Industrial robot operation and maintenance management method, device, computer equipment and storage medium

Abstract

The present application relates to an industrial robot operation and maintenance management method, apparatus, computer equipment, storage medium, and computer program product. The method comprises: obtaining a target fault analysis model corresponding to the industrial robot; determining the failure probability of each component in the geometric model corresponding to the industrial robot using the target fault analysis model; determining the failure risk level of the industrial robot corresponding to the geometric model based on the failure probability of each component in the geometric model; and selecting an industrial robot that meets the failure risk level criteria as a target robot; collecting multiple target state data corresponding to each component in the geometric model corresponding to the target robot, and obtaining a health assessment result for the target robot based on the multiple target state data; and predicting the operation and maintenance cycle and service life of the target robot based on the health assessment result, thereby adjusting the production schedule of the target robot. This method can improve the efficiency of industrial robot operation and maintenance management.

Description

Technical Field

[0001] The present application relates to the field of intelligent manufacturing technology, and in particular to an industrial robot operation and maintenance management method, apparatus, computer equipment, storage medium, and computer program product. Background Art

[0002] With the development of intelligent manufacturing, industrial robots are gradually gaining widespread application. Industrial robots are characterized by complex structures, high precision, and high reliability. However, they are subject to issues such as precision degradation and component failure. To ensure the normal operation of industrial robots, timely operation and maintenance are necessary.

[0003] The operation and maintenance model in traditional technology generally installs sensors on industrial robots, monitors relevant data of industrial robots, and thus predicts the failure of industrial robots.

[0004] However, traditional technologies require a lot of resources and time to process the relevant data monitored by sensors, which affects the operation and maintenance efficiency of industrial robots. Summary of the Invention

[0005] Based on this, it is necessary to provide an industrial robot operation and maintenance management method, device, computer equipment, computer-readable storage medium and computer program product to address the above technical problems.

[0006] In a first aspect, the present application provides an industrial robot operation and maintenance management method. The method comprises:

[0007] Obtain a target fault analysis model corresponding to the industrial robot, and determine the failure probability of each component in the geometric model corresponding to the industrial robot through the target fault analysis model;

[0008] Based on the failure probability of each component in the geometric model, the failure risk level of the industrial robot corresponding to the geometric model is determined, and the industrial robot that meets the failure risk level conditions is selected as the target robot;

[0009] Collecting multiple target state data corresponding to each component in the geometric model corresponding to the target robot, and obtaining a health assessment result of the target robot based on the multiple target state data;

[0010] Based on the health assessment results, the operation and maintenance cycle and service life of the target robot are predicted, and based on the operation and maintenance cycle and service life of the target robot, the production schedule of the target robot is adjusted.

[0011] In one embodiment, obtaining a target fault analysis model corresponding to the industrial robot includes:

[0012] Constructing a geometric model corresponding to the industrial robot in a virtual space, and obtaining a plurality of historical state data corresponding to each component in the geometric model and a plurality of eigenvalues ​​corresponding to each historical state data;

[0013] Based on the multiple characteristic values ​​corresponding to each historical state data, a characteristic data set is constructed, where the characteristic data set includes the multiple characteristic values ​​corresponding to each historical state data;

[0014] Determine multiple feature values ​​corresponding to each component in the feature data set, and establish an association relationship between multiple feature values ​​corresponding to the same component and the corresponding component;

[0015] A data model is constructed based on the feature data set and the association relationship, and based on the data model, a target fault analysis model is obtained.

[0016] In one embodiment, obtaining a target fault analysis model based on the data model includes:

[0017] Divide the feature data set in the data model into a training set and a test set, obtain multiple initial fault analysis models based on the training set, and determine the failure probability prediction result of each initial fault analysis model for each component based on the association relationship in the data model;

[0018] Based on the test set in the data model, verify the failure probability prediction results of each component by each initial fault analysis model to obtain the accuracy of each initial fault analysis model;

[0019] Selecting an initial fault analysis model that meets the accuracy condition as an alternative fault analysis model, where the number of the alternative fault analysis models is multiple;

[0020] When the industrial robot corresponding to the geometric model fails, the state data corresponding to each component in the geometric model at the time of the failure is collected, and the alternative fault analysis models are verified based on the state data at the time of the failure to determine the accuracy of each alternative fault analysis model;

[0021] The alternative fault analysis model that meets the accuracy conditions is selected as the target fault analysis model.

[0022] In one embodiment, there are multiple target fault analysis models, and determining the failure probability of each component in the geometric model corresponding to the industrial robot through the target fault analysis model includes:

[0023] Collect the real-time status data of each component of the industrial robot corresponding to the geometric model, and obtain the characteristic value corresponding to each real-time status data;

[0024] Inputting the characteristic value corresponding to each real-time status data into multiple target fault analysis models to obtain the failure probability prediction result of each target fault analysis model for each component in the geometric model;

[0025] Based on the failure probability prediction results, the failure probability of each component in the geometric model is determined.

[0026] In one embodiment, collecting a plurality of target state data corresponding to each component in a geometric model corresponding to a target robot and obtaining a health assessment result of the target robot based on the plurality of target state data includes:

[0027] Collecting multiple historical state data corresponding to each component in the geometric model multiple times, and obtaining an eigenvalue mean and an eigenvalue standard deviation corresponding to each eigenvalue based on the historical state data collected multiple times;

[0028] Based on the eigenvalue mean and eigenvalue standard deviation corresponding to each eigenvalue, a health assessment model is constructed, wherein the health assessment model is configured with a health assessment method;

[0029] Collect multiple target state data corresponding to each component in the geometric model corresponding to the target robot, and obtain the health assessment results of the target robot based on the multiple target state data, including:

[0030] Obtaining multiple eigenvalues ​​corresponding to each target state data, and inputting the multiple eigenvalues ​​corresponding to each target state data into a health assessment model to obtain a difference between each eigenvalue and a corresponding eigenvalue mean;

[0031] Compare the difference between each eigenvalue and the corresponding eigenvalue mean with the corresponding eigenvalue standard deviation to obtain a comparison result;

[0032] Based on the comparison results and the health assessment method, the health assessment results of the target robot are obtained.

[0033] In one embodiment, after collecting a plurality of target state data corresponding to each component in the geometric model corresponding to the target robot and obtaining a health assessment result of the target robot based on the plurality of target state data, the method further includes:

[0034] When the health assessment result indicates that the target robot is in a faulty state, the target robot is controlled to stop the production task and run the target distance without load;

[0035] State data of the target robot during no-load operation is collected, and based on the state data of the target robot during no-load operation, a faulty component of the target robot is determined.

