CN115674272A - Robot fault diagnosis method, device, equipment and storage medium - Google Patents
Robot fault diagnosis method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN115674272A CN115674272A CN202211370397.4A CN202211370397A CN115674272A CN 115674272 A CN115674272 A CN 115674272A CN 202211370397 A CN202211370397 A CN 202211370397A CN 115674272 A CN115674272 A CN 115674272A
- Authority
- CN
- China
- Prior art keywords
- fault diagnosis
- robot
- torque
- fault
- information
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Manipulator (AREA)
Abstract
The application discloses a robot fault diagnosis method, a device, equipment and a storage medium, which relate to the field of industrial intelligent manufacturing, and the method comprises the following steps: acquiring torque information of the robot; inputting the torque information into a trained fault diagnosis model to obtain fault type information output by the fault diagnosis model; wherein the fault diagnosis model is an LSTM-Resnet hybrid deep learning model. The method and the device improve the efficiency of determining the fault information of the robot.
Description
Technical Field
The application relates to the field of industrial intelligent manufacturing, in particular to a robot fault diagnosis method, device, equipment and storage medium.
Background
In the correlation technique, the application of industrial robot will be promoted from traditional manufacturing industry to other manufacturing industries, and industrial robot can produce with high efficiency around the clock, therefore, how to make industrial robot keep high-efficient operating condition receives the general attention of researcher. Currently, important factors affecting the work of an industrial robot include industrial robot failure.
The current existing methods for solving the faults of the industrial robot are as follows: a large amount of human resources are invested to complete daily, weekly and monthly inspection and maintenance of the industrial robot, an inspection table is patrolled according to the final equipment working state of the inspection and maintenance recording stroke, an equipment maintenance manual of the industrial robot is formed according to the inspection and maintenance recording stroke, and equipment parameters of the industrial robot in a fault state are summarized. After data analysis is carried out, the fault occurrence frequency of each device, the fault rule and the fault reason are obtained, and practical experience is accumulated for later fault condition treatment; or a robot remote service platform is developed, the working state of the robot is evaluated through remote monitoring and real-time data recording of the industrial robot working in an enterprise workshop, and corresponding technical support of fault maintenance and daily maintenance of the industrial robot is provided for a client; or the independently established TCP/IP network is used for carrying out data communication with the industrial robot participating in workshop production to assist the customer maintenance personnel to complete the fault problem processing of the robot.
However, the above-described way of solving the problem of failure of an industrial robot is inefficient in determining the failure information of the robot.
Content of application
The main purpose of the present application is to provide a fault diagnosis method, apparatus, device and storage medium, which aim to solve the technical problem that the efficiency of determining fault information of a robot is too low.
In a first aspect, to achieve the above object, the present application provides a robot fault diagnosis method, including:
acquiring torque information of the robot;
inputting the torque information into a trained fault diagnosis model to obtain fault type information output by the fault diagnosis model; the fault diagnosis model is an LSTM-Resnet hybrid deep learning model.
Optionally, after acquiring the torque information of the robot, the method further includes:
controlling a preset sample segmentation frame to perform multiple translations in the torque vibration information scale according to a preset movement scale, and intercepting a plurality of sample torque data from the torque information; the preset moving scale is smaller than the torque vibration information ruler of the torque information and smaller than the scale of the preset sample segmentation frame;
inputting the torque information into a trained fault diagnosis model to obtain fault type information output by the fault diagnosis model; wherein, the fault diagnosis model is an LSTM-Resnet hybrid deep learning model, and comprises the following steps:
inputting the sample torque data into a trained fault diagnosis model to obtain fault type information output by the fault diagnosis model; wherein the fault diagnosis model is an LSTM-Resnet hybrid deep learning model.
Optionally, the fault diagnosis model includes a preprocessing network, configured to preprocess the torque information to obtain a torque fault signal;
the preprocessing network comprises a first convolution layer, a first pooling layer, a second convolution layer and a second pooling layer which are connected in sequence;
the first convolution layer and the second convolution layer are both one-dimensional convolution layers with an inclusion structure, and the first pooling layer and the second pooling layer are both batch standard layers.
Optionally, the preprocessing network is further configured to perform denoising and normalization processing on the torque information to obtain a torque fault signal.
