Industrial robot monitoring method, device and equipment and storage medium

Abstract

The invention discloses an industrial robot monitoring method, device and equipment and a storage medium. The method comprises the steps of obtaining operation data of a target robot; inputting the operation data into a preset digital twinborn model, and performing operation simulation by the preset digital twinborn model according to the operation data to obtain the residual life of each part of the target robot in a preset future time period; based on the residual life, performing color rendering on a simulation robot corresponding to the preset digital twin model to obtain a rendered robot; monitoring the target robot according to the rendering robot to obtain a monitoring result; and generating an execution strategy in the preset future time period under the condition that the monitoring result meets a preset alarm condition. According to the embodiment of the invention, prediction can be carried out in advance before the industrial robot breaks down, so that processing can be carried out in advance before the fault occurs, and the loss caused by the robot fault is reduced.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of industrial robot monitoring, and particularly relates to an industrial robot monitoring method, device, equipment and storage medium. BACKGROUND

[0002] With the rapid development of society, industrial robots are applied more and more widely in production, and production efficiency is greatly improved. The higher the degree of automation of a production enterprise is, the lower the labor cost will be, and the more residual value can be created. When an industrial robot produces, a standard action path is designed first, and then the robot performs basic actions such as grabbing a product, transporting a product, installing a product and resetting according to the fixed standard action path to complete a production task.

[0003] In actual production, an industrial robot may fail due to various program reasons, affecting production. Therefore, it is necessary to monitor the health of the industrial robot. At present, detection is mainly performed according to the running state of the robot, that is, whether the robot fails is determined according to the motion path of the mechanical arm of the robot.

[0004] In related technologies, once the motion path of the mechanical arm deviates, it can be determined that the robot fails at this time, but this is after the robot fails, resulting in unplanned downtime and causing huge production losses. SUMMARY

[0005] The embodiments of the application provide an industrial robot monitoring method, device, equipment and storage medium, which can predict in advance before an industrial robot fails, so as to process in advance before the failure occurs and reduce losses caused by the robot failure.

[0006] In one aspect, the embodiments of the application provide an industrial robot monitoring method, which comprises: obtaining running data of a target robot; inputting the running data into a preset digital twin model, performing running simulation on the preset digital twin model according to the running data, and obtaining residual life of each part of the target robot in a preset future period; based on the residual life, performing color rendering on a simulation robot corresponding to the preset digital twin model to obtain a rendered robot; monitoring the target robot according to the rendered robot to obtain a monitoring result; generating an execution strategy in the preset future period when the monitoring result meets a preset alarm condition.

[0007] Optionally, in the step of inputting the running data into the preset digital twin model, the preset digital twin model performs running simulation according to the running data to obtain the residual life of each part of the target robot before a preset future period, and the method further comprises: obtaining a structural drawing of the target robot and a target scene, the target scene including a working environment and a working pipeline of the target robot; building a preset digital twin model based on the structural drawing, the working environment and the working pipeline; in the preset digital twin model, being configured with: a real-time connection module for obtaining running data of the target robot; a parsing driving module for parsing the received running data to obtain parsed data, and updating the motion state of each part of the preset digital twin model according to the parsed data.

[0008] Optionally, in the step of inputting the running data into the preset digital twin model, the preset digital twin model performs running simulation according to the running data to obtain the residual life of each part of the target robot before a preset future period, and the method further comprises: performing feature extraction on the running data to obtain multi-dimensional running features; performing feature fusion on the multi-dimensional running features to obtain a fusion health index of each part of the target robot; performing curve fitting on the fusion health index according to a preset time sequence to obtain a fitted degradation trajectory; determining the residual life of each part based on the fitted degradation trajectory and a preset threshold.

[0009] Optionally, the step of determining the residual life of each part based on the fitted degradation trajectory and a preset threshold comprises: obtaining an intersection point of the fitted degradation trajectory and a plurality of preset thresholds; determining the residual life according to a difference between the intersection point and a current time.

[0010] Optionally, the multi-dimensional running features include vibration features, current features, temperature features and sound features, and the step of performing feature fusion on the multi-dimensional running features to obtain a fusion health index of each part of the target robot comprises: normalizing the multi-dimensional running features to obtain standard multi-dimensional running features; scoring the standard multi-dimensional running features using a preset fault symptom correlation matrix to obtain correlation scores, the preset fault symptom correlation matrix being used to calculate the correlation degree of each multi-dimensional running feature and a fault; According to the correlation score and a preset correlation threshold, strong correlation features are screened from a plurality of standard multi-dimensional operation features; The strong correlation features are screened by using a preset principal component analysis algorithm to obtain a plurality of fault key features; The plurality of fault key features are combined into a vector to obtain a to-be-diagnosed vector; The to-be-diagnosed vector is input into a preset diagnosis model to obtain a fusion health index.

[0011] Optionally, according to the rendering robot, the target robot is monitored to obtain a monitoring result, including: According to the color rendering value of each part in the rendering robot within a preset time period, a failure occurrence probability of the rendering robot in a preset future time period is determined; Based on the failure occurrence probability, failure warning information is generated; The failure warning information is determined as the monitoring result.

