or a web browser through which the user can interact with an implementation of the systems and techniques described herein), or a computing system including any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected to each other by digital data communication in any form or medium (e.g., a communication network). Examples of communication networks include: local area network (LAN), wide area network (WAN), blockchain network, and the Internet.

[0093] A computing system can include a client and a server. The client and the server are generally far from each other and typically interact through a communication network. The client-server relationship is created by computer programs running on respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or a cloud host, which is a host product in the cloud computing service system, and solves the defects of difficult management and weak business scalability existing in traditional physical hosts and VPS services.

[0094] It should be understood that various forms of the processes shown above can be used, with steps reordered, added, or deleted. For example, the steps recited in the present invention can be executed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution of the present invention can be achieved, and no limitation is imposed herein.

[0095] The above specific embodiments do not constitute a limitation on the protection scope of the present invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A fault diagnosis method for an industrial robot, characterized in that, Including: Obtain the modeling parameters of the target industrial robot, and establish a digital twin model corresponding to the target industrial robot according to the modeling parameters; Collect the real-time operation data corresponding to the target industrial robot, and determine the fault diagnosis result corresponding to the target industrial robot according to the real-time operation data and the fault diagnosis model, wherein the fault diagnosis model is a neural network model pre-trained; Generate a simulation display image corresponding to the target industrial robot according to the real-time operation data and the digital twin model, and display the simulation display image and the fault diagnosis result.

2. The method according to claim 1, wherein The collecting the real-time operation data corresponding to the target industrial robot includes: Collect the vibration acceleration data corresponding to the target industrial robot through a three-axis vibration sensor arranged at the vibration data collection position of the target industrial robot, wherein the vibration data collection position includes at least one of the RV reducer, the joint bearing, and the surface of the robotic arm rod.

3. The method according to claim 1, wherein The collecting the real-time operation data corresponding to the target industrial robot includes: Collect the temperature data corresponding to the target industrial robot through a temperature sensor arranged at the temperature data collection position of the target industrial robot, wherein the temperature data collection position includes at least one of the outside of the RV reducer, the motor winding housing, and the encoder top cover.

4. The method according to claim 1, characterized in that Before the collecting the real-time operation data corresponding to the target industrial robot, it further includes: Obtain the historical fault data corresponding to the target industrial robot, and perform fault simulation according to the digital twin model to generate standard fault data corresponding to the fault type; Determine a training data set based on the historical fault data and the standard fault data, and train the to-be-trained fault diagnosis model according to the training data set to obtain the fault diagnosis model.

5. The method according to claim 1, wherein The determining the fault diagnosis result corresponding to the target industrial robot according to the real-time operation data and the fault diagnosis model includes: Preprocess the real-time operation data to obtain preprocessed real-time operation data, and extract the time-domain features corresponding to the preprocessed real-time operation data; Segment the time-domain features based on the sliding window processing parameters, input the segmented time-domain features into the fault diagnosis model, and use the output of the fault diagnosis model as the fault diagnosis result.

6. The method according to claim 1, wherein After the collecting the real-time operation data corresponding to the target industrial robot, it further includes: Send the real-time operation data to the cloud server for storage; The cloud server, when receiving a data access request, determines the user permission information corresponding to the data access request, and determines the feedback information corresponding to the data access request according to the user permission information.

7. The method according to claim 1, wherein The fault diagnosis model is a convolutional neural network; the fault diagnosis model includes four convolutional layers, four pooling layers, and a fully connected layer.

8. A fault diagnosis device for an industrial robot, characterized in that, Including: A twin model construction module, configured to obtain the modeling parameters of the target industrial robot, and establish a digital twin model corresponding to the target industrial robot according to the modeling parameters; A diagnosis result determination module, configured to collect real-time operation data corresponding to the target industrial machine, and determine a fault diagnosis result corresponding to the target industrial robot according to the real-time operation data and a fault diagnosis model, wherein the fault diagnosis model is a neural network model pre-trained; A real-time status monitoring module, configured to generate a simulated display image corresponding to the target industrial robot according to the real-time operation data and the digital twin model, and display the simulated display image and the fault diagnosis result.

9. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the fault diagnosis method of the industrial robot according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions, and the computer instructions are used to implement the fault diagnosis method of the industrial robot according to any one of claims 1-7 when executed by a processor.

Images