Optionally, the fault diagnosis model includes a feature extraction network, configured to perform feature extraction on the torque fault signal to obtain a torque vibration signal feature;
the feature extraction network is FB-LSTM Resnet network, and the feature extraction layer in the feature extraction network is bidirectional LSTM meta-feature extraction layer.
Optionally, the fault diagnosis model includes a fault diagnosis network, configured to perform fault diagnosis on the torque vibration signal characteristic to obtain fault type information;
the fault diagnosis module comprises a global pooling layer and an extreme learning machine layer which are connected in sequence;
the global pooling layer is used for flattening the torque vibration signal to obtain a flattened torque vibration signal;
and the extreme learning machine layer is used for outputting fault type information according to the flattened torque vibration signal.
Optionally, the extreme learning machine is further configured to output the optimal parameter.
After inputting the torque information into the trained fault diagnosis model, the method further comprises:
and updating the fault diagnosis model according to the optimal parameters to obtain an updated fault diagnosis model.
In a second aspect, the present application also provides a robot fault diagnosis device, including:
the acquisition module is used for acquiring the torque information of the robot;
the input module is used for inputting the torque information into the trained fault diagnosis model to obtain fault type information output by the fault diagnosis model; the fault diagnosis model is an LSTM-Resnet hybrid deep learning model.
In a third aspect, the present application provides a robot fault diagnosis apparatus including: the robot fault diagnosis system comprises a processor, a memory and a robot fault diagnosis program stored in the memory, wherein the robot fault diagnosis program realizes the steps of the robot fault diagnosis method according to the first aspect when being executed by the processor.
In a fourth aspect, the present application provides a computer-readable storage medium having a robot fault diagnosis program stored thereon, the robot fault diagnosis program implementing the robot fault diagnosis method according to the first aspect when executed by a processor.
The application provides a fault diagnosis method of a robot, which comprises the steps of obtaining torque information of the robot; inputting the torque information into a trained fault diagnosis model to obtain fault type information output by the fault diagnosis model; wherein the fault diagnosis model is an LSTM-Resnet hybrid deep learning model.
Therefore, the fault information of the robot can be rapidly obtained by inputting the torque information of the robot into the LSTM-Resnet mixed deep learning model, and the efficiency of determining the fault information of the robot is improved.
Drawings
Fig. 1 is a schematic diagram of an architecture of a robot fault diagnosis system of the robot fault diagnosis method of the present application;
fig. 2 is a schematic structural diagram of a robot fault diagnosis device in a hardware operating environment according to the robot fault diagnosis method of the present application;
fig. 3 is a schematic flowchart of a first embodiment of a robot fault diagnosis method provided in the present application;
fig. 4 is a schematic flowchart of a robot fault diagnosis method according to a second embodiment of the present disclosure;
fig. 5 is a schematic structural diagram of a preprocessing network of a robot fault diagnosis method provided in the present application;
fig. 6 is a schematic diagram of a signal acquisition and processing module of a robot fault diagnosis method provided in the present application;
FIG. 7 is a schematic diagram of a fault feature extraction module of a robot fault diagnosis method provided in the present application;
FIG. 8 is a schematic diagram of a fault diagnosis module of a robot fault diagnosis method provided in the present application;
fig. 9 is a block diagram of the robot fault diagnosis apparatus according to the present invention.
The implementation, functional features and advantages of the objectives of the present application will be further explained with reference to the accompanying drawings.
Detailed Description
It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
In the correlation technique, industrial robot's application will be promoted to other manufacturing industries from traditional manufacturing industry, and industrial robot can produce with high efficiency around the clock, consequently, how to make industrial robot keep high-efficient operating condition receives the general attention of researcher. Currently, important factors affecting the work of an industrial robot include industrial robot failure.
The current existing methods for solving the faults of the industrial robot are as follows: a large amount of human resources are invested to complete daily, weekly and monthly inspection and maintenance of the industrial robot, an inspection table is patrolled according to the final equipment working state of the inspection and maintenance recording stroke, an equipment maintenance manual of the industrial robot is formed according to the inspection and maintenance recording stroke, and equipment parameters of the industrial robot in a fault state are summarized. After data analysis is carried out, the fault occurrence frequency of each device, the fault rule and the fault reason are obtained, and practical experience is accumulated for later fault condition treatment; or a robot remote service platform is developed, the working state of the robot is evaluated through remote monitoring and real-time data recording of the industrial robot working in an enterprise workshop, and corresponding technical support of fault maintenance and daily maintenance of the industrial robot is provided for a client; or the independently established TCP/IP network is used for carrying out data communication with the industrial robot participating in workshop production to assist the customer maintenance personnel to complete the fault problem processing of the robot.