[0012] Optionally, based on the remaining life, the simulation robot corresponding to the preset digital twin model is color rendered to obtain a rendering robot, including: Based on the remaining life, the life grade of each part is determined; According to the life grade and the fusion health index, the color rendering value of each part is calculated; According to the color rendering value, the simulation robot is color rendered to obtain a rendering robot.

[0013] On the other hand, the embodiment of the present application provides an industrial robot monitoring device, the device comprising: An acquisition module is configured to acquire operation data of a target robot; An input module is configured to input the operation data into a preset digital twin model, and the preset digital twin model is configured to simulate the operation according to the operation data to obtain the remaining life of each part of the target robot within a preset future time period; A color rendering module is configured to color render a simulation robot corresponding to the preset digital twin model based on the remaining life to obtain a rendering robot; A monitoring module is configured to monitor the target robot according to the rendering robot to obtain a monitoring result; A generation module is configured to generate an execution strategy within the preset future time period when the monitoring result meets a preset alarm condition.

[0014] In still another aspect, the embodiment of the present application provides an electronic device, the device comprising a processor and a memory storing computer program instructions; The processor implements the industrial robot monitoring method according to any one of the first aspect when executing the computer program instructions.

[0015] In another aspect, an embodiment of the present application provides a computer storage medium, and computer program instructions are stored on the computer readable storage medium, and the computer program instructions are executed by a processor to implement the industrial robot monitoring method according to any one of the first aspect.

[0016] The industrial robot monitoring method, device, equipment and storage medium provided by the embodiments of the present application can summarize the running data through the preset digital twin model, break through the limitation of a single data source, convert abstract sensor data into intuitive model states through information complementation, enable a user to intuitively focus on future faults and component life of a target robot through a rendered robot in the preset digital twin model, make early decisions for subsequent operation and maintenance and fault processing of the target robot, and reduce losses caused by unplanned downtime of the target robot. BRIEF DESCRIPTION OF DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments of the present application. Those skilled in the art can obtain other drawings according to these drawings without creative labor.

[0018] Figure 1 is a flow diagram of an industrial robot monitoring method provided by an embodiment of the present application; Figure 2 is a structural diagram of an industrial robot monitoring device provided by another embodiment of the present application; Figure 3 is a structural diagram of an electronic device provided by another embodiment of the present application. DETAILED DESCRIPTION

[0019] The features and exemplary embodiments of various aspects of the present application will be described in detail below, in order to make the purposes, technical solutions and advantages of the present application more clear and apparent, the following will further describe the present application in combination with the drawings and specific embodiments. It should be understood that the specific embodiments described herein are only intended to explain the present application, but not to limit the present application. The present application can be implemented without some of these specific details by those skilled in the art. The following description of the embodiments is only to provide a better understanding of the present application by showing examples of the present application.

[0020] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes the element.

[0021] To address the problems of existing technologies, embodiments of this application provide an industrial robot monitoring method, apparatus, equipment, storage medium, and program product. In these embodiments, operational data can be aggregated using a preset digital twin model, overcoming the limitations of a single data source. Through information complementarity, abstract sensor data is transformed into an intuitive model state. Users can intuitively monitor future failures and component lifespans of the target robot through a rendered robot within the preset digital twin model, enabling advance decision-making for subsequent maintenance and fault handling of the target robot and reducing losses incurred during unplanned downtime.

[0022] The industrial robot monitoring method provided in the embodiments of this application will be introduced first below.

[0023] Figure 1 A flowchart illustrating an embodiment of the industrial robot monitoring method provided in this application is shown. Figure 1 As shown, industrial robot monitoring methods may include S101-S105: S101, acquire the target robot's operating data.

[0024] In this embodiment, different sensors can be deployed at various locations of the target robot to construct its perceptual neural network. These sensors can be vibration sensors, high-precision current / voltage sensors, acoustic microphone / ultrasonic sensors, temperature sensors, and vision sensors. Vibration sensors can be installed at key joints and gearboxes to monitor bearing wear, abnormal gear meshing, and mechanical loosening. High-precision current / voltage sensors can monitor the current and voltage fluctuations of the motor driver, analyze load changes, abnormal torque, and electrical faults. Acoustic microphone / ultrasonic sensors can collect operating noise and detect early anomalies that are imperceptible to the human ear through voiceprint recognition technology. Temperature sensors can be used to monitor the temperature of the motor, reducer, and control cabinet to prevent overheating damage. Vision sensors can be equipped with industrial cameras to monitor end effector vibration, trajectory deviation, and external collisions.

[0025] In this embodiment, the electronic device can collect sensor data transmitted by all sensors through an embedded intelligent acquisition terminal, perform preliminary edge computing processing, and transmit the data to the cloud server behind the electronic device via communication signals, providing a data foundation for the monitoring of the target robot.

[0026] S102, input the running data into the preset digital twin model, and the preset digital twin model performs a running simulation according to the running data to obtain the remaining lifespan of each part of the target robot in a preset future time period.