[0036] In a second aspect, the present application also provides an industrial robot operation and maintenance management device. The device includes:

[0037] The first processing module is used to obtain a target fault analysis model corresponding to the industrial robot, and determine the failure probability of each component in the geometric model corresponding to the industrial robot in the virtual space through the target fault analysis model;

[0038] a second processing module, configured to determine a failure risk level of the industrial robot corresponding to the geometric model based on the failure probability of each component in the geometric model, and select an industrial robot that meets the failure risk level conditions as a target robot;

[0039] a third processing module, configured to collect a plurality of target state data corresponding to each component in a geometric model corresponding to the target robot, and obtain a health assessment result of the target robot based on the plurality of target state data;

[0040] The fourth processing module is used to predict the operation and maintenance cycle and service life of the target robot based on the health assessment result, and adjust the production schedule of the target robot based on the operation and maintenance cycle and service life of the target robot.

[0041] In a third aspect, the present application further provides a computer device. The computer device includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are performed:

[0042] Obtain a target fault analysis model corresponding to the industrial robot, and determine the failure probability of each component in the geometric model corresponding to the industrial robot in the virtual space through the target fault analysis model;

[0043] Based on the failure probability of each component in the geometric model, the failure risk level of the industrial robot corresponding to the geometric model is determined, and the industrial robot that meets the failure risk level conditions is selected as the target robot;

[0044] Collecting multiple target state data corresponding to each component in the geometric model corresponding to the target robot, and obtaining a health assessment result of the target robot based on the multiple target state data;

[0045] Based on the health assessment results, the operation and maintenance cycle and service life of the target robot are predicted, and based on the operation and maintenance cycle and service life of the target robot, the production schedule of the target robot is adjusted.

[0046] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the following steps:

[0047] Obtain a target fault analysis model corresponding to the industrial robot, and determine the failure probability of each component in the geometric model corresponding to the industrial robot in the virtual space through the target fault analysis model;

[0048] Based on the failure probability of each component in the geometric model, the failure risk level of the industrial robot corresponding to the geometric model is determined, and the industrial robot that meets the failure risk level conditions is selected as the target robot;

[0049] Collecting multiple target state data corresponding to each component in the geometric model corresponding to the target robot, and obtaining a health assessment result of the target robot based on the multiple target state data;

[0050] Based on the health assessment results, the operation and maintenance cycle and service life of the target robot are predicted, and based on the operation and maintenance cycle and service life of the target robot, the production schedule of the target robot is adjusted.

[0051] In a fifth aspect, the present application further provides a computer program product. The computer program product includes a computer program that, when executed by a processor, implements the following steps:

[0052] Obtain a target fault analysis model corresponding to the industrial robot, and determine the failure probability of each component in the geometric model corresponding to the industrial robot in the virtual space through the target fault analysis model;

[0053] Based on the failure probability of each component in the geometric model, the failure risk level of the industrial robot corresponding to the geometric model is determined, and the industrial robot that meets the failure risk level conditions is selected as the target robot;

[0054] Collecting multiple target state data corresponding to each component in the geometric model corresponding to the target robot, and obtaining a health assessment result of the target robot based on the multiple target state data;

[0055] Based on the health assessment results, the operation and maintenance cycle and service life of the target robot are predicted, and based on the operation and maintenance cycle and service life of the target robot, the production schedule of the target robot is adjusted.

[0056] The above-mentioned industrial robot operation and maintenance management method, apparatus, computer equipment, storage medium, and computer program product first obtain a target fault analysis model corresponding to the industrial robot and, using the target fault analysis model, determine the failure probability of each component in the geometric model corresponding to the industrial robot. This allows the industrial robot to subsequently accurately and efficiently formulate a corresponding operation and maintenance plan based on the determined failure probability of each component, thereby improving the efficiency of operation and maintenance management. Based on the failure probability of each component in the geometric model, the industrial robot corresponding to the geometric model is then determined to have a failure risk level. An industrial robot that meets the failure risk level criteria is selected as a target robot. Multiple target state data corresponding to each component in the geometric model corresponding to the target robot are then collected. Based on the multiple target state data, a health assessment result of the target robot is obtained. Based on the health assessment result, the operation and maintenance cycle and service life of the target robot are predicted. This allows the target robot's operation and maintenance tasks to be efficiently scheduled. Based on the operation and maintenance cycle and service life of the target robot, the production schedule of the target robot is adjusted to avoid conflicts between the target robot's operation and maintenance tasks and production tasks, ensuring timely maintenance of the target robot. Furthermore, the actual usage time of the target robot is prevented from exceeding its service life, which could lead to failures. This improves the efficiency of industrial robot operation and maintenance management. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 This is a diagram of an application environment of an industrial robot operation and maintenance management method in one embodiment;

[0058] Figure 2 1 is a flow chart of an industrial robot operation and maintenance management method according to an embodiment;

[0059] Figure 3 A schematic diagram of a process for generating a target fault analysis model in one embodiment;

[0060] Figure 4 Schematic diagram of a process for target robot operation and maintenance management in one embodiment;

[0061] Figure 5 A schematic diagram of a process for constructing a digital twin model of an industrial robot in one embodiment;

[0062] Figure 6 This is a structural block diagram of an industrial robot operation and maintenance management device in one embodiment;

[0063] Figure 7 FIG. 1 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0064] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0065] The industrial robot operation and maintenance management method provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown, the industrial robot 102 communicates with the server 104 via a network. A data storage system can store data that the server 104 needs to process. The data storage system can be integrated on the server 104, or placed on a cloud or other network server. The server 104 obtains a target fault analysis model corresponding to the industrial robot 102 and, using the target fault analysis model, determines the failure probability of each component in the geometric model corresponding to the industrial robot 102. Based on the failure probability of each component in the geometric model, the server 104 determines the failure risk level of the industrial robot 102 corresponding to the geometric model. The server 104 selects the industrial robot 102 that meets the failure risk level criteria as the target robot. Multiple target state data corresponding to each component in the geometric model corresponding to the target robot are then collected. Based on the multiple target state data, a health assessment result for the target robot is obtained. Based on the health assessment result, the target robot's operation and maintenance cycle and service life are predicted. The target robot's production schedule is adjusted based on the target robot's operation and maintenance cycle and service life. The industrial robot 102 can be, but is not limited to, various industrial machine devices used in the field, possessing a certain degree of automation and capable of performing various industrial processing and manufacturing functions through its own power source and control capabilities. The server 104 may be implemented as an independent server or a server cluster consisting of multiple servers.

[0066] In one embodiment, Figure 2 As shown, a method for operation and maintenance management of industrial robots is provided, which is applied to Figure 1 Taking the server 104 in the example as an example, the following steps are included:

[0067] Step 202 : Obtain a target fault analysis model corresponding to the industrial robot, and determine the failure probability of each component in the geometric model corresponding to the industrial robot using the target fault analysis model.

[0068] Specifically, the geometric model of an industrial robot can be a visual 3D model, enabling visualization of twin objects, twin structures, and twin processes of the physical entity. It can also describe the industrial robot body and its accessory architecture using 3D visualization, defining the physical entity's geometric properties, motion properties, geometric shape, and mechanical structure. Its actions can represent the industrial robot's spatial position and posture. Furthermore, graphical visualization of geometric model-related data enables remote status monitoring, process parameter visualization, and historical status backtracking. Each component in the geometric model corresponds to a component in the physical entity of the industrial robot, which can specifically include a reducer, servo motor, generator, controller, etc. This does not limit the components of the industrial robot.

[0069] Specifically, the server can first obtain the target fault analysis model corresponding to each industrial robot, and then obtain the real-time data corresponding to each component in each industrial robot that can be used for fault analysis, and associate the real-time data of each industrial robot with the corresponding component in the corresponding geometric model, and then use the target fault analysis model to perform fault analysis on each component in each geometric model, so as to determine the failure probability of each component in the geometric model corresponding to the industrial robot.

[0070] Among them, the target fault analysis model corresponding to the same model of industrial robots is the same. The server can pre-train the target fault analysis model corresponding to each model of industrial robots based on the historical operation data of each industrial robot, and store the obtained target fault analysis model in the database, so that the corresponding target fault analysis model can be selected based on the model of the industrial robot in the future to improve the operation and maintenance management efficiency.

[0071] Step 204 : Based on the failure probability of each component in the geometric model, determine the failure risk level of the industrial robot corresponding to the geometric model, and select an industrial robot that meets the failure risk level conditions as a target robot.

[0072] The fault risk level can be used to classify industrial robots, allowing the server to provide monitoring methods that better meet the needs of industrial robots with different fault risk levels. This can improve fault monitoring efficiency and avoid wasting fault monitoring resources. Meeting the fault risk level conditions specifically indicates that the industrial robot's status is "high fault risk," meaning it is prone to failure. The server needs to monitor the industrial robot for faults in a timely manner, so that the server can issue an early warning before a fault occurs, or issue a timely fault alarm when a fault occurs, thereby reducing the impact of industrial robot failures on the production line.

[0073] Specifically, the server can determine the failure risk level of the geometric model based on the failure probability of each component in the geometric model and, in accordance with a failure risk level assessment method, further determine the failure risk level of the industrial robot corresponding to the geometric model and select the industrial robot that meets the failure risk level criteria as the target robot. Since different models of industrial robots have different core components, the failure risk level assessment methods for different models of industrial robots vary. The specific failure risk level assessment method can be configured based on the actual application scenario and is not limited here.

[0074] In a specific application, the server can first obtain the task data corresponding to the geometric model. This task data can specifically include the task action, load size, production cycle, and the qualified rate of the workpieces produced by the industrial robot corresponding to the geometric model. The production cycle is the time interval between the production of two workpieces by the industrial robot. The server can then assess the overall failure risk level of the geometric model based on the failure probability of each component in the geometric model and the task data corresponding to the geometric model, using a failure risk assessment method, to determine the failure risk level of the geometric model.

[0075] For example, for multiple industrial robots of the same model, since the failure probabilities of their components are similar, the server can assess the failure risk level of each robot based on the pass rate of the workpieces produced by each robot. The lower the pass rate of the workpieces produced by the robot, the higher the failure risk level of the robot.

[0076] In a specific application, after selecting an industrial robot that meets the fault risk level criteria as a target robot, the server can perform fault monitoring on the target robot using a monitoring method that meets the target robot's actual needs. For example, the server can perform online monitoring of the target robot to track its operating status in real time, and the server can also add operation and maintenance tasks for the target robot.

[0077] Step 206 : Collect multiple target state data corresponding to each component in the geometric model corresponding to the target robot, and obtain a health assessment result of the target robot based on the multiple target state data.

[0078] Each geometric model includes multiple components, and each component corresponds to multiple pieces of status data. The status data can specifically be signals storing data. The server can collect the status data of the industrial robot using external sensors installed on the industrial robot. The status data corresponding to the target robot is the target status data, and there are multiple target robots. For example, for a reducer in a component, the multiple pieces of status data corresponding to the reducer can specifically include the reducer bearing speed and reducer vibration amplitude. For example, for a servo motor in a component, the multiple pieces of status data corresponding to the servo motor can specifically include the servo motor speed, current, voltage, and power. For example, for a controller in a component, the multiple pieces of status data corresponding to the controller can specifically include the torque and speed controlled by the controller. For example, for a generator in a component, the multiple pieces of status data corresponding to the generator can specifically include the generator front axle temperature and rear axle temperature. Health assessment results can be categorized into healthy, degraded, and faulty states. Health assessment results can also be categorized based on actual application scenarios.

[0079] Specifically, during the online monitoring of multiple target robots, the server can collect multiple target state data corresponding to each component of each target robot in real time, and associate the multiple target state data of each target robot with the corresponding components in the corresponding geometric model. In other words, the server collects multiple target state data corresponding to each component in the geometric model corresponding to the target robot. Then, based on the multiple target state data corresponding to each component, the server performs a health assessment on each component in the geometric model corresponding to each target robot, obtaining a health assessment result for each target robot's corresponding geometric model, and thereby obtaining health assessment results for each of the multiple target robots.

[0080] Step 208 : Based on the health assessment result, predict the operation and maintenance cycle and service life of the target robot, and adjust the production schedule of the target robot based on the operation and maintenance cycle and service life of the target robot.

[0081] Specifically, the server can predict the operation and maintenance cycle and service life of the target robot based on the health assessment results, and input the operation and maintenance cycle and service life of the target robot as constraints into the production scheduling system, constrain the operating time of the target robot, and optimize the scheduling of the target robot's production tasks. It assists the production scheduling system in adjusting the operating time of the target robot to avoid time conflicts between the target robot's production tasks and operation and maintenance tasks, and promptly remove target robots with insufficient service life from the production line.

[0082] The production scheduling system may specifically be a system for managing the operation time of industrial robot production tasks, such as a PLM (Product Lifecycle Management) system.

[0083] In specific applications, for industrial robots of the same model, the server can use machine learning algorithms, based on extensive historical data, to pre-train a decision model that can be used to predict the degradation trend of each type of industrial robot. If the health assessment results of a target robot indicate that the target robot is in a degraded state, the server can retrieve the decision model corresponding to the target robot based on the model. Based on this decision model, the server can predict the degradation trend of the target robot, thereby predicting the target robot's operation and maintenance cycle and service life.

[0084] Among them, machine learning algorithms can learn patterns from complex data based on the accumulation of large amounts of data to predict future behavioral outcomes and trends. They can be used to accurately predict the service life and operation and maintenance cycle of target robots.

[0085] In the above-mentioned industrial robot operation and maintenance management method, a target fault analysis model corresponding to the industrial robot is first obtained. The target fault analysis model is used to determine the failure probability of each component in the geometric model corresponding to the industrial robot. This allows the operator to subsequently accurately and efficiently formulate a corresponding operation and maintenance plan based on the determined failure probability of each component, thereby improving the efficiency of operation and maintenance management. Based on the failure probability of each component in the geometric model, the fault risk level of the industrial robot corresponding to the geometric model is determined, and an industrial robot that meets the failure risk level criteria is selected as the target robot. Subsequently, multiple target state data corresponding to each component in the geometric model corresponding to the target robot are collected, and a health assessment result of the target robot is obtained based on the multiple target state data. Based on the health assessment result, the operation and maintenance cycle and service life of the target robot are predicted, so that the operation and maintenance tasks of the target robot can be efficiently scheduled. Based on the operation and maintenance cycle and service life of the target robot, the production schedule of the target robot is adjusted to avoid conflicts between the operation and maintenance tasks of the target robot and production tasks, ensuring timely maintenance of the target robot. Furthermore, the actual usage time of the target robot is prevented from exceeding its service life, which may lead to failures. This improves the operation and maintenance management efficiency of the industrial robot.

[0086] In one embodiment, obtaining a target fault analysis model corresponding to the industrial robot includes:

[0087] Constructing a geometric model corresponding to the industrial robot in a virtual space, and obtaining a plurality of historical state data corresponding to each component in the geometric model and a plurality of eigenvalues ​​corresponding to each historical state data;

[0088] Based on the multiple characteristic values ​​corresponding to each historical state data, a characteristic data set is constructed, where the characteristic data set includes the multiple characteristic values ​​corresponding to each historical state data;

[0089] Determine multiple feature values ​​corresponding to each component in the feature data set, and establish an association relationship between multiple feature values ​​corresponding to the same component and the corresponding component;

[0090] A data model is constructed based on the feature data set and the association relationship, and based on the data model, a target fault analysis model is obtained.

[0091] Among them, the geometric models corresponding to industrial robots of the same type are the same. The multiple eigenvalues ​​corresponding to each historical state data can specifically be the time domain characteristics, frequency domain characteristics, and spectral kurtosis characteristics of each historical data. The time domain characteristics can specifically include maximum value, minimum value, peak value, mean, variance, standard deviation, mean square value, root mean square value, kurtosis, skewness, form factor, crest factor, pulse factor, and margin factor. The frequency domain eigenvalues ​​can specifically include the center of gravity frequency, mean square frequency, root mean square frequency, frequency variance, and frequency standard deviation. The spectral kurtosis characteristics can specifically include the mean, standard deviation, skewness, and kurtosis of the spectral kurtosis.

[0092] Specifically, the server can first use a three-dimensional modeling tool to construct a geometric model corresponding to each model of industrial robot in the virtual space, and obtain multiple historical state data corresponding to each component in the geometric model, and then extract multiple eigenvalues ​​corresponding to each historical state data. Then, based on the multiple eigenvalues ​​corresponding to each historical state data, a feature data set is constructed, and the multiple eigenvalues ​​corresponding to each component in the feature data set are determined, and the association relationship between the multiple eigenvalues ​​corresponding to the same component and the corresponding component is constructed. Finally, based on the feature data set and the association relationship, a data model is constructed, and the feature data set in the data model is trained based on the machine learning algorithm to obtain the target fault analysis model.

[0093] Specifically, the feature data set may include multiple feature data subsets, and multiple feature values ​​corresponding to the same component belong to the same feature data subset. Establishing an association relationship between the multiple feature values ​​corresponding to the same component and the corresponding component, that is, establishing an association relationship between the feature data subset and the corresponding component.

[0094] In a specific application, when constructing a data model based on a feature dataset and its associated relationships, the server can use the multiple feature values ​​corresponding to each piece of historical state data in the feature dataset as the first layer of the data model. The server can then use the multiple feature values ​​corresponding to the same component and the associated relationships between the components as the second layer of the data model. Finally, the source file storing the multiple pieces of historical state data corresponding to each component can be used as the third layer of the data model. When a component in an industrial robot fails, the server can search the first layer of the data model for the feature values ​​corresponding to the failed component based on the associated relationships stored in the second layer of the data model to analyze the correlation between the feature values ​​and the component failure.

[0095] In a specific application, the server can also determine the collection time of multiple historical status data corresponding to each component based on the source files in the third layer of the data model, and establish a correlation between the collection time and the corresponding historical status data, that is, establish a correlation between the collection time and the corresponding feature value, and store this correlation between the collection time and the corresponding feature value in the second layer of the data model. When a component of the industrial robot fails at a certain point in time, the server can search the first layer of the data model for feature values ​​collected near that time point based on the correlation between the collection time and the corresponding feature value in the second layer of the data model, in order to analyze the correlation between the feature value and the component failure.

[0096] In this embodiment, a data model is constructed by historical status data, and characteristic values ​​related to faults can be quickly queried based on the data model, which can improve the efficiency of fault analysis. A target fault analysis model is obtained based on data model training, which can achieve accurate and efficient prediction of the failure probability of each component of the industrial robot.

[0097] In one embodiment, obtaining a target fault analysis model based on the data model includes:

[0098] Divide the feature data set in the data model into a training set and a test set, obtain multiple initial fault analysis models based on the training set, and determine the failure probability prediction result of each initial fault analysis model for each component based on the association relationship in the data model;

[0099] Based on the test set in the data model, verify the failure probability prediction results of each component by each initial fault analysis model to obtain the accuracy of each initial fault analysis model;

[0100] Selecting an initial fault analysis model that meets the accuracy condition as an alternative fault analysis model, where the number of the alternative fault analysis models is multiple;

[0101] When the industrial robot corresponding to the geometric model fails, the state data corresponding to each component in the geometric model at the time of the failure is collected, and the alternative fault analysis models are verified based on the state data at the time of the failure to determine the accuracy of each alternative fault analysis model;

[0102] The alternative fault analysis model that meets the accuracy conditions is selected as the target fault analysis model.

[0103] Satisfying the accuracy condition indicates that the model accuracy has reached an accuracy threshold. The accuracy threshold can be configured based on the actual application scenario. In this embodiment, the accuracy threshold can be 90%. When the number of model iterations reaches the iteration threshold, the server also determines that the model meets the accuracy condition. The iteration threshold can be configured based on the actual application scenario. In this embodiment, the iteration threshold can be 1000. The accuracy threshold and iteration threshold are not limited here.

[0104] Specifically, the server can collect multiple historical state data of various components of the industrial robot multiple times to obtain a feature data set, and construct a data model based on the feature data set. The feature data set in the data model is then divided into a training set and a test set. Then, each feature data subset in the training set is trained using a machine learning algorithm to obtain multiple initial fault analysis models. Based on the association between each feature data subset in the data model and the corresponding component, the server determines the failure probability prediction result of each initial fault analysis model for each component. The server can also verify the failure probability prediction result of each initial fault analysis model for each component based on the test set in the data model to obtain the accuracy of each initial fault analysis model. Then, the initial fault analysis model that meets the accuracy condition is selected as the alternative fault analysis model. When the industrial robot corresponding to the geometric model fails, the state data corresponding to each component in the geometric model at the time of the failure is collected, and the alternative fault analysis model is verified based on the state data at the time of the failure to determine the accuracy of each alternative fault analysis model. Finally, the alternative fault analysis model that meets the accuracy condition is selected as the target fault analysis model.

[0105] In a specific application, such as Figure 3 As shown, a method for generating a target fault analysis model is provided, which includes the following steps:

[0106] Step 302: Obtain historical status data.

[0107] Specifically, the server collects multiple historical status data corresponding to each component of the industrial robot multiple times and generates a feature data set.

[0108] Step 304: model training.

[0109] Specifically, the server divides the feature data set into a training set and a test set, first obtains multiple initial fault analysis models through the training set, and then uses the test set to verify the multiple initial fault analysis models to obtain the accuracy of each initial fault analysis model, and evaluates the accuracy of each initial fault analysis model, and selects the initial fault analysis model that meets the accuracy conditions as the recommended model, that is, the alternative fault analysis model, and the number of alternative fault analysis models is multiple.

[0110] Step 306: Acquire real-time status data at the time of the fault.

[0111] Specifically, when the industrial robot corresponding to the geometric model fails, the server collects real-time status data corresponding to each component in the geometric model when the failure occurs. When the next failure occurs, the real-time status data at the time of the current failure can be used as historical status data.

[0112] Step 308: Verify the alternative fault analysis model based on the real-time status data at the time of the fault.

[0113] Specifically, the server obtains the accuracy of each candidate fault analysis model, sorts the accuracy of each candidate fault analysis model, and selects multiple candidate fault analysis models that meet the accuracy conditions and have higher accuracy as target fault analysis models, wherein the number of target fault analysis models is multiple.

[0114] Step 310: Update the target fault analysis model to a recommended model.

[0115] In this embodiment, a target fault analysis model is obtained through multiple accuracy verifications, so that the prediction of fault analysis in the operation and maintenance management process is more accurate.

[0116] In one embodiment, there are multiple target fault analysis models, and determining the failure probability of each component in the geometric model corresponding to the industrial robot through the target fault analysis model includes:

[0117] Collect the real-time status data of each component of the industrial robot corresponding to the geometric model, and obtain the characteristic value corresponding to each real-time status data;

[0118] Inputting the characteristic value corresponding to each real-time status data into multiple target fault analysis models to obtain the failure probability prediction result of each target fault analysis model for each component in the geometric model;

[0119] Based on the failure probability prediction results, the failure probability of each component in the geometric model is determined.

[0120] Specifically, after obtaining multiple target fault analysis models based on training of historical status data, the server can collect real-time status data of each component of the industrial robot corresponding to the geometric model, and obtain the eigenvalue corresponding to each real-time status data, and then input the eigenvalue corresponding to each real-time status data into multiple target fault analysis models to obtain the failure probability prediction results of each component in the geometric model by each target fault analysis model, wherein each component corresponds to the failure probability prediction results of multiple target fault analysis models, and then based on the failure probability prediction results of multiple target fault analysis models, the failure probability of each component in the geometric model is determined.

[0121] In a specific application, the server can perform a weighted summation of the failure probability prediction results for each component to determine the failure probability of each component in the geometric model. The weight of each target fault analysis model for each component can be configured based on the actual application scenario. This embodiment does not limit the weight of each target fault analysis model.

[0122] In this embodiment, the failure probability of each component is determined by integrating the failure probability prediction results of multiple target failure probability analysis models, thereby achieving the effect of improving the accuracy of failure prediction.

[0123] In one embodiment, collecting a plurality of target state data corresponding to each component in a geometric model corresponding to a target robot and obtaining a health assessment result of the target robot based on the plurality of target state data includes:

[0124] Collecting multiple historical state data corresponding to each component in the geometric model multiple times, and obtaining an eigenvalue mean and an eigenvalue standard deviation corresponding to each eigenvalue based on the historical state data collected multiple times;

[0125] Based on the eigenvalue mean and eigenvalue standard deviation corresponding to each eigenvalue, a health assessment model is constructed, wherein the health assessment model is configured with a health assessment method;

[0126] Collect multiple target state data corresponding to each component in the geometric model corresponding to the target robot, and obtain the health assessment results of the target robot based on the multiple target state data, including:

[0127] Obtaining multiple eigenvalues ​​corresponding to each target state data, and inputting the multiple eigenvalues ​​corresponding to each target state data into a health assessment model to obtain a difference between each eigenvalue and a corresponding eigenvalue mean;

[0128] Compare the difference between each eigenvalue and the corresponding eigenvalue mean with the corresponding eigenvalue standard deviation to obtain a comparison result;

[0129] Based on the comparison results and the health assessment method, the health assessment results of the target robot are obtained.

[0130] Specifically, the server collects multiple historical status data corresponding to each component in the geometric model multiple times, and obtains multiple eigenvalues ​​corresponding to each historical status data based on the historical status data collected multiple times, and then obtains the eigenvalue mean and eigenvalue standard deviation corresponding to each eigenvalue, and then constructs a health assessment model configured with a health assessment method based on the eigenvalue mean and eigenvalue standard deviation corresponding to each eigenvalue, and then obtains multiple eigenvalues ​​corresponding to each target state data, and inputs the multiple eigenvalues ​​corresponding to each target state data into the health assessment model to obtain the difference between each eigenvalue and the corresponding eigenvalue mean, and then compares the difference between each eigenvalue and the corresponding eigenvalue mean with the corresponding eigenvalue standard deviation to obtain a comparison result. Finally, based on the comparison result and the health assessment method, the health assessment result of each component is obtained, and based on the health assessment result of each component in the target robot, the health assessment result of the target robot as a whole is obtained.

[0131] In a specific application, assuming that the server collects multiple historical state data corresponding to each component in the geometric model for a cumulative number of n times, the server can specifically obtain the mean of each eigenvalue through formula (1) and the standard deviation of each eigenvalue through formula (2):

[0132]

[0133]

[0134] Among them, μ is the mean of the eigenvalue, n is the number of acquisitions, i is the number of acquisitions, x i is the eigenvalue at the time of the i-th acquisition, and σ is the standard deviation of the eigenvalue.

[0135] Taking the servo motor in the geometric model as an example, the server can collect multiple historical state data of the servo motor multiple times. Taking the current data in the historical state data of the servo motor as an example, after each collection of the servo motor current data, the server will obtain the time domain characteristics, frequency domain characteristics, and spectral kurtosis characteristics corresponding to the servo motor current data, thereby obtaining multiple eigenvalues ​​corresponding to the servo motor current data. Taking the peak value in the time domain characteristics corresponding to the servo motor current data as an example, after multiple collections, the server can obtain multiple peak values ​​corresponding to the servo motor current data, and then can obtain the peak mean and peak standard deviation corresponding to the servo motor current data based on formula (1) and formula (2). Based on a similar method, the server can obtain the mean and standard deviation corresponding to each eigenvalue.

[0136] In specific applications, in the process of constructing a health assessment model configured with a health assessment method based on the eigenvalue mean and eigenvalue standard deviation corresponding to each eigenvalue, the server can specifically configure the health assessment method in the health assessment model by selecting o=μ as the health baseline, a=μ±σ as the degradation baseline, and b=μ±3σ as the fault baseline.

[0137] For example, when the absolute values ​​of the differences between multiple eigenvalues ​​corresponding to a component and their mean are no greater than σ, the component is between the healthy baseline and the degraded baseline and is in a healthy state. When the absolute values ​​of the differences between multiple eigenvalues ​​corresponding to a component and their mean are greater than σ but no greater than 3σ, the component is between the degraded baseline and the faulty baseline and is in a degraded state. When the absolute values ​​of the differences between multiple eigenvalues ​​corresponding to a component and their mean are greater than 3σ, the component is outside the faulty baseline and is in a faulty state.

[0138] In a specific application, in the process of obtaining the health assessment results of the target robot, the server can obtain multiple eigenvalues ​​corresponding to each target state data in the target robot in real time, and input the multiple eigenvalues ​​corresponding to each target state data into the health assessment model to obtain the difference between each eigenvalue and the corresponding eigenvalue mean, and compare the difference between each eigenvalue and the corresponding eigenvalue mean with the corresponding eigenvalue standard deviation to obtain a comparison result. Finally, based on the comparison result and the configured health assessment method, the health assessment results of each component are obtained, and the health assessment results of each component of the target robot are combined to obtain the health assessment result of the target robot. For example, for a target robot whose core component is a reducer, when the health assessment result of the reducer indicates that the reducer is in a degraded state, the server will determine that the target robot is also in a degraded state.

[0139] In this embodiment, by constructing a health assessment model through historical data, the purpose of efficiently and accurately determining the health assessment results of the target robot can be achieved, thereby improving the efficiency of operation and maintenance management.

[0140] In one embodiment, after collecting a plurality of target state data corresponding to each component in the geometric model corresponding to the target robot and obtaining a health assessment result of the target robot based on the plurality of target state data, the method further includes:

[0141] When the health assessment result indicates that the target robot is in a faulty state, the target robot is controlled to stop the production task and run the target distance without load;

[0142] State data of the target robot during no-load operation is collected, and based on the state data of the target robot during no-load operation, a faulty component of the target robot is determined.

[0143] Among them, the target distance can be selected according to actual needs.

[0144] Specifically, when the health assessment result indicates that the target robot is in a faulty state, the server will control the target robot to stop the production task, and control the target robot to unload the load (workpiece or cargo), and run the target distance without load. At the same time, the real-time status data generated by the target robot during the no-load operation is collected, and the faulty parts of the target robot are determined based on the real-time status data of the target robot during the no-load operation.

[0145] In a specific application, when the health assessment results indicate that the target robot is in a faulty state, the server will perform offline testing and diagnosis on the target robot, that is, testing and diagnosis off the production line. During the offline testing and diagnosis phase, the server will control the target robot to stop operating and conduct single-axis testing and manual troubleshooting. Specifically, the single-axis test can include: the server controls the target robot to run a target distance without load, from position A to position B, and controls the angle of each axis of the target robot to be no less than 90 degrees. The target robot's operating speed is set to 50% of the rated speed, and the operating time is controlled within 2 minutes. The server then obtains the real-time status data generated by the target robot during the single-axis test, performs fault analysis on the target robot based on this real-time status data, and sends the fault analysis results to the operation and maintenance personnel, who can then manually troubleshoot the target robot's fault point based on the fault analysis results.

[0146] In this embodiment, by performing offline testing and diagnosis on the target robot, the fault location, fault mode and fault cause can be determined more accurately, thereby improving the accuracy of fault location and fault assessment during the operation and maintenance management process.

[0147] In one embodiment, Figure 4 As shown, a target robot operation and maintenance management method that meets the target robot operation and maintenance requirements is provided, and the method mainly includes the following steps:

[0148] Step S402 : Select an industrial robot that meets the fault risk level conditions as a target robot, and issue an early warning for the fault risk level of the target robot.

[0149] Step S404: remotely monitor the target robot, collect real-time status data of the target robot based on the remote monitoring, and obtain a health assessment result of the target robot based on the real-time status data of the target robot.

[0150] Step S406 : When the health assessment result of the target robot indicates that the target robot is in a fault state, an offline test diagnosis is performed on the target robot to improve the accuracy of the fault diagnosis.

[0151] In this embodiment, an operation and maintenance management solution that meets the corresponding needs can be provided for operation and maintenance needs at different levels, thereby improving the operation and maintenance management efficiency.

[0152] In one embodiment, Figure 5 As shown, a method for constructing a digital twin model of an industrial robot for operation and maintenance management of an industrial robot is provided. The method mainly includes the following steps:

[0153] Step S502: constructing a geometric model.

[0154] Step S504: construct a data model.

[0155] Specifically, the server obtains state data of the industrial robot, and constructs a data model corresponding to the geometric model of the industrial robot based on the obtained state data.

[0156] Step S506: construct a fault analysis model.

[0157] Specifically, the server trains the feature data set in the data model to obtain a fault analysis model, and determines the target robot that meets the fault risk level conditions based on the fault analysis model.

[0158] Step S508: construct a health assessment model.

[0159] Specifically, the server constructs a health assessment model configured with a health assessment method to obtain a health assessment result of the target robot.

[0160] Step S510: construct a decision model.

[0161] Specifically, based on the health assessment results of the target robot, the operation and maintenance cycle and service life of the target robot are predicted to assist the production scheduling system in scheduling the production tasks of the target robot.

[0162] It should be understood that, although the various steps in the flowcharts involved in the various embodiments described above are displayed in sequence according to the instructions of the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the flowcharts involved in the various embodiments described above can include multiple steps or multiple stages, and these steps or stages are not necessarily executed and completed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of steps or stages in other steps.

[0163] Based on the same inventive concept, embodiments of the present application also provide an industrial robot operation and maintenance management device for implementing the aforementioned industrial robot operation and maintenance management method. The solution provided by this device is similar to the solution described in the aforementioned method. Therefore, the specific limitations in one or more embodiments of the industrial robot operation and maintenance management device provided below can be found in the above-mentioned limitations on the industrial robot operation and maintenance management method, and will not be further elaborated here.

[0164] In one embodiment, Figure 6 As shown, an industrial robot operation and maintenance management device is provided, including: a first processing module 602, a second processing module 604, a third processing module 606 and a fourth processing module 608, wherein:

[0165] A first processing module 602 is configured to obtain a target fault analysis model corresponding to the industrial robot, and determine the failure probability of each component in the geometric model corresponding to the industrial robot in the virtual space using the target fault analysis model;

[0166] The second processing module 604 is configured to determine a failure risk level of the industrial robot corresponding to the geometric model based on the failure probability of each component in the geometric model, and select an industrial robot that meets the failure risk level conditions as a target robot;

[0167] The third processing module 606 is configured to collect a plurality of target state data corresponding to each component in the geometric model corresponding to the target robot, and obtain a health assessment result of the target robot based on the plurality of target state data;

[0168] The fourth processing module 608 is used to predict the operation and maintenance cycle and service life of the target robot based on the health assessment result, and adjust the production schedule of the target robot based on the operation and maintenance cycle and service life of the target robot.

[0169] The above-mentioned industrial robot operation and maintenance management device first obtains a target fault analysis model corresponding to the industrial robot and determines the failure probability of each component in the geometric model corresponding to the industrial robot using the target fault analysis model. This allows the device to subsequently accurately and efficiently formulate a corresponding operation and maintenance plan based on the determined failure probability of each component, thereby improving the efficiency of operation and maintenance management. The device then determines the failure risk level of the industrial robot corresponding to the geometric model based on the failure probability of each component in the geometric model, and selects an industrial robot that meets the failure risk level conditions as a target robot. Furthermore, multiple target state data corresponding to each component in the geometric model corresponding to the target robot are collected, and a health assessment result of the target robot is obtained based on the multiple target state data. Based on the health assessment result, the operation and maintenance cycle and service life of the target robot are predicted, so that the operation and maintenance tasks of the target robot can be efficiently arranged. The target robot's production schedule is adjusted based on the operation and maintenance cycle and service life of the target robot to avoid conflicts between the target robot's operation and maintenance tasks and production tasks, ensuring timely maintenance of the target robot. Furthermore, the device prevents the target robot's actual usage time from exceeding its service life, which could lead to failures. This improves the operation and maintenance management efficiency of the industrial robot.

[0170] In one embodiment, the industrial robot operation and maintenance management device also includes a fifth processing module, which is used to construct a geometric model corresponding to the industrial robot in the virtual space, and obtain multiple historical status data corresponding to each component in the geometric model and multiple eigenvalues ​​corresponding to each historical status data, and then construct a feature data set based on the multiple eigenvalues ​​corresponding to each historical status data, the feature data set including multiple eigenvalues ​​corresponding to each historical status data, and then determine the multiple eigenvalues ​​corresponding to each component in the feature data set, and construct an association relationship between the multiple eigenvalues ​​corresponding to the same component and the corresponding component, and finally, construct a data model based on the feature data set and the association relationship, and obtain a target fault analysis model based on the data model.

[0171] In one embodiment, the industrial robot operation and maintenance management device also includes a sixth processing module, which is used to divide the feature data set in the data model into a training set and a test set, obtain multiple initial fault analysis models based on the training set, and determine the failure probability prediction result of each initial fault analysis model for each component based on the association relationship in the data model. Then, based on the test set in the data model, the failure probability prediction result of each initial fault analysis model for each component is verified to obtain the accuracy of each initial fault analysis model, and the initial fault analysis model that meets the accuracy condition is selected as the alternative fault analysis model. When the industrial robot corresponding to the geometric model fails, the status data corresponding to each component in the geometric model at the time of the failure is collected, and the alternative fault analysis model is verified based on the status data at the time of the failure to determine the accuracy of each alternative fault analysis model. Finally, the alternative fault analysis model that meets the accuracy condition is selected as the target fault analysis model.

[0172] In one embodiment, there are multiple target fault analysis models, and the first processing module is also used to collect real-time status data of each component of the industrial robot corresponding to the geometric model, and obtain the characteristic value corresponding to each real-time status data, and then input the characteristic value corresponding to each real-time status data into multiple target fault analysis models to obtain the failure probability prediction result of each target fault analysis model for each component in the geometric model, and then, based on the failure probability prediction result, determine the failure probability of each component in the geometric model.

[0173] In one embodiment, the industrial robot operation and maintenance management device also includes a seventh processing module, which is used to collect multiple historical status data corresponding to each component in the geometric model multiple times, and based on the historical status data collected multiple times, obtain the eigenvalue mean and eigenvalue standard deviation corresponding to each eigenvalue, and then, based on the eigenvalue mean and eigenvalue standard deviation corresponding to each eigenvalue, construct a health assessment model, and the health assessment model is configured with a health assessment method.

[0174] In one embodiment, the third processing module is also used to obtain multiple eigenvalues ​​corresponding to each target state data, and input the multiple eigenvalues ​​corresponding to each target state data into the health assessment model to obtain the difference between each eigenvalue and the corresponding eigenvalue mean, and then compare the difference between each eigenvalue and the corresponding eigenvalue mean with the corresponding eigenvalue standard deviation to obtain a comparison result. Finally, based on the comparison result and the health assessment method, the health assessment result of the target robot is obtained.

[0175] In one embodiment, the industrial robot operation and maintenance management device also includes an eighth processing module, which is used to control the target robot to stop production tasks and run a target distance without load when the health assessment result indicates that the target robot is in a fault state, and to collect status data of the target robot during the no-load operation process, and determine the faulty component of the target robot based on the status data of the target robot during the no-load operation process.

[0176] Each module in the aforementioned industrial robot operation and maintenance management device can be implemented in whole or in part through software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in hardware form, or can be stored in a computer device's memory in software form, so that the processor can call and execute the corresponding operations of each module.

[0177] In one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as follows: Figure 7 As shown. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. The processor, the memory and the input / output interface are connected via a system bus, and the communication interface, the display unit and the input device are connected to the system bus via the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, an industrial robot operation and maintenance management method is implemented. The display unit of the computer device is used to form a visually visible picture, which can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device casing, or an external keyboard, touchpad or mouse.

[0178] Those skilled in the art will understand that Figure 7 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0179] In one embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, and the processor implements the steps in the above method embodiments when executing the computer program.

[0180] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0181] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0182] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of relevant countries and regions.

[0183] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, database or other media used in the embodiments provided in this application may include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory may include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in the various embodiments provided herein may include at least one of a relational database and a non-relational database. Non-relational databases may include, but are not limited to, distributed databases based on blockchains. The processor involved in the various embodiments provided herein may be, but are not limited to, a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic unit, a data processing logic unit based on quantum computing, and the like.

[0184] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0185] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present application. It should be noted that a person of ordinary skill in the art may make various modifications and improvements without departing from the spirit of the present application, and these modifications and improvements fall within the scope of protection of the present application. Therefore, the scope of protection of the present application shall be determined by the appended claims.

Claims

1. A method for operation and maintenance management of an industrial robot, characterized in that: The method comprises: Obtaining a target fault analysis model corresponding to the industrial robot, and determining the failure probability of each component in the geometric model corresponding to the industrial robot through the target fault analysis model, wherein, before obtaining the target fault analysis model corresponding to the industrial robot, the method includes: obtaining a feature data set by repeatedly collecting multiple historical status data of various components of the industrial robot, constructing a data model based on the feature data set, dividing the feature data set in the data model into a training set and a test set, obtaining multiple initial fault analysis models based on the training set, and determining a failure probability prediction result of each initial fault analysis model for each component based on the association relationship in the data model; Based on the test set in the data model, verify the failure probability prediction result of each component of each initial fault analysis model to obtain the accuracy of each initial fault analysis model; select the initial fault analysis model that meets the accuracy condition as the alternative fault analysis model, and the number of the alternative fault analysis models is multiple; when the industrial robot corresponding to the geometric model fails, collect the state data corresponding to each component in the geometric model at the time of the failure, and verify the alternative fault analysis model based on the state data at the time of the failure to determine the accuracy of each alternative fault analysis model; select the alternative fault analysis model that meets the accuracy condition as the target fault analysis model; Determining a failure risk level of an industrial robot corresponding to the geometric model based on a failure probability of each component in the geometric model, and selecting an industrial robot that meets the failure risk level conditions as a target robot; Collecting a plurality of target state data corresponding to each component in the geometric model corresponding to the target robot, and obtaining a health assessment result of the target robot based on the plurality of target state data; Based on the health assessment result, the operation and maintenance cycle and service life of the target robot are predicted, and based on the operation and maintenance cycle and service life of the target robot, the production schedule of the target robot is adjusted.

2. The method according to claim 1, characterized in that The step of obtaining the target fault analysis model corresponding to the industrial robot includes: Constructing a geometric model corresponding to the industrial robot in a virtual space, and obtaining a plurality of historical state data corresponding to each component in the geometric model and a plurality of eigenvalues ​​corresponding to each historical state data; Constructing a feature data set based on the multiple feature values ​​corresponding to each piece of historical status data, wherein the feature data set includes the multiple feature values ​​corresponding to each piece of historical status data; Determining a plurality of feature values ​​corresponding to each component in the feature data set, and establishing an association relationship between the plurality of feature values ​​corresponding to the same component and the corresponding component; A data model is constructed based on the feature data set and the association relationship, and the target fault analysis model is obtained based on the data model.

3. The method according to claim 1, characterized in that There are multiple target fault analysis models; Determining the failure probability of each component in the geometric model corresponding to the industrial robot using the target fault analysis model includes: Collecting real-time status data of various components of the industrial robot corresponding to the geometric model, and obtaining characteristic values ​​corresponding to each real-time status data; Inputting the characteristic value corresponding to each real-time status data into multiple target fault analysis models to obtain a failure probability prediction result of each target fault analysis model for each component in the geometric model; Based on the failure probability prediction result, the failure probability of each component in the geometric model is determined.

4. The method according to claim 1, wherein The step of collecting a plurality of target state data corresponding to each component in the geometric model corresponding to the target robot and obtaining a health assessment result of the target robot based on the plurality of target state data includes: Collecting a plurality of historical state data corresponding to each component in the geometric model multiple times, and obtaining an eigenvalue mean and an eigenvalue standard deviation corresponding to each eigenvalue based on the historical state data collected multiple times; Constructing a health assessment model based on the eigenvalue mean and the eigenvalue standard deviation corresponding to each eigenvalue, wherein the health assessment model is configured with a health assessment method; The collecting of a plurality of target state data corresponding to each component in the geometric model corresponding to the target robot, and obtaining a health assessment result of the target robot based on the plurality of target state data includes: Acquire multiple characteristic values ​​corresponding to each target state data, and input the multiple characteristic values ​​corresponding to each target state data into the health assessment model to obtain the difference between each characteristic value and the corresponding characteristic value mean; Compare the difference between each eigenvalue and the corresponding eigenvalue mean with the corresponding eigenvalue standard deviation to obtain a comparison result; Based on the comparison result and the health assessment method, a health assessment result of the target robot is obtained.

5. The method according to claim 1, wherein After collecting a plurality of target state data corresponding to each component in the geometric model corresponding to the target robot and obtaining a health assessment result of the target robot based on the plurality of target state data, the method further includes: When the health assessment result indicates that the target robot is in a fault state, controlling the target robot to stop the production task and run the target distance without load; State data of the target robot during no-load operation is collected, and a faulty component of the target robot is determined based on the state data of the target robot during no-load operation.

6. An industrial robot operation and maintenance management device, characterized in that: The device comprises: The first processing module is used to obtain a target fault analysis model corresponding to the industrial robot, and determine the failure probability of each component in the geometric model corresponding to the industrial robot in the virtual space through the target fault analysis model, wherein, before obtaining the target fault analysis model corresponding to the industrial robot, it also includes: obtaining a feature data set by collecting multiple historical status data of each component of the industrial robot multiple times, and constructing a data model based on the feature data set, dividing the feature data set in the data model into a training set and a test set, obtaining multiple initial fault analysis models based on the training set, and determining the failure probability of each component for each initial fault analysis model based on the association relationship in the data model. Failure probability prediction results; based on the test set in the data model, verify the failure probability prediction results of each component of each initial fault analysis model to obtain the accuracy of each initial fault analysis model; select the initial fault analysis model that meets the accuracy conditions as the alternative fault analysis model, and the number of the alternative fault analysis models is multiple; when the industrial robot corresponding to the geometric model fails, collect the state data corresponding to each component in the geometric model at the time of the failure, and verify the alternative fault analysis model based on the state data when the failure occurs, and determine the accuracy of each alternative fault analysis model; select the alternative fault analysis model that meets the accuracy conditions as the target fault analysis model; a second processing module, configured to determine a failure risk level of the industrial robot corresponding to the geometric model based on the failure probability of each component in the geometric model, and select an industrial robot that meets the failure risk level conditions as a target robot; a third processing module, configured to collect a plurality of target state data corresponding to each component in the geometric model corresponding to the target robot, and obtain a health assessment result of the target robot based on the plurality of target state data; The fourth processing module is used to predict the operation and maintenance cycle and service life of the target robot based on the health assessment result, and adjust the production schedule of the target robot based on the operation and maintenance cycle and service life of the target robot.

7. The device according to claim 6, characterized in that There are multiple target fault analysis models, and the first processing module is also used to collect real-time status data of each component of the industrial robot corresponding to the geometric model, and obtain the characteristic value corresponding to each real-time status data, input the characteristic value corresponding to each real-time status data into multiple target fault analysis models, and obtain the failure probability prediction result of each target fault analysis model for each component in the geometric model; based on the failure probability prediction result, determine the failure probability of each component in the geometric model.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 5 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

Images