However, the above-described way of solving the problem of failure of an industrial robot is inefficient in determining the failure information of the robot.
The application provides a fault diagnosis method of a robot, which comprises the steps of obtaining torque information of the robot; inputting the torque information into a trained fault diagnosis model to obtain fault type information output by the fault diagnosis model; wherein the fault diagnosis model is an LSTM-Resnet hybrid deep learning model.
Therefore, the fault information of the robot can be rapidly obtained by inputting the torque information of the robot into the LSTM-Resnet mixed deep learning model, and the efficiency of determining the fault information of the robot is improved.
In the following, a multimedia playing system applied in the implementation of the present application will be described:
referring to fig. 1, fig. 1 is a schematic diagram of an architecture of a robot fault diagnosis system according to an exemplary embodiment. As shown in fig. 1, the robot fault diagnosis system may include a server 11, a network 12, and a robot fault diagnosis device 13.
The server 11 may be a physical server comprising a separate host, or the server 11 may be a virtual server carried by a cluster of hosts. During operation, the server 11 may run a server-side program of an application to implement a related business function of the application, for example, when the robot fault diagnosis device 13 acquires the torque information of the robot, the server 11 may serve as the server for acquiring the torque information of the robot to support the robot fault diagnosis device 13 to complete the task of acquiring the torque information of the robot.
The network 12 may include various types of wired or wireless networks. In one embodiment, the Network 12 may include the Public Switched Telephone Network (PSTN) and the Internet. The robot fault diagnosis device 13 may interact with the server 11 through the network 12.
The robot fault diagnosis device 13 may include electronic devices of the types such as: workstations, smart phones, tablets, laptops, palmtops (PDAs), etc., to which one or more embodiments of the present disclosure are not limited.
Referring to fig. 2, fig. 2 is a schematic structural diagram of a robot fault diagnosis device in a hardware operating environment according to an embodiment of the present application.
As shown in fig. 2, the robot fault diagnosis apparatus may include: a processor 1001, such as a Central Processing Unit (CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. Wherein a communication bus 1002 is used to enable connective communication between these components. The user interface 1003 may include a Display screen (Display), an input unit such as a Keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface, a wireless interface. The network interface 1004 may optionally include a standard wired interface, a WIreless interface (e.g., a WIreless-FIdelity (WI-FI) interface). The Memory 1005 may be a Random Access Memory (RAM) Memory, or may be a Non-Volatile Memory (NVM), such as a disk Memory. The memory 1005 may alternatively be a storage device separate from the processor 1001.
Those skilled in the art will appreciate that the configuration shown in fig. 2 does not constitute a limitation of the cast terminal and may include more or fewer components than those shown, or some components may be combined, or a different arrangement of components.
As shown in fig. 2, the memory 1005, which is a storage medium, may include therein an operating system, a data storage module, a network communication module, a user interface module, and a robot fault diagnosis program.
In the robot failure diagnosis apparatus shown in fig. 2, the network interface 1004 is mainly used for data communication with a network server; the user interface 1003 is mainly used for data interaction with a user; the processor 1001 and the memory 1005 of the robot fault diagnosis device of the present application may be provided in the robot fault diagnosis device, and the robot fault diagnosis device calls the robot fault diagnosis program stored in the memory 1005 through the processor 1001 and executes the robot fault diagnosis method provided by the embodiment of the present application.
Based on the above hardware structure of the robot fault diagnosis device but not limited to the above hardware structure, the present application provides a first embodiment of robot fault diagnosis. Referring to fig. 3, fig. 3 shows a schematic flow chart of the first embodiment for applying robot fault diagnosis.
It should be noted that, although a logical order is shown in the flow chart, in some cases, the steps shown or described may be performed in an order different than that shown or described herein.
In this embodiment, the robot fault diagnosis method includes:
s10, acquiring torque information of the robot;
the execution subject of the present embodiment is a robot failure diagnosis apparatus.