[0027] In this embodiment, in order to construct an accurate preset digital twin model, the method may further include the following steps before S102: Obtain the structural drawings of the target robot and the target scene; Based on structural drawings, working environment, and workflow, a pre-set digital twin model is built; The preset digital twin model is configured with: The real-time connection module is used to acquire the target robot's operational data. The parsing driver module is used to parse the received running data, obtain parsed data, and update the motion state of each part of the preset digital twin model based on the parsed data.

[0028] In this embodiment, the user can input the target robot and the target scenario in which the target robot is deployed into the electronic device to build a preset digital twin model to more closely resemble the actual use scenario. The target scenario may include the working environment and workflow of the target robot. For example, the target robot is a robotic arm, the working environment may be a logistics, welding and handling scenario, and the workflow may be a corresponding workflow.

[0029] A pre-defined digital twin model is built using the structural drawings of the target robot and the target scene.

[0030] Furthermore, this real-time connection module can establish a connection with electronic devices via a network and directly acquire sensor data through the corresponding communication interface, so that the preset digital twin model can be synchronized according to the actual situation. The parsing and driving module can update the operating status of each part of the robot in the preset digital twin model according to the received operating data, enabling the virtual model to move synchronously with the physical entity.

[0031] In this embodiment, after the preset digital twin model is constructed and can move according to the physical entity, S102 may specifically include: Feature extraction is performed on the operational data to obtain multi-dimensional operational features; Feature fusion is performed on multi-dimensional operational characteristics to obtain the fused health index of each part of the target robot; Based on the integrated health indicators, curve fitting is performed according to a preset time series to obtain the fitted degradation trajectory; Based on the fitted degradation trajectory and preset threshold, the remaining lifespan of each part is determined.

[0032] In this embodiment, in order to monitor the target robot using a preset digital twin model, it is necessary to extract multi-dimensional operational features from the operational data, identify features associated with the target robot's faults from these multi-dimensional operational features, and focus on these features to predict the target robot's performance.

[0033] Specifically, the purpose of feature extraction from operational data is to compress and extract the most relevant and quantifiable information about equipment degradation and failures from the massive amount of operational data. As an example, analysis can be performed in the time domain, frequency domain, and video domain. Examples will be provided below: Time-domain characteristics: Statistics can be directly calculated from the waveform signal where the vibration amplitude changes over time, quickly reflecting energy changes and the distribution pattern of the signal. For example, the average value of a mean signal reflects the DC component of the signal. Root mean square (RMS) value: Also known as the effective value, it is the most important indicator for measuring the magnitude of vibration energy. A trend of increasing value usually indicates the development of faults such as imbalance or loosening. Peak value: The maximum absolute value of the signal, highly sensitive to impact-related faults (such as localized cracks). Kurtosis: An indicator measuring the steepness of the signal distribution. The vibration signal distribution in a healthy state approximates a normal distribution (kurtosis ≈ 3). When an impact fault occurs, the kurtosis value increases significantly, making it an excellent early fault indicator. Skewness: An indicator measuring the asymmetry of the signal distribution. Waveform factor, peak factor, and impulse factor: These are composite factors with different sensitivities to different fault modes (e.g., the peak factor is sensitive to impact).

[0034] Regarding frequency domain characteristics, the spectrum can be used for analysis. For example, the time domain signal can be converted into a frequency domain signal through Fourier transform to obtain a spectrum diagram for spectral analysis.

[0035] Specifically, the vibration signal can be filtered and denoised first. The signal can then be transformed from the time domain to the frequency domain using the Fast Fourier Transform algorithm to obtain the spectrum. After obtaining the spectrum, the fault frequency point can be calculated based on the fault frequency of the rotating parts of the machine. It can be observed whether the amplitude in the spectrum significantly exceeds the standard or new spectral peaks appear. For example, for bearings, the fault frequency of the inner ring, outer ring, rolling elements, and cage can be calculated. For gears, the meshing frequency and its sidebands can be calculated.

[0036] Regarding the time-frequency domain, envelope spectrum analysis can be performed. The specific analysis steps are as follows: Bandpass filtering: Select a frequency band that includes the high-frequency resonance caused by the fault impulse (usually determined through spectrum analysis).

[0037] Hilbert Transform: Performing a Hilbert transform on the filtered signal yields its analytic signal.

[0038] Envelope analysis: Calculate the amplitude of the analytical signal; this is the envelope signal. It represents the "profile" of the high-frequency signal, i.e., the rhythm of the fault impact.

[0039] Perform an FFT on the envelope to obtain the envelope spectrum. The frequencies corresponding to the peaks in this spectrum are the repetition frequencies of the fault impact.

[0040] For current signal analysis, a Fast Fourier Transform (FFT) can be performed on the current to obtain a detailed current spectrum. This spectrum can be used to analyze faults such as broken bars, eccentricity, and load torque fluctuations. Broken bars generate sidebands (1±2ks)fs on both sides of the fundamental frequency (where s is the slip and fs is the supply frequency). Eccentricity faults produce specific harmonic components. Load torque fluctuations (such as those from reducer faults) also modulate the current, generating sidebands. Faults are diagnosed by monitoring changes in the amplitude of these characteristic harmonics.