It should be understood that the robot fault diagnosis apparatus includes a torque sensor, and torque information of the robot is acquired through the torque sensor.
S20, inputting the torque information into the trained fault diagnosis model to obtain fault type information output by the fault diagnosis model; wherein the fault diagnosis model is an LSTM-Resnet hybrid deep learning model.
It should be understood that the robot fault diagnosis device inputs the torque information into the trained fault diagnosis model, and obtains the fault type information output by the fault diagnosis model. The fault diagnosis model is an LSTM-Resnet hybrid deep learning model.
In the embodiment, the current mode for positioning the fault of the industrial robot comprises the steps of manually positioning the fault of the robot, manually checking the fault position of the robot after the fault of the robot and the fault of the robot are determined manually according to monitoring data of the robot, and by inputting the torque information of the robot into the LSTM-Resnet mixed deep learning model, the fault information of the robot can be obtained rapidly, so that the efficiency of determining the fault information of the robot is improved.
Based on the above embodiment, the present application also provides a second embodiment of a robot fault diagnosis method. Fig. 4 is a flowchart illustrating a robot fault diagnosis method according to a second embodiment of the present disclosure.
In this embodiment, after the torque information of the robot is acquired, the method further includes:
s101, controlling a preset sample segmentation frame to translate for multiple times in the torque vibration information scale according to a preset movement scale, and intercepting multiple sample torque data from the torque information; the preset moving scale is smaller than the torque vibration information ruler of the torque information and smaller than the scale of the preset sample segmentation frame;
it should be understood that the preset movement scale can be set by the user according to actual conditions. The torque vibration information scale is length information of the torque information, and the preset moving scale is smaller than the torque vibration information scale of the torque information and smaller than the scale of the preset sample segmentation frame.
According to the preset movement scale and the torque vibration information scale of the torque information, the preset sample segmentation frame is controlled to translate in the torque vibration information scale, namely a vibration signal window translation method. Because the interval sampling can not represent the condition of all vibration signals, the robot fault diagnosis equipment adopts a vibration signal window translation method to segment the torque information sample, and the obtained torque data of the extended sample can maximally represent the condition of all torque data, thereby being beneficial to exploring the learning potential of a neural network.
The method for dividing the torque information samples by adopting the vibration signal window translation method can save the continuity existing among the time sequence vibration signals, and can avoid that the equal-interval sampling can not represent all the vibration signals. The robot fault diagnosis equipment adopts a vibration signal window translation method, can finish signal segmentation, and obtains sample torque data.
Specifically, in one example, as for vibration information in torqueDimension L z For the torque information, a block size L is used y The preset sample segmentation frame is used for carrying out vibration signal window translation, wherein the step length of each translation, namely the moving scale, is p. Therefore, after the preset sample segmentation frame is translated every time, the preset sample segmentation frame can intercept a section of sample torque data from the torque information.
At this time, it can be understood that the number of sample torque data that can be obtained is E, and the E sample torque data constitutes a sample set. E is calculated according to equation 1.
The formula 1 is:
wherein a "-" represents a calculation of a floor rounding.
It can be understood that, since the preset movement scale is smaller than the torque vibration information scale of the torque information and smaller than the scale of the preset sample segmentation frame, the sample torque data intercepted in the two adjacent translations has an overlapping part, and the scale of the overlapping signal segment is L y -p. In this case, the torque information is processed by the vibration signal window shifting method, and the sample torque data of the obtained sample set is increased by α times. α can be calculated according to equation 2:
the formula 2 is:
wherein, the position of the ith sample torque data in the sample set in the torque signal can be expressed as:
in the embodiment, the torque information is input into a trained fault diagnosis model, and fault type information output by the fault diagnosis model is obtained; wherein, the fault diagnosis model is an LSTM-Resnet hybrid deep learning model, and the adaptability is changed into:
s102, inputting the sample torque data into a trained fault diagnosis model to obtain fault type information output by the fault diagnosis model; wherein the fault diagnosis model is an LSTM-Resnet hybrid deep learning model.
The robot fault diagnosis equipment inputs the torque data of the expansion sample into a trained fault diagnosis model to obtain fault type information output by the fault diagnosis model; wherein the fault diagnosis model is an LSTM-Resnet hybrid deep learning model.
In the embodiment, the robot fault diagnosis equipment can maximally represent the conditions of all torque data by adopting a vibration signal window translation method, and is beneficial to exploring neural network learning potential. Meanwhile, the continuity existing among time sequence vibration signals can be stored by adopting a vibration signal window translation method to divide the torque information samples, and the condition that all vibration signals cannot be represented by equidistant sampling can be avoided.
Referring to fig. 5 as a specific implementation, fig. 5 is a schematic structural diagram of a preprocessing network of a robot fault diagnosis method provided in the present application, where a fault diagnosis model includes the preprocessing network, and is configured to preprocess torque information to obtain a torque fault signal;
the preprocessing network comprises a first convolution layer, a first pooling layer, a second convolution layer and a second pooling layer which are connected in sequence;
the first convolution layer and the second convolution layer are both one-dimensional convolution layers with an inclusion structure, and the first pooling layer and the second pooling layer are both batch standard layers.
It should be understood that the torque fault signal is a one-dimensional sample, and the robot fault diagnosis device completes the preprocessing of the torque fault signal according to the processing network, so that the signal noise can be effectively reduced. The preprocessing network comprises a first convolution layer, a first pooling layer, a second convolution layer and a second pooling layer which are sequentially connected, wherein a batch of standardization layers are added after the convolution layers, so that the convergence rate of the network is accelerated, the training precision is improved, and the reliability of subsequent processing is ensured.
In the embodiment, the robot fault diagnosis device completes the preprocessing of the torque fault signal according to the preprocessing network, and can effectively reduce the signal noise. In addition, the addition of the batch standardization layer after the convolution layer is beneficial to accelerating the convergence rate of the network and improving the training precision, thereby ensuring the reliability of subsequent processing.
As a specific embodiment, referring to fig. 6, fig. 6 is a schematic diagram of a signal acquiring and processing module of a robot fault diagnosis method provided in the present application.
And the preprocessing network is also used for carrying out denoising processing and normalization processing on the torque information to obtain a torque fault signal.
It is to be understood that the denoising technique is a technique for removing noise from a signal. All signal processing devices have characteristics that make them susceptible to noise. The noise may be random noise or white noise having a uniform frequency distribution, or may be frequency-dependent noise introduced by a device mechanism or a signal processing algorithm.
The normalization processing can map the torque information into a range of 0-1, and can process the torque information more conveniently and rapidly. The normalization processing is beneficial to improving the model precision and the convergence rate of the model.
In the embodiment, the torque fault signal is obtained by performing denoising and normalization processing on the torque information, which is beneficial to improving the accuracy of the fault diagnosis result.
Referring to fig. 7 as a specific implementation, fig. 7 is a schematic diagram of a fault feature extraction module of a robot fault diagnosis method provided in the present application. The fault diagnosis model comprises a feature extraction network, and is used for extracting the features of the torque fault signal to obtain the torque vibration signal features;
the feature extraction network is FB-LSTM Resnet network, and the feature extraction layer in the feature extraction network is bidirectional LSTM meta-feature extraction layer.
It should be understood that the fault feature extraction module includes a bidirectional LSTM element, an addition layer and a Dropput layer, where the first torque fault information sequentially enters the first bidirectional LSTM element, the first Dropout layer, the second bidirectional LSTM element and the second Dropout layer to obtain a first torque vibration signal feature; the second torque fault information sequentially enters a third bidirectional LSTM element and a third Dropout layer to obtain a second torque vibration information characteristic; and adding the first torque vibration signal characteristic and the second torque vibration information characteristic to obtain a torque vibration signal characteristic.
The FB-LSTM Resnet network is used for carrying out feature extraction on the torque fault signal to obtain the torque vibration signal feature, and the FB-LSTM Resnet network is composed of a bidirectional feature extraction layer consisting of LSTM elements and a fusion type residual error network. The bidirectional LSTM can comprehensively acquire the characteristics of the torque vibration signal from the front direction and the rear direction and adjust parameters in time. The adopted fusion type residual error structure can improve the training speed and effect and effectively process the degradation condition caused by the depth of layer on the premise of not increasing additional parameters and operation amount.
In this embodiment, the FB-LSTM Resnet network is adopted to facilitate optimization of data flow, thereby alleviating the problem of gradient dispersion, that is, solving the problem that the output falls into a function saturation region, and the gradient rapidly becomes small and effective learning cannot be achieved.