[0041] For audio signals, the following time-frequency analysis techniques can be used: Short-Time Fourier Transform (STFT): Principle: Divide the long signal into many short segments, perform FFT on each segment, and finally stack the series of spectra into a two-dimensional time-frequency graph.

[0042] Result: A spectrogram is obtained, which can visually show how the frequency components change over time.

[0043] Advantages and disadvantages: It is simple and intuitive, but time resolution and frequency resolution cannot be achieved at the same time (Heisenberg uncertainty principle).

[0044] Wavelet transform: Principle: A scalable and shiftable mother wavelet function is used to fit the signal, which can provide good localization characteristics in both the time and frequency domains.

[0045] Results: The energy distribution of the signal at different scales (corresponding frequencies) and times was obtained.

[0046] Applications: It is very suitable for extracting transient impact features and is often used for noise reduction and feature extraction of sound and vibration signals.

[0047] Mel frequency cepstral coefficients (MFCC): Principle: This is a classic feature in speech recognition, but it is equally applicable to industrial sound analysis. It simulates the human ear's ability to perceive different frequencies, compressing the spectrum to a lower dimension.

[0048] How to extract: Perform FFT on the signal -> Map to Mel scale -> Take logarithm -> Perform Discrete Cosine Transform.

[0049] Advantages: The extracted features have low dimensionality and best represent the "perceptual" characteristics of sound, making them very suitable as input for machine learning models.

[0050] In the embodiments described in this application, feature extraction and analysis can be performed in the manner described above to provide a data foundation for subsequent accurate monitoring of the target robot.

[0051] In other embodiments, the multi-dimensional operational features include vibration features, current features, temperature features, and sound features. Feature fusion is performed on these multi-dimensional operational features to obtain fused health indicators for each part of the target robot, including: The multi-dimensional operational characteristics are normalized to obtain standard multi-dimensional operational characteristics; The standard multi-dimensional operational characteristics are scored using a pre-set fault symptom association matrix to obtain an association score; Based on the correlation score and the preset correlation threshold, strong correlation features are selected from multiple standard multi-dimensional operational features. By using a pre-defined principal component analysis algorithm to filter strongly correlated features, several key fault features were obtained. Multiple key fault features are merged into vectors to obtain the vector to be diagnosed; The vector to be diagnosed is input into the preset diagnostic model to obtain the fused health indicators.

[0052] In this embodiment, the above-mentioned multi-dimensional operational features are merged to obtain a high-dimensional joint feature vector, which is then input into the classifier for decision-making. Specifically, the following can be extracted from the vibration features: time-domain statistics (RMS, kurtosis), frequency-domain features (fault frequency amplitude), and envelope spectrum features; current features: harmonic component amplitude and current ripple effective value; temperature features: current temperature and temperature rise rate; and sound: MFCC feature mean.

[0053] Then, all feature values ​​are normalized to the same range, such as 0-1, to avoid the model being biased towards features with large values ​​due to inconsistent dimensions. Then, the standard multi-dimensional operation features are scored using a preset fault correlation matrix to obtain the correlation score. The preset fault symptom correlation matrix is ​​used to calculate the degree of correlation between each multi-dimensional operation feature and the fault.

[0054] Specifically, users can list all possible faults of industrial robots and the faults they want to detect, such as: Mechanical issues: worn joint bearings, broken / worn gears in joint reducers, loose connecting bolts on robotic arms, worn guide rails, and insufficient belt tension.

[0055] Electrical components: Servo motor winding insulation failure, encoder failure, brake failure, driver IGBT aging.

[0056] Other issues include insufficient lubrication, cooling fan malfunction, collision of the robotic arm, and abnormal external load.

[0057] Then list all the feature metrics that can be extracted from sensors and data, such as: Vibration: Total vibration value (RMS), kurtosis, peak value, amplitude of specific fault frequencies (such as BPFI, BPFO), and envelope spectrum energy.

[0058] Current: RMS current value, THD (Total Harmonic Distortion) of current, amplitude of specific harmonic components, and RMS torque ripple.

[0059] Temperature: motor temperature, reducer temperature, ambient temperature, temperature rise rate (ΔT / Δt).

[0060] Sound: Overall sound pressure level, sound energy in a specific frequency band.

[0061] Process parameters: tracking error (difference between actual position and commanded position), completion cycle time, and load rate.

[0062] The relationship between each failure mode and each symptom indicator can then be scored, using probability to describe it. The specific scoring criteria are as follows: 0: No correlation. This symptom usually does not occur when the fault occurs.

[0063] 1: Weak association. May occur, but is unstable or not a primary feature.

[0064] 2: Moderate correlation. It is a relatively reliable indicative symptom.

[0065] 3: Strong correlation. This is a very typical and sensitive symptom of the fault.

[0066] N / A: Not applicable.

[0067] In this embodiment, the failure modes and symptoms can be found in Table 1 below:

[0068] The correlation degree can be determined by referring to Table 1 above, thus obtaining the strong correlation feature.