Referring to fig. 8 as a specific embodiment, fig. 8 is a schematic diagram of a fault diagnosis module of a robot fault diagnosis method provided in the present application. The fault diagnosis model comprises a fault diagnosis network and is used for carrying out fault diagnosis on the torque vibration signal characteristics to obtain fault type information;
the fault diagnosis module comprises a global pooling layer and an extreme learning machine layer which are connected in sequence;
the global pooling layer is used for flattening the torque vibration signal to obtain a flattened torque vibration signal;
and the extreme learning machine layer is used for outputting fault type information according to the torque vibration signal which is subjected to the flattening processing.
It is understood that in the prior art, each neuron of the fully-connected layer is connected with all neurons of the previous layer, and is used for fusing data acquired by the convolutional layer. The data volume of the full connection layer is large and complicated, the training rate of the model is lost, and an overfitting condition is easily caused. Therefore, the embodiment adopts the global pooling layer to carry out flattening processing on the torque vibration signal, the flattened torque vibration signal is obtained, the flattening processing of the vibration signal can be completed, and the network parameters are reduced, so that the overfitting condition is effectively avoided.
Based on the same application concept, the application also provides a third embodiment of the robot fault diagnosis method.
In this embodiment, the extreme learning machine is further configured to output an optimal parameter according to the flattened torque vibration signal.
In this embodiment, after step S20, the method further includes:
and S30, updating the fault diagnosis model according to the optimal parameters to obtain an updated fault diagnosis model.
It should be understood that the conventional Softmax (flexible maximum transfer function) method has difficulty in effectively applying the features obtained from FB-LSTM respet because it needs to adjust the layer in advance according to the classification target, so that the model obtained by training is not good in efficiency, accuracy and robustness, and the training time is also increased. And the optimal parameters can be obtained by only one training by adopting the extreme learning machine, and the extreme learning machine has good generalization.
Therefore, in each fault diagnosis process, the extreme learning machine can also output the optimal parameters, and then the fault diagnosis model is updated according to the optimal parameters to obtain the updated fault diagnosis model.
In the embodiment, the global pooling layer is adopted to replace the full connection layer, and the layer can complete flattening processing of the vibration signal and reduce network parameters, so that the overfitting condition is effectively avoided. The extreme learning machine is adopted to obtain relevant parameters through the set hidden layer stage, the weight of the output layer is obtained by utilizing a regular method, and the method has good generalization. Meanwhile, after the extreme learning machine is used for outputting the optimal parameters, the fault diagnosis model is updated according to the optimal parameters to obtain an updated fault diagnosis model, and the accuracy of the fault diagnosis model is improved.
Based on the same application concept, the application also provides a robot fault diagnosis device. Referring to fig. 5, fig. 5 is a schematic structural diagram of a robot fault diagnosis device provided in the present application, where the device specifically includes:
an obtaining module 500, configured to obtain torque information of a robot;
an input module 510, configured to input torque information into a trained fault diagnosis model, and obtain fault type information output by the fault diagnosis model; the fault diagnosis model is an LSTM-Resnet hybrid deep learning model.
According to the technical scheme of the embodiment, the torque information of the robot is acquired through mutual matching of all the functional modules; inputting the torque information into a trained fault diagnosis model to obtain fault type information output by the fault diagnosis model; wherein the fault diagnosis model is an LSTM-Resnet hybrid deep learning model.
Therefore, the fault information of the robot can be rapidly obtained by inputting the torque information of the robot into the LSTM-Resnet mixed deep learning model, and the efficiency of determining the fault information of the robot is improved.
In addition, the present application further provides a computer storage medium, where a robot fault diagnosis program is stored on the storage medium, and when executed by a processor, the method for diagnosing robot fault as above is implemented. Therefore, a detailed description thereof will be omitted. In addition, the beneficial effects of the same method are not described in detail. For technical details not disclosed in embodiments of the computer-readable storage medium referred to in the present application, reference is made to the description of embodiments of the method of the present application. It is determined that the program instructions may be deployed to be executed on one computing device or on multiple computing devices located at one site or distributed across multiple sites and interconnected by a communication network, as examples.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by a computer program, which can be stored in a computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. The storage medium may be a magnetic disk, an optical disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), or the like.
It should be noted that the above-described embodiments of the apparatus are merely schematic, where units illustrated as separate components may or may not be physically separate, and components illustrated as units may or may not be physical units, may be located in one place, or may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. In addition, in the drawings of the embodiments of the apparatus provided in the present application, the connection relationship between the modules indicates that there is a communication connection therebetween, and may be implemented as one or more communication buses or signal lines. One of ordinary skill in the art can understand and implement it without inventive effort.
Through the above description of the embodiments, those skilled in the art will clearly understand that the present application can be implemented by software plus necessary general-purpose hardware, and certainly can also be implemented by special-purpose hardware including special-purpose integrated circuits, special-purpose CPUs, special-purpose memories, special-purpose components and the like. Generally, functions performed by computer programs can be easily implemented by corresponding hardware, and specific hardware structures for implementing the same functions may be various, such as analog circuits, digital circuits, or dedicated circuits. However, for the present application, the implementation of a software program is more preferable. Based on such understanding, the technical solutions of the present application may be embodied in the form of a software product, where the computer software product is stored in a readable storage medium, such as a floppy disk, a usb disk, a removable hard disk, a Read-only memory (ROM), a random-access memory (RAM), a magnetic disk or an optical disk of a computer, and includes instructions for enabling a computer device (which may be a personal computer, a server, or a network device) to execute the methods of the embodiments of the present application.
The above description is only a preferred embodiment of the present application, and not intended to limit the scope of the present application, and all modifications of equivalent structures and equivalent processes, which are made by the contents of the specification and the drawings of the present application, or which are directly or indirectly applied to other related technical fields, are included in the scope of the present application.
Claims (10)
1. A method of diagnosing a fault in a robot, the method comprising:
acquiring torque information of the robot;
inputting the torque information into a trained fault diagnosis model to obtain fault type information output by the fault diagnosis model; wherein the fault diagnosis model is an LSTM-Resnet hybrid deep learning model.
2. The robot fault diagnosis method according to claim 1, wherein after the torque information of the robot is acquired, the method further comprises:
controlling a preset sample segmentation frame to perform multiple translations in the torque vibration information scale according to a preset movement scale, and intercepting a plurality of sample torque data from the torque information; the preset movement scale is smaller than the torque vibration information scale of the torque information and smaller than the scale of the preset sample segmentation frame;
inputting the torque information into a trained fault diagnosis model to obtain fault type information output by the fault diagnosis model; wherein, the fault diagnosis model is an LSTM-Resnet hybrid deep learning model, and comprises the following steps:
inputting the sample torque data into a trained fault diagnosis model to obtain fault type information output by the fault diagnosis model; wherein the fault diagnosis model is an LSTM-Resnet hybrid deep learning model.
3. The robot fault diagnosis method according to claim 1, wherein the fault diagnosis model includes a preprocessing network for preprocessing the torque information to obtain a torque fault signal;
the preprocessing network comprises a first convolution layer, a first pooling layer, a second convolution layer and a second pooling layer which are connected in sequence;
the first convolution layer and the second convolution layer are both one-dimensional convolution layers with Inceptation structures, and the first pooling layer and the second pooling layer are both batch standard layers.
4. The robot fault diagnosis method according to claim 3, wherein the preprocessing network is further configured to perform denoising and normalization processing on the torque information to obtain the torque fault signal.
5. The robot fault diagnosis method according to claim 1, wherein the fault diagnosis model includes a feature extraction network for performing feature extraction on the torque fault signal to obtain a torque vibration signal feature;
the feature extraction network is a FB-LSTM Resnet network, and a feature extraction layer in the feature extraction network is a bidirectional LSTM meta-feature extraction layer.
6. The robot fault diagnosis method according to claim 5, wherein the fault diagnosis model includes a fault diagnosis network for performing fault diagnosis on the torque vibration signal characteristics to obtain fault type information;
the fault diagnosis module comprises a global pooling layer and an extreme learning machine layer which are connected in sequence;
the global pooling layer is used for flattening the torque vibration signal to obtain a flattened torque vibration signal;
and the extreme learning machine layer is used for outputting fault category information according to the flattened torque vibration signal.