[0069] Then, the pre-set principal component analysis algorithm is used to filter the strongly correlated features to obtain multiple key fault features. Specifically, a new set of orthogonal axes (principal components PC1, PC2, PC3...) is found through the pre-set principal component analysis algorithm and sorted in order according to the size of the variance of the retained running data. Among them, PC1 is the direction with the largest variance, PC2 is the direction with the second largest variance that is orthogonal to PC1, and so on.

[0070] Then plot the variance explained by each principal component. Select the components before the point where the curve flattens out. For example, if the curve becomes an "elbow" after the first k components, select those k components.

[0071] Cumulative variance explained: Typically, the minimum number of principal components required to achieve a cumulative variance explained > 85% or 90% is chosen. This means that these k principal components capture 90% of the information in the original data.

[0072] View the loading matrix of each principal component.

[0073] Loadings represent the contribution of each original feature to the principal component. The larger the absolute value of the loading, the more important the original feature is to the principal component.

[0074] Finally, you can either directly output the key features of the fault, or retain the original features that contribute highly to the principal components and use them as the key features of the fault.

[0075] Then, multiple key fault features are merged into a vector to obtain the vector to be diagnosed, which is then input into a preset diagnostic model to obtain a fused health index. The preset diagnostic model can be a gradient boosting tree (GBDT), a support vector machine (SVM), or a neural network.

[0076] For example, the joint feature vector for diagnosing "joint reducer wear" in a robot might be: [Vibration_RMS, Vibration_kurtosis, Vibration_meshing frequency amplitude, Current_specific sideband energy, Temperature, Temperature rise rate].

[0077] A good balance can be achieved between information retention and computational efficiency.

[0078] In this embodiment, the integrated health indicator can be a value that approaches 0 from 1.

[0079] In this embodiment, the fused health indicators can be fitted into a mathematical model over time / operation cycle, i.e., fitted degradation trajectory, wherein the mathematical model can be a direct model, a linear model, or a Wiener process or a gamma process.

[0080] In this embodiment, determining the remaining lifespan of each component based on the fitted degradation trajectory and a preset threshold can specifically include: Obtain the intersection points of the fitted degenerate trajectory and multiple preset thresholds; The remaining lifetime is determined based on the difference between the intersection point and the current time.

[0081] In this embodiment, when the remaining lifespan of each part is determined, multiple preset thresholds can be set on the time axis. These preset thresholds represent different lifespan levels reached by each part. The remaining lifespan can be determined through these lifespan levels, which can also provide support for subsequent maintenance and early warning.

[0082] Specifically, the lifespan level can include a normal level, a warning level, and a caution level. The normal level corresponds to a first preset threshold, the warning level corresponds to a second preset threshold, and the caution level corresponds to a third preset threshold. The first and second preset thresholds decrease sequentially.

[0083] S103, based on the remaining lifespan, perform color rendering on the simulation robot corresponding to the preset digital twin model to obtain the rendered robot.

[0084] In some embodiments, specifically, S103 may include: Determine the lifespan class of each part based on its remaining lifespan; Calculate the color rendering value for each part based on its lifespan rating and fusion health index; The simulated robot is rendered using color rendering values ​​to obtain the rendered robot.

[0085] In this embodiment, the lifespan level can be the lifespan level mentioned above. More specifically, it can be calculated using three-channel RGB color based on the remaining lifespan and fused health indicators. The color rendering value of each part can be obtained through different RGB values, so as to see the lifespan of each part more intuitively. It can be understood that the color of the simulation robot can be initialized to white, and then the simulation robot can be rendered according to the color rendering value of each part to obtain the rendered robot, so as to understand the usage status of the parts more quickly.

[0086] Furthermore, the preset digital twin model can save the lifespan of each part, providing a basis for subsequent maintenance and repair. It can also allow users to understand the replacement frequency or failure frequency of each part, thus enabling them to add these parts as key focus parts.

[0087] S104, Based on the rendering robot, monitor the target robot and obtain the monitoring results.

[0088] In other embodiments, after obtaining the rendering robot, monitoring can be performed according to preset alarm rules. Specifically, S104 may include: Based on the color rendering values ​​of each part in the rendering robot within a preset time period, determine the probability of failure of the rendering robot in a preset future time period. Based on the probability of failure occurrence, generate fault warning information; The fault warning information is identified as a monitoring result.

[0089] Specifically, in this embodiment, since the color rendering value is determined by the fusion of health indicators and lifespan level, the probability of failure in a preset future time period can be analyzed by the remaining lifespan and the fusion of health indicators, thereby generating fault warning information to determine that the part corresponding to the target robot may fail in a certain time period in the future and to remind the user.

[0090] In other embodiments, the preset alarm rule can be a single symptom threshold rule, a multi-symptom combined rule, or an exclusionary rule. These alarm rules will be discussed in detail below: Single symptom threshold rule: If the vibration kurtosis is greater than the preset threshold, an impulsive vibration anomaly can be triggered. If the motor temperature is greater than the temperature threshold, a motor overheating alarm can be triggered. If the tracking error is greater than the error threshold, a positioning deviation alarm can be triggered. The preset threshold can be 5, the temperature threshold can be 90℃, and the error threshold can be 0.5rad.

[0091] Multiple symptom combination rule: and rules: If (vibration BPFO amplitude > threshold) AND (vibration kurtosis > threshold); Alarm message: High confidence: Suspected failure of the outer ring of the spherical plain bearing; Note: A high amplitude at a single frequency may be interference, but it is often accompanied by an impulse.

[0092] Or rules: If (current_THD>threshold) OR motor temperature>threshold); An alarm will sound: the motor's electrical performance is abnormal.

[0093] Note: Motor problems may manifest as abnormal current or overheating.

[0094] Sequence rules: If (the vibration peak increases sharply within 1 second) AND (the current increases sharply within 1 second) AND (the current returns to normal) AND (the vibration peak remains at a high level); An alarm will sound: suspected mechanical collision.

[0095] Note: Captures the impact and current overload at the moment of collision, as well as the continuous abnormal vibration that may be caused after the collision.

[0096] Exclusionary rules: used to rule out other possibilities and further refine fault location.

[0097] If (the vibration BPFO amplitude is high) AND (the motor temperature is normal); Therefore, the conclusion is that the fault is more likely to be in the mechanical bearing itself, rather than caused by motor overheating.

[0098] If (alarm "positioning deviation too large") AND (current normal); Therefore, the problem is likely in the mechanical transmission (such as the reducer or belt) or the encoder, rather than insufficient motor power.

[0099] S105: If the monitoring results meet the preset alarm conditions, generate the execution strategy for the preset future time period.

[0100] In this embodiment, if a fault is detected in the target robot within a preset future time period through a preset digital twin model, it can be handled by referring to the previous execution strategy, or it can be handled manually, or the strategy can be executed in advance in the preset digital twin model to achieve accurate handling of the fault.

[0101] In this embodiment, operational data can be aggregated through a preset digital twin model, breaking through the limitations of a single data source. Through information complementarity, abstract sensor data can be transformed into an intuitive model state. Users can intuitively monitor the future failures and component lifespan of the target robot through the rendered robot in the preset digital twin model, making advance decisions for the subsequent operation and maintenance and fault handling of the target robot, and reducing the losses caused by unplanned downtime of the target robot.

[0102] Based on the industrial robot monitoring method provided in the above embodiments, this application also provides specific implementation methods for an industrial robot monitoring device. Please refer to the following embodiments.

[0103] First see Figure 2 The industrial robot monitoring device 200 provided in this application embodiment includes: The acquisition module 201 is used to acquire the operating data of the target robot; The input module 202 is used to input the running data into the preset digital twin model. The preset digital twin model performs a running simulation according to the running data to obtain the remaining lifespan of each part of the target robot in a preset future time period. Color rendering module 203 is used to perform color rendering on the simulation robot corresponding to the preset digital twin model based on the remaining lifespan, so as to obtain the rendered robot. The monitoring module 204 is used to monitor the target robot based on the rendering robot and obtain the monitoring results; The generation module 205 is used to generate an execution strategy for a preset future time period when the monitoring results meet the preset alarm conditions.

[0104] As an optional implementation, the input module 202 can also be used for: Obtain the structural drawings and target scene of the target robot, including the working environment and production line of the target robot; Based on structural drawings, working environment, and workflow, a pre-set digital twin model is built; The preset digital twin model is configured with: The real-time connection module is used to acquire the target robot's operational data. The parsing driver module is used to parse the received running data, obtain parsed data, and update the motion state of each part of the preset digital twin model based on the parsed data.

[0105] As an optional implementation, the input module 202 can also be used for: Feature extraction is performed on the operational data to obtain multi-dimensional operational features; Feature fusion is performed on multi-dimensional operational characteristics to obtain the fused health index of each part of the target robot; Based on the integrated health indicators, curve fitting is performed according to a preset time series to obtain the fitted degradation trajectory; Based on the fitted degradation trajectory and preset threshold, the remaining lifespan of each part is determined.

[0106] As an optional implementation, the input module 202 can also be used for: Obtain the intersection points of the fitted degenerate trajectory and multiple preset thresholds; The remaining lifetime is determined based on the difference between the intersection point and the current time.

[0107] As an optional implementation, the multi-dimensional operating characteristics include vibration characteristics, current characteristics, temperature characteristics, and sound characteristics. The input module 202 can also be used for: The multi-dimensional operational characteristics are normalized to obtain standard multi-dimensional operational characteristics; The standard multi-dimensional operating features are scored using a preset fault symptom association matrix to obtain association scores. The preset fault symptom association matrix is ​​used to calculate the degree of association between each multi-dimensional operating feature and the fault. Based on the correlation score and the preset correlation threshold, strong correlation features are selected from multiple standard multi-dimensional operational features. By using a pre-defined principal component analysis algorithm to filter strongly correlated features, several key fault features were obtained. Multiple key fault features are merged into vectors to obtain the vector to be diagnosed; The vector to be diagnosed is input into the preset diagnostic model to obtain the fused health indicators.

[0108] As an optional implementation, the monitoring module 204 can also be used for: Based on the color rendering values ​​of each part in the rendering robot within a preset time period, determine the probability of failure of the rendering robot in a preset future time period. Based on the probability of failure occurrence, generate fault warning information; The fault warning information is identified as a monitoring result.

[0109] As an alternative implementation, the color rendering module 203 can also be used for: Determine the lifespan class of each part based on its remaining lifespan; Calculate the color rendering value for each part based on its lifespan rating and fusion health index; The simulated robot is rendered using color rendering values ​​to obtain the rendered robot.

[0110] Figure 3A schematic diagram of the hardware structure of the electronic device provided in an embodiment of this application is shown.

[0111] An electronic device may include a processor 301 and a memory 302 storing computer program instructions.

[0112] Specifically, the processor 301 may include a central processing unit (CPU), an application specific integrated circuit (ASIC), or one or more integrated circuits that can be configured to implement the embodiments of this application.

[0113] Memory 302 may include mass storage for data or instructions. For example, and not limitingly, memory 302 may include a hard disk drive (HDD), floppy disk drive, flash memory, optical disk, magneto-optical disk, magnetic tape, or Universal Serial Bus (USB) drive, or a combination of two or more of these. In one instance, memory 302 may include removable or non-removable (or fixed) media, or memory 302 may be non-volatile solid-state memory. Memory 302 may be internal or external to the integrated gateway disaster recovery device.

[0114] In one instance, memory 302 may be read-only memory (ROM). In one instance, the ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), an electrically rewritable ROM (EAROM), or flash memory, or a combination of two or more of these.

[0115] Memory 302 may include read-only memory (ROM), random access memory (RAM), disk storage media device, optical storage media device, flash memory device, electrical, optical, or other physical / tangible memory storage device. Therefore, typically, memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the industrial robot monitoring method according to the first aspect of this disclosure.

[0116] The processor 301 reads and executes computer program instructions stored in the memory 302 to achieve... Figure 1 An industrial robot monitoring method is shown in the embodiment.

[0117] In one example, the electronic device may also include a communication interface 303 and a bus 304. For example, Figure 3 As shown, the processor 301, memory 302, and communication interface 303 are connected through bus 304 and complete communication with each other.

[0118] The communication interface 303 is mainly used to realize communication between various modules, devices, units and / or equipment in the embodiments of this application.

[0119] Bus 304 includes hardware, software, or both, that couples components of an electronic device together. For example, and not as a limitation, the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Extended Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a Hyper Transport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an Infinite Bandwidth Interconnect, a Low Pin Count (LPC) bus, a memory bus, a Microchannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or other suitable buses, or a combination of two or more of these. Where appropriate, bus 304 may include one or more buses. Although specific buses are described and illustrated in embodiments of this application, this application contemplates any suitable bus or interconnect.

[0120] The electronic device can execute the industrial robot monitoring method in the embodiments of this application, thereby achieving a combination of Figures 1-2 The described industrial robot monitoring method and device.

[0121] Furthermore, in conjunction with the industrial robot monitoring methods in the above embodiments, this application embodiment can provide a computer storage medium for implementation. The computer storage medium stores computer program instructions; when these computer program instructions are executed by a processor, they implement any of the industrial robot monitoring methods in the above embodiments.

[0122] In an optional embodiment, in conjunction with the industrial robot monitoring methods in the above embodiments, this application embodiment can provide a computer program product to implement the method. The instructions in the computer program product are executed by the processor of an electronic device, enabling the electronic device to implement any of the industrial robot monitoring methods in the above embodiments.

[0123] It should be clarified that this application is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of this application is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of this application.

[0124] The functional blocks shown in the above block diagram can be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, they can be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of this application are programs or code segments used to perform the required tasks. Programs or code segments can be stored on a machine-readable medium or transmitted over a transmission medium or communication link via data signals carried on a carrier wave. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, radio frequency (RF) links, etc. Code segments can be downloaded via computer networks such as the Internet, intranets, etc.

[0125] It should also be noted that the exemplary embodiments mentioned in this application describe methods or systems based on a series of steps or apparatus. However, this application is not limited to the order of the above steps; that is, the steps can be performed in the order mentioned in the embodiments, or in a different order, or several steps can be performed simultaneously.

[0126] It should be clarified that this application is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of this application is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of this application.

[0127] The functional blocks shown in the above-described structural diagram can be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, they can be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of this application are programs or code segments used to perform the required tasks. Programs or code segments can be stored on a machine-readable medium or transmitted over a transmission medium or communication link via data signals carried on a carrier wave. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, radio frequency (RF) links, etc. Code segments can be downloaded via computer networks such as the Internet, intranets, etc.

[0128] It should also be noted that the exemplary embodiments mentioned in this application describe methods or systems based on a series of steps or apparatus. However, this application is not limited to the order of the above steps; that is, the steps can be performed in the order mentioned in the embodiments, or in a different order, or several steps can be performed simultaneously.

[0129] The aspects of this disclosure have been described above with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this disclosure. It should be understood that each block in the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that these instructions, executable via the processor of the computer or other programmable data processing apparatus, enable the implementation of the functions / actions specified in one or more blocks of the flowchart illustrations and / or block diagrams. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field-programmable logic circuit. It is also understood that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can also be implemented by special-purpose hardware performing the specified functions or actions, or can be implemented by a combination of special-purpose hardware and computer instructions.

[0130] The above description is merely a specific implementation of this application. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, modules, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here. It should be understood that the protection scope of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the protection scope of this application.

Claims

1. A method for monitoring industrial robots, characterized in that, include: Obtain operational data of the target robot; The operational data is input into a preset digital twin model, which performs an operational simulation based on the operational data to obtain the remaining lifespan of each part of the target robot within a preset future time period. Based on the remaining lifespan, the simulation robot corresponding to the preset digital twin model is color-rendered to obtain the rendered robot; Based on the rendering robot, the target robot is monitored to obtain monitoring results; If the monitoring results meet the preset alarm conditions, an execution strategy for the preset future time period is generated.

2. The method according to claim 1, characterized in that, Before inputting the operational data into a preset digital twin model, and the preset digital twin model performing operational simulations according to the operational data to obtain the remaining lifespan of each part of the target robot within a preset future time period, the method further includes: Obtain the structural drawings and target scene of the target robot, wherein the target scene includes the working environment and production line of the target robot; Based on the aforementioned structural drawings, working environment, and workflow, a pre-defined digital twin model is constructed. The preset digital twin model is configured with: A real-time connection module, which is used to acquire the operating data of the target robot; The parsing drive module is used to parse the received running data, obtain parsed data, and update the motion state of each part of the preset digital twin model according to the parsed data.

3. The method according to claim 2, characterized in that, The step of inputting the operational data into a preset digital twin model, wherein the preset digital twin model performs operational simulation according to the operational data, to obtain the remaining lifespan of each part of the target robot within a preset future time period, includes: Feature extraction is performed on the operational data to obtain multi-dimensional operational features; The multi-dimensional operational features are fused to obtain the fused health index of each part of the target robot. Based on the fused health indicators, curve fitting is performed according to a preset time series to obtain the fitted degradation trajectory; Based on the fitted degradation trajectory and the preset threshold, the remaining lifespan of each part is determined.

4. The method according to claim 3, characterized in that, The determination of the remaining lifespan of each part based on the fitted degradation trajectory and a preset threshold includes: Obtain the intersection points of the fitted degenerate trajectory and multiple preset thresholds; The remaining lifetime is determined based on the difference between the intersection point and the current time.

5. The method according to claim 3, characterized in that, The multi-dimensional operational features include vibration features, current features, temperature features, and sound features. The feature fusion of these multi-dimensional operational features to obtain the fused health indicators for each part of the target robot includes: The multi-dimensional operational features are normalized to obtain standard multi-dimensional operational features; The standard multi-dimensional operating features are scored using a preset fault symptom association matrix to obtain association scores. The preset fault symptom association matrix is ​​used to calculate the degree of association between each multi-dimensional operating feature and the fault. Based on the correlation score and the preset correlation threshold, strong correlation features are selected from multiple standard multi-dimensional operating features. The strongly correlated features were filtered using a preset principal component analysis algorithm to obtain multiple key fault features; Multiple key fault features are merged into vectors to obtain the vector to be diagnosed; The vector to be diagnosed is input into a preset diagnostic model to obtain a fused health index.

6. The method according to claim 5, characterized in that, The step of monitoring the target robot based on the rendering robot and obtaining monitoring results includes: Based on the color rendering values ​​of each part in the rendering robot within a preset time period, the probability of failure of the rendering robot in a preset future time period is determined. Based on the probability of the fault occurring, generate fault warning information; The aforementioned fault warning information is identified as a monitoring result.

7. The method according to claim 1, characterized in that, The step of color rendering the simulation robot corresponding to the preset digital twin model based on the remaining lifespan to obtain the rendered robot includes: Based on the remaining lifespan, determine the lifespan rating of each component; Calculate the color rendering value for each part based on the lifespan level and fusion health index; The simulated robot is rendered using the color rendering values ​​to obtain the rendered robot.

8. An industrial robot monitoring device, characterized in that, The device includes: The acquisition module is used to acquire the operational data of the target robot; An input module is used to input the running data into a preset digital twin model, which performs a running simulation according to the running data to obtain the remaining lifespan of each part of the target robot in a preset future time period. The color rendering module is used to perform color rendering on the simulation robot corresponding to the preset digital twin model based on the remaining lifespan, so as to obtain the rendered robot. The monitoring module is used to monitor the target robot based on the rendering robot and obtain monitoring results; The generation module is used to generate the execution strategy for the preset future time period when the monitoring results meet the preset alarm conditions.

9. An electronic device, characterized in that, The device includes: a processor and a memory storing computer program instructions; When the processor executes the computer program instructions, it implements the industrial robot monitoring method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer program instructions, which, when executed by a processor, implement the industrial robot monitoring method as described in any one of claims 1-7.

Images