7. The robot fault diagnosis method according to claim 6, wherein the extreme learning machine is further configured to output an optimum parameter;
after the inputting the torque information into the trained fault diagnosis model, the method further comprises:
and updating the fault diagnosis model according to the optimal parameters to obtain an updated fault diagnosis model.
8. A robot fault diagnosis apparatus characterized by comprising:
the acquisition module is used for acquiring the torque information of the robot;
the input module is used for inputting the torque information into a trained fault diagnosis model to obtain fault type information output by the fault diagnosis model; the fault diagnosis model is an LSTM-Resnet hybrid deep learning model.
9. A robot malfunction diagnosis apparatus characterized by comprising: a processor, a memory and a robot fault diagnosis program stored in the memory, the robot fault diagnosis program when executed by the processor implementing the steps of the robot fault diagnosis method according to any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that a robot fault diagnosis program is stored thereon, which when executed by a processor implements the robot fault diagnosis method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211370397.4A CN115674272A (en) | 2022-11-03 | 2022-11-03 | Robot fault diagnosis method, device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211370397.4A CN115674272A (en) | 2022-11-03 | 2022-11-03 | Robot fault diagnosis method, device, equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115674272A true CN115674272A (en) | 2023-02-03 |
Family
ID=85047610
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211370397.4A Pending CN115674272A (en) | 2022-11-03 | 2022-11-03 | Robot fault diagnosis method, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115674272A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117390571A (en) * | 2023-12-11 | 2024-01-12 | 深圳市潼芯传感科技有限公司 | Fault removal method and system based on industrial equipment |
| CN117953588A (en) * | 2024-03-26 | 2024-04-30 | 南昌航空大学 | Badminton player action intelligent recognition method integrating scene information |
-
2022
- 2022-11-03 CN CN202211370397.4A patent/CN115674272A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117390571A (en) * | 2023-12-11 | 2024-01-12 | 深圳市潼芯传感科技有限公司 | Fault removal method and system based on industrial equipment |
| CN117390571B (en) * | 2023-12-11 | 2024-03-29 | 深圳市潼芯传感科技有限公司 | Fault removal method and system based on industrial equipment |
| CN117953588A (en) * | 2024-03-26 | 2024-04-30 | 南昌航空大学 | Badminton player action intelligent recognition method integrating scene information |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115674272A (en) | Robot fault diagnosis method, device, equipment and storage medium | |
| CN108229591B (en) | Neural network adaptive training method and apparatus, device, program, and storage medium | |
| WO2021151336A1 (en) | Road image target detection method based on attentional mechanism and related device | |
| CN114998695A (en) | Method and system for improving image recognition speed | |
| CN111507985A (en) | Image instance segmentation optimization processing method and device and computer equipment | |
| CN109982088B (en) | Image processing method and device | |
| CN111325697B (en) | Color image restoration method based on tensor eigen transformation | |
| CN114266894A (en) | Image segmentation method and device, electronic equipment and storage medium | |
| CN111340025A (en) | Character recognition method, character recognition device, computer equipment and computer-readable storage medium | |
| Yang et al. | Unsupervised blind image quality assessment via joint spatial and transform features | |
| CN109902720B (en) | Image classification and identification method for depth feature estimation based on subspace decomposition | |
| CN116680401A (en) | Document processing method, document processing device, apparatus and storage medium | |
| CN114882273B (en) | Visual identification method, device, equipment and storage medium applied to narrow space | |
| WO2022257433A1 (en) | Processing method and apparatus for feature map of image, storage medium, and terminal | |
| CN115546554A (en) | Sensitive image identification method, device, equipment and computer readable storage medium | |
| CN113065585B (en) | Training method and device of image synthesis model and electronic equipment | |
| CN114820755A (en) | Depth map estimation method and system | |
| CN111192252B (en) | Image segmentation result optimization method and device, intelligent terminal and storage medium | |
| CN115147469A (en) | Registration method, device, equipment and storage medium | |
| CN114494856A (en) | Equipment model detection model training method and equipment model detection method | |
| CN113688931A (en) | Ultrasonic image screening method and device based on deep learning | |
| CN111899239A (en) | Image processing method and device | |
| CN113610856A (en) | Method and device for training image segmentation model and image segmentation | |
| CN110955369A (en) | Focus judgment method, device and equipment based on click position and storage medium | |
| CN111539420A (en) | Panoramic image saliency prediction method and system based on attention perception features |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |