CN119915503B - Linear guide rail fault diagnosis method and system based on deep learning - Google Patents
Linear guide rail fault diagnosis method and system based on deep learningInfo
- Publication number
- CN119915503B CN119915503B CN202510379438.3A CN202510379438A CN119915503B CN 119915503 B CN119915503 B CN 119915503B CN 202510379438 A CN202510379438 A CN 202510379438A CN 119915503 B CN119915503 B CN 119915503B
- Authority
- CN
- China
- Prior art keywords
- fault
- guide rail
- linear guide
- data
- model
- 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.)
- Active
Links
Landscapes
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The invention relates to the technical field of linear guide rail operation and maintenance management, in particular to a linear guide rail fault diagnosis method and system based on deep learning, which can improve the stability and production efficiency of equipment and reduce maintenance cost; the method comprises the steps of obtaining first operation state data of a target linear guide rail and second operation state data of at least one relevant linear guide rail, inputting the first operation state data into a preset guide rail fault evaluation model to obtain a first fault evaluation result, respectively inputting at least one group of second operation state data into the guide rail fault evaluation model to obtain at least one group of second fault evaluation result, correcting the first fault evaluation result according to the at least one group of second fault evaluation result by considering the influence of homologous faults, and obtaining a target linear guide rail fault diagnosis result.
Description
Technical Field
The invention relates to the technical field of linear guide rail operation and maintenance management, in particular to a method and a system for diagnosing a linear guide rail fault based on deep learning.
Background
In a linear guide rail system, fault diagnosis is a key link for ensuring stable operation and maintenance production efficiency of mechanical equipment, the linear guide rail is used as a precision mechanical part and widely applied to various automatic equipment such as a digital machine tool, an industrial robot and the like, and various faults are easy to occur in the linear guide rail due to long-term bearing of load, friction and vibration, so that the precision and the stability of the equipment are affected, the equipment is stopped and the production loss is caused.
The existing linear guide rail fault diagnosis method is mostly carried out on single equipment or components, the mutual relevance and fault propagation characteristics among the equipment in the actual industrial scene are ignored, particularly in a linear guide rail system, adjacent linear guide rails are often under similar operation environments and working conditions, the fault of one guide rail can generate homologous fault influence on the adjacent guide rail, and the homologous fault influence increases the complexity of fault diagnosis, so that the conventional method is difficult to accurately judge the true source and degree of the fault.
Disclosure of Invention
In order to solve the technical problems, the invention provides a linear guide rail fault diagnosis method and system based on deep learning, which can improve the stability and production efficiency of equipment and reduce the maintenance cost.
In a first aspect, the present invention provides a linear guide rail fault diagnosis method based on deep learning, the method comprising:
acquiring first running state data of a target linear guide rail and second running state data of at least one relevant linear guide rail;
Inputting the first running state data into a preset guide rail fault evaluation model to obtain a first fault evaluation result;
Respectively inputting at least one group of second running state data into the guide rail fault evaluation model to obtain at least one group of second fault evaluation results;
and correcting the first fault evaluation result according to at least one group of second fault evaluation results in consideration of the influence of homologous faults, and obtaining a target linear guide rail fault diagnosis result.
Further, the associated linear guide is a linear guide of the same type as the target linear guide and positioned adjacent thereto.
Further, the method for collecting the running state data comprises the following steps:
the linear guide rail and the adjacent linear guide rail are provided with various sensors for monitoring the running state of the linear guide rail in real time, wherein the sensors comprise a temperature sensor, an acceleration sensor, a vibration sensor and a noise sensor;
setting data acquisition frequency;
according to the data acquisition frequency, acquiring data in real time through a sensor arranged on the linear guide rail;
preprocessing the collected sensor data;
And storing the preprocessed sensor data into a preset database.
Further, the first fault assessment result includes a ball wear fault probability value, a rail corrosion fault probability value, a slider deformation fault probability value, and a slider end cap drop fault probability value.
Further, the method for constructing the guide rail fault evaluation model comprises the following steps:
Collecting operation state data of a historical linear guide rail, and preprocessing the collected data;
Selecting a deep learning model as an infrastructure of a guide rail fault evaluation model, wherein the deep learning model comprises a cyclic neural network and a convolutional neural network;
Dividing the preprocessed running state data of the historical linear guide rail into a training set, a verification set and a test set;
training the model by using the training set, and adjusting parameters of the model;
Verifying the model by using verification set data, and adjusting parameters and structures of the model according to verification results;
evaluating the model using the test set data;
and deploying the trained model into a production environment, receiving real-time data streams from the sensors, and performing online fault assessment.
Further, a mathematical calculation formula for correcting the first fault evaluation result according to at least one group of the second fault evaluation results is as follows: Wherein P c represents a corrected fault probability value, P t represents an original target rail fault probability value, alpha is an adjustment coefficient between 0 and 1 for balancing the proportion of original evaluation and homologous fault influence, n represents the number of associated rails, w i represents the weight of the ith associated rail, and P ai represents the fault probability value of the ith associated rail.
Further, the target linear guide rail fault diagnosis result comprises a corrected fault probability value, fault type confirmation, fault severity assessment, homologous fault influence analysis and maintenance advice.
In another aspect, the present application also provides a linear guide rail fault diagnosis system based on deep learning, the system comprising:
the data acquisition module acquires first running state data of the target linear guide rail and second running state data of at least one associated linear guide rail;
the guide rail fault evaluation module is used for inputting the first running state data into a preset guide rail fault evaluation model to obtain a first fault evaluation result;
The guide rail fault evaluation module is associated, at least one group of second running state data are respectively input into the guide rail fault evaluation model, and at least one group of second fault evaluation results are obtained;
And the fault diagnosis module is used for correcting the first fault evaluation result according to at least one group of second fault evaluation results in consideration of the influence of homologous faults, and obtaining a target linear guide rail fault diagnosis result.
In a third aspect, the present application provides an electronic device comprising a bus, a transceiver, a memory, a processor and a computer program stored on the memory and executable on the processor, the transceiver, the memory and the processor being connected by the bus, the computer program when executed by the processor implementing the steps of any of the methods described above.
In a fourth aspect, the application also provides a computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of any of the methods described above.
Compared with the prior art, the method has the advantages that the method not only evaluates the state of a single linear guide rail, but also analyzes the mutual influence between adjacent guide rails, can more accurately identify the influence of homologous faults by acquiring and processing the running state data of at least one related linear guide rail, thereby improving the diagnosis precision;
the accurate fault diagnosis is beneficial to planning maintenance activities in advance, so that maintenance work is more predictive and targeted, unexpected downtime is reduced, maintenance cost is reduced, and production efficiency is improved;
The deep learning model can process complex nonlinear relations, is suitable for different types of fault mode recognition, can keep good performance even under complex working condition environments such as long-term load, friction, vibration and the like, and can perform self optimization through continuous learning and training along with accumulation of more running state data, so that the accuracy and reliability of diagnosis of the preset guide rail fault evaluation model are improved along with time;
the whole diagnosis process from data acquisition to final fault diagnosis result generation is automatic, so that the requirement of manual intervention is greatly reduced, the response speed and the working efficiency of the system are improved, and the risk of human errors is reduced;
the detailed fault diagnosis result provided by the method can be directly used for guiding operation and maintenance personnel to make correct maintenance decisions, so that stable operation and efficient production of mechanical equipment are ensured; through accurate fault diagnosis, enterprises can avoid unnecessary comprehensive inspection and excessive maintenance, thereby saving manpower and material resources, and improving the availability and service life of equipment;
In summary, the linear guide rail fault diagnosis method based on deep learning overcomes the defects of the traditional fault diagnosis method, can improve the stability and production efficiency of equipment, and reduces maintenance cost.
Drawings
FIG. 1 is a flow chart of the present invention;
FIG. 2 is a flow chart of a method of constructing a rail fault assessment model;
FIG. 3 is a block diagram of a deep learning based linear guide rail fault diagnosis system;
FIG. 4 is a sample configuration diagram for four different operating conditions.
Detailed Description
In the description of the present application, those skilled in the art will appreciate that the present application may be embodied as methods, apparatus, electronic devices, and computer-readable storage media. Accordingly, the present application may be embodied in the form of hardware entirely, software (including firmware, resident software, micro-code, etc.), a combination of hardware and software. Furthermore, in some embodiments, the application may also be embodied in the form of a computer program product in one or more computer-readable storage media, which contain computer program code.
Any combination of one or more computer-readable storage media may be employed by the computer-readable storage media described above. The computer readable storage medium includes an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer readable storage medium include a portable computer diskette, a hard disk, a random access memory, a read-only memory, an erasable programmable read-only memory, a flash memory, an optical fiber, a compact disc read-only memory, an optical storage device, a magnetic storage device, or any combination thereof. In the context of this document, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, device.
The technical scheme of the application obtains, stores, uses, processes and the like the data, which all meet the relevant regulations of national laws.
The application provides a method, a device and electronic equipment through flow charts and/or block diagrams.
It will be understood that each block of the flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer-readable program instructions. These computer-readable program instructions may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions/acts specified in the flowchart and/or block diagram block or blocks.
These computer readable program instructions may also be stored in a computer readable storage medium that can cause a computer or other programmable data processing apparatus to function in a particular manner. Thus, instructions stored in a computer-readable storage medium produce an instruction means which implement the functions/acts specified in the flowchart and/or block diagram block or blocks.
The computer readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other devices to cause a series of operational steps to be performed on the computer, other programmable apparatus or other devices to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide processes for implementing the functions/acts specified in the flowchart and/or block diagram block or blocks.
The present application will be described below with reference to the drawings in the present application.
The first embodiment of the invention is a linear guide rail fault diagnosis method based on deep learning, as shown in fig. 1 to 2, specifically comprising the following steps:
S1, acquiring first running state data of a target linear guide rail and second running state data of at least one associated linear guide rail;
The first running state data and the second running state data have the same data content and at least comprise any one of temperature data, acceleration data, vibration data and noise data;
the method for collecting the running state data comprises the following steps:
The linear guide rail and the adjacent linear guide rail are provided with various sensors for monitoring the running state of the linear guide rail in real time, wherein the sensors comprise a temperature sensor, an acceleration sensor, a vibration sensor and a noise sensor;
The temperature sensor is used for monitoring temperature changes of the linear guide rail and the surrounding environment, and abnormal temperature rise is a sign of abrasion and insufficient lubrication;
the acceleration sensor is capable of detecting sudden movements or shocks due to friction or mechanical problems;
the vibration sensor is used for recording vibration modes generated when the equipment is operated, and is helpful for identifying mechanical defects or unbalance;
the noise sensor is capable of capturing abnormal sounds caused by a malfunction, such as abnormal friction sounds between metals or impact sounds;
Setting data acquisition frequency, and monitoring the change of long-term running state of the linear guide rail;
according to the data acquisition frequency, acquiring data in real time through a sensor arranged on a linear guide rail, so as to ensure timeliness and accuracy of the data;
preprocessing the collected sensor data to ensure the quality and usability of the data, wherein the preprocessing comprises the steps of data cleaning, data format conversion and data normalization;
And in order to ensure the safety and reliability of the data, taking data backup and recovery measures for the stored data through the database.
In the step, through acquiring first operation state data and second operation state data of a target linear guide rail and at least one associated linear guide rail, the comprehensive monitoring of the operation state of the linear guide rail system is realized, the operation state of the linear guide rail can be comprehensively reflected, potential problems can be found timely, due to the introduction of the associated linear guide rail, the problem can be more accurately positioned, when the target linear guide rail is abnormal, the problem of local problems or systematic problems can be more quickly judged by comparing the target linear guide rail with the data of the associated linear guide rail, thereby taking targeted maintenance measures, by installing various sensors on the linear guide rail and the adjacent linear guide rail and setting data acquisition frequency, the real-time monitoring of the operation state of the linear guide rail is realized, the abnormal conditions can be timely found, early warning is sent out in advance, the stable operation of equipment is guaranteed, the quality and the usability of the acquired sensor data are guaranteed, meanwhile, the data after the preprocessing is stored in a database, reliable data support is provided for subsequent analysis and processing, through data and data recovery, the safety and maintenance cost of the whole system can be further improved, the maintenance cost is further improved, the safety and the maintenance cost is further improved, the maintenance cost is prolonged, the whole maintenance cost is further, the maintenance cost is reduced, and the maintenance cost is further improved, and the real-time is guaranteed, and the maintenance cost is guaranteed through the performance is better.
S2, inputting the first running state data into a preset guide rail fault evaluation model to obtain a first fault evaluation result;
The first fault evaluation result comprises a ball wear fault probability value, a guide rail corrosion fault probability value, a sliding block deformation fault probability value and a sliding block end cover falling fault probability value;
The ball wear failure probability value is used to evaluate whether the ball is likely to be excessively worn;
the guide rail corrosion fault probability value is used for judging whether the probability of corrosion phenomenon exists on the surface of the guide rail or not;
the deformation fault probability value of the sliding block is used for detecting whether the sliding block has abnormal deformation or not;
The sliding block end cover falling fault probability value is used for predicting whether the sliding block end cover is loose or falls off;
the construction method of the guide rail fault evaluation model comprises the following steps:
Collecting historical linear guide rail running state data, including temperature data, acceleration data, vibration data and noise data;
Preprocessing the collected data, including data cleaning, data format conversion and data normalization, to ensure the quality and consistency of the data;
Extracting the characteristics of the running state data of the preprocessed historical linear guide rail, and extracting the characteristics capable of reflecting the faults of the linear guide rail;
Selecting a deep learning model as an infrastructure of a guide rail fault evaluation model, wherein the deep learning model comprises a cyclic neural network and a convolutional neural network;
The input layer is used for receiving the preprocessed data, the hidden layer is used for carrying out feature extraction and pattern recognition, and the output layer is used for outputting fault evaluation results;
Dividing the preprocessed running state data of the historical linear guide rail into a training set, a verification set and a test set, wherein the training set is used for training a model, the verification set is used for verifying and adjusting the model, and the test set is used for evaluating the performance of the model;
Training the model by using a training set, adjusting parameters of the model through a back propagation algorithm to minimize errors between a prediction result of the model and an actual fault type;
The model is verified by using verification set data, and parameters and structures of the model are adjusted according to verification results;
the model is evaluated by using test set data, and indexes such as accuracy, recall rate, F1 score and the like of the model are calculated;
and deploying the trained model into a production environment, receiving real-time data streams from the sensors, and performing online fault assessment.
In the step, the operation state data of the historical linear guide rail can be fully utilized through constructing a guide rail fault assessment model based on deep learning, comprehensive processing and analysis of information can be achieved, the model can accurately predict fault types such as ball abrasion, guide rail corrosion, sliding block deformation and sliding block end cover falling off, accuracy of fault prediction is improved, fault probability values output by the model provide clear fault early warning information for operation staff, the operation staff can take measures in advance to prevent faults or timely discover and process the faults at the initial stage of the faults, equipment shutdown and production loss caused by fault expansion are avoided, on-line fault assessment based on the model can enable the operation staff to know the operation state of the linear guide rail in real time, frequency and strength of manual inspection are reduced, meanwhile, the model can automatically identify fault types, time for judging fault causes by the operation staff is shortened, operation efficiency is improved, intelligent monitoring and fault prediction of the operation state are achieved through preventing faults, maintenance cost and production interruption cost caused by faults are reduced, intelligent operation state is improved through the introduction of the deep learning model, intelligent operation state is achieved, intelligent operation state monitoring and intelligent operation state prediction is achieved, and intelligent operation state fault prediction is achieved through the aid of the intelligent operation state prediction is achieved, and intelligent operation state prediction is improved.
S3, respectively inputting at least one group of second running state data into the guide rail fault evaluation model to obtain at least one group of second fault evaluation results, wherein the second fault evaluation results comprise a ball wear fault probability value, a guide rail corrosion fault probability value, a sliding block deformation fault probability value and a sliding block end cover falling fault probability value;
preprocessing the second running state data, including data cleaning, data format conversion and data normalization;
inputting the preprocessed second running state data into a preset guide rail fault evaluation model, wherein the model can automatically extract the characteristics in the data and predict the probability of fault types;
The model outputs at least one group of second fault evaluation results, wherein each group of results corresponds to one associated linear guide rail, and the results comprise a ball wear fault probability value, a guide rail corrosion fault probability value, a sliding block deformation fault probability value and a sliding block end cover falling fault probability value;
and recording and storing the output fault evaluation result in a database for subsequent analysis and comparison.
In the step, the operation conditions of at least one group of related linear guide rails can be comprehensively obtained by collecting and preprocessing second operation state data of the guide rails, the quality and consistency of the data are ensured through preprocessing steps such as data cleaning, format conversion and normalization, a reliable basis is provided for subsequent analysis, the preprocessed data are input into a preset guide rail fault evaluation model, the model is constructed based on deep learning, key features can be automatically extracted from complex data, probability prediction of fault types is performed, the efficiency and precision of fault diagnosis are greatly improved, the second fault evaluation result output by the model intuitively reflects the possibility that the related linear guide rails have corresponding fault types, clear fault early warning information is provided for operation staff, the operation staff can analyze the transmission and influence of homologous faults compared with the fault evaluation result of the target linear guide rails and the related linear guide rails, so that the true source and degree of faults can be accurately judged, the operation staff can make a more scientific and reasonable maintenance plan and fault processing strategy based on detailed fault probability evaluation and homologous fault influence analysis, and the utilization rate of equipment and the utilization of fault caused by faults is reduced.
S4, considering the influence of homologous faults, correcting the first fault evaluation result according to at least one group of second fault evaluation results, and obtaining a target linear guide rail fault diagnosis result;
analyzing a second fault evaluation result of the associated linear guide rail, and identifying whether a similar fault mode exists or not by comparing fault probability value distribution and fault type proportion of each associated guide rail;
Correcting a first fault evaluation result of the target linear guide rail according to the identified homologous fault mode;
if it is determined that the homologous fault exists and the fault probability value of the associated guide rail is higher than the corresponding fault probability value of the target guide rail, the fault probability value of the target guide rail is considered to be properly adjusted upwards so as to reflect the extra risk brought by the homologous fault;
Conversely, if the fault probability value of the associated rail is low and such low probability value is due to homologous factors, carefully evaluating the true fault condition of the target rail;
In the correction process, different weights are distributed to fault evaluation results of different associated guide rails, wherein the distribution of the weights is performed based on factors such as physical distance between the associated guide rail and a target guide rail, similarity of operating environment, difference of working conditions and the like;
correcting the fault evaluation result of the target guide rail by combining the fault probability value and the corresponding weight of each associated guide rail through a weighted average method;
The mathematical calculation formula for correcting the first fault evaluation result according to at least one group of the second fault evaluation results is as follows: Wherein P c represents a corrected fault probability value, P t represents an original target rail fault probability value, alpha is an adjustment coefficient between 0 and 1 for balancing the proportion of original evaluation and homologous fault influence, n represents the number of associated rails, w i represents the weight of the ith associated rail, and P ai represents the fault probability value of the ith associated rail;
the target linear guide rail fault diagnosis result comprises:
the corrected fault probability value gives the possibility of the fault of the target linear guide rail on each fault type, and the probability value already considers the influence of homologous faults and is more accurate than the original first fault evaluation result;
confirming the fault type, namely determining the fault type of the target linear guide rail which is about to happen according to the corrected fault probability value, facilitating operation and maintenance management personnel to quickly locate the fault point and taking corresponding maintenance measures;
the fault severity assessment, which is to combine the history data and experience, can carry out preliminary assessment on the severity of the fault, and is helpful for operation and maintenance management personnel to know the specific influence of the fault on the operation and production efficiency of the equipment;
In the correction process, the influence degree of the homologous faults on the target guide rail can be analyzed by comparing the fault probability values of the target linear guide rail and the associated linear guide rail;
and (3) maintenance advice, namely, based on a fault diagnosis result, an operation and maintenance manager can make targeted maintenance plans and advice.
In the step, the influence of the homologous faults on the fault probability value of the target guide rail can be identified and corrected through the comprehensive correlation of the fault evaluation result of the linear guide rail, so that the real fault condition of the target guide rail can be reflected more accurately, the accuracy and the reliability of fault diagnosis can be improved, potential faults can be predicted and identified earlier after the influence of the homologous faults is considered, sufficient time is provided for operation and maintenance managers to carry out maintenance and repair work, the fault expansion or equipment shutdown is avoided, the accurate fault diagnosis result provides scientific decision basis for the operation and maintenance managers, the operation and maintenance strategy and resource allocation can be optimized, the operation and maintenance cost is reduced, the production efficiency is improved, the stability and the reliability of the whole linear guide rail system are improved through timely correction and prediction of the faults, the production interruption and loss caused by the faults are reduced, the correction formula and the weight distribution method in the step have certain flexibility and the scalability, can be adjusted and optimized according to the actual scene and the characteristics of the linear guide rail system so as to adapt to different fault diagnosis industrial requirements, and the stability and the operation and the stability of the linear guide rail system can be obviously improved.
In the second embodiment, as shown in fig. 3, the linear guide rail fault diagnosis system based on deep learning of the invention specifically comprises the following modules;
the data acquisition module acquires first running state data of the target linear guide rail and second running state data of at least one associated linear guide rail;
the guide rail fault evaluation module is used for inputting the first running state data into a preset guide rail fault evaluation model to obtain a first fault evaluation result;
The guide rail fault evaluation module is associated, at least one group of second running state data are respectively input into the guide rail fault evaluation model, and at least one group of second fault evaluation results are obtained;
And the fault diagnosis module is used for correcting the first fault evaluation result according to at least one group of second fault evaluation results in consideration of the influence of homologous faults, and obtaining a target linear guide rail fault diagnosis result.
The system not only pays attention to the running state of the target linear guide rail, but also collects and analyzes the data of the linear guide rail related to the target linear guide rail, thereby being beneficial to capturing the influence of homologous faults and improving the accuracy of fault diagnosis;
The fault diagnosis module can judge the source and degree of the fault more accurately by introducing the influence analysis of the homologous fault, so that misdiagnosis or missed diagnosis possibly caused by neglecting the relevance among the devices is avoided;
With the collection and analysis of more running state data, the guide rail fault assessment model can perform self-optimization through continuous learning and training, thereby improving the accuracy and reliability of fault diagnosis along with the time;
The whole process from data acquisition to fault diagnosis result generation is automatic, reduces the requirement of manual intervention, improves the response speed and the working efficiency of the system, and reduces the risk of human errors;
the system provides a more intelligent, accurate and efficient fault diagnosis solution by integrating deep learning technology and understanding of interaction between linear guide rails, is beneficial to improving the stability and production efficiency of equipment and reducing maintenance cost.
In the third embodiment, as shown in fig. 4, taking the acceleration data of the target linear guide as an example, the steps of diagnosing the fault of the target linear guide include:
In order to simulate the real running environment of the linear guide rail and the state when faults occur, in the embodiment, each fault type is tested under four different working conditions;
The method comprises the steps of determining the sampling point number of each sensor, putting data acquired by the sensors into two channels, preprocessing the acquired data because vibration data acquired by the acceleration sensors often contain a large amount of noise and directly reduce the fault recognition rate of a model, decomposing signals into modes with different frequency bands, and effectively separating noise and useful information in the signals;
Dividing the converted data into data sets according to the fault category, wherein the dividing ratio is 8:1:1, the training set is used for iterative training of the model, the test set is used for evaluating the change of the accuracy of the model in the training process, and the verification set is used for evaluating the generalization effect of the model;
inputting data in a training set into a model for training, and determining model parameters by a verification set and a test set, wherein the model comprises a multi-channel extraction module, a feature screening module and a fault classification module;
Optimizing the mode number and penalty parameters, processing the original data, and selecting the mode function which is obtained and contains different information quantity and noise; when the modal function contains more fault characteristic information, the sparsity of the signal is stronger, the envelope entropy is smaller, and conversely, the envelope entropy is larger, so that the minimum envelope entropy is used as a standard for screening the modal function, and the screened three modal components are input into the model for training; the convolution layer extracts the characteristic information, the pooling layer reduces the parameter quantity calculated by the size reduction of the characteristic map, and residual error connection can relieve the problems of gradient disappearance and explosion when gradient counter-propagation is carried out, and can accelerate the training speed and improve the model performance;
and performing fault diagnosis tests under different working conditions by using the trained model.
The various modifications and embodiments of the deep learning-based linear guide fault diagnosis method in the first embodiment are equally applicable to the deep learning-based linear guide fault diagnosis system of the present embodiment, and those skilled in the art will be aware of the implementation method of the deep learning-based linear guide fault diagnosis system of the present embodiment through the foregoing detailed description of the deep learning-based linear guide fault diagnosis method, so that the details of the implementation method of the deep learning-based linear guide fault diagnosis system of the present embodiment will not be described in detail herein for brevity.
In addition, the application also provides an electronic device, which comprises a bus, a transceiver, a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the transceiver, the memory and the processor are respectively connected through the bus, and when the computer program is executed by the processor, the processes of the method embodiment for controlling output data are realized, and the same technical effects can be achieved, so that repetition is avoided and redundant description is omitted.
The foregoing is merely a preferred embodiment of the present invention, and it should be noted that it will be apparent to those skilled in the art that modifications and variations can be made without departing from the technical principles of the present invention, and these modifications and variations should also be regarded as the scope of the invention.
Claims (9)
1. A linear guide rail fault diagnosis method based on deep learning, the method comprising:
acquiring first running state data of a target linear guide rail and second running state data of at least one relevant linear guide rail;
Inputting the first running state data into a preset guide rail fault evaluation model to obtain a first fault evaluation result;
Respectively inputting at least one group of second running state data into the guide rail fault evaluation model to obtain at least one group of second fault evaluation results;
Correcting the first fault evaluation result according to at least one group of second fault evaluation results in consideration of the influence of homologous faults, and obtaining a target linear guide rail fault diagnosis result;
the associated linear guide rail is a linear guide rail which is the same as the target linear guide rail in type and adjacent in position;
In the correction process, different weights are distributed to the second fault evaluation results of different associated linear guide rails, and the distribution of the weights is determined based on the physical distance between the associated linear guide rails and the target linear guide rail, the similarity of the operating environment and the difference degree of working conditions.
2. The deep learning-based linear guide rail fault diagnosis method according to claim 1, wherein the operation state data acquisition method comprises:
the linear guide rail and the adjacent linear guide rail are provided with various sensors for monitoring the running state of the linear guide rail in real time, wherein the sensors comprise a temperature sensor, an acceleration sensor, a vibration sensor and a noise sensor;
setting data acquisition frequency;
according to the data acquisition frequency, acquiring data in real time through a sensor arranged on the linear guide rail;
preprocessing the collected sensor data;
And storing the preprocessed sensor data into a preset database.
3. The deep learning based linear rail fault diagnosis method according to claim 1, wherein the first fault assessment result comprises a ball wear fault probability value, a rail corrosion fault probability value, a slider deformation fault probability value, and a slider end cap drop fault probability value.
4. The deep learning-based linear guide rail fault diagnosis method according to claim 1, wherein the guide rail fault evaluation model construction method comprises the steps of:
Collecting operation state data of a historical linear guide rail, and preprocessing the collected data;
Selecting a deep learning model as an infrastructure of a guide rail fault evaluation model, wherein the deep learning model comprises a cyclic neural network and a convolutional neural network;
Dividing the preprocessed running state data of the historical linear guide rail into a training set, a verification set and a test set;
training the model by using the training set, and adjusting parameters of the model;
Verifying the model by using verification set data, and adjusting parameters and structures of the model according to verification results;
evaluating the model using the test set data;
and deploying the trained model into a production environment, receiving real-time data streams from the sensors, and performing online fault assessment.
5. The deep learning-based linear guide rail fault diagnosis method according to claim 1, wherein a mathematical calculation formula for correcting the first fault evaluation result according to at least one set of the second fault evaluation results is:
Wherein P c represents a corrected fault probability value, P t represents an original target linear guide fault probability value, alpha is an adjustment coefficient between 0 and 1 for balancing the proportion of original evaluation and homologous fault influence, n represents the number of associated linear guide tracks, w i represents the weight of the ith associated linear guide track, and P ai represents the fault probability value of the ith associated linear guide track.
6. The deep learning-based linear guide rail fault diagnosis method according to claim 1, wherein the target linear guide rail fault diagnosis result includes a corrected fault probability value, a fault type confirmation, a fault severity assessment, a homologous fault impact analysis, and a maintenance recommendation.
7. A deep learning-based linear guide rail fault diagnosis system, wherein the system is applied to the deep learning-based linear guide rail fault diagnosis method of claim 1, the system comprising:
the data acquisition module acquires first running state data of the target linear guide rail and second running state data of at least one associated linear guide rail;
the guide rail fault evaluation module is used for inputting the first running state data into a preset guide rail fault evaluation model to obtain a first fault evaluation result;
The guide rail fault evaluation module is associated, at least one group of second running state data are respectively input into the guide rail fault evaluation model, and at least one group of second fault evaluation results are obtained;
And the fault diagnosis module is used for correcting the first fault evaluation result according to at least one group of second fault evaluation results in consideration of the influence of homologous faults, and obtaining a target linear guide rail fault diagnosis result.
8. An electronic device comprising a bus, a transceiver, a memory, a processor and a computer program stored on the memory and operable on the processor, the transceiver, the memory and the processor being connected by the bus, characterized in that the computer program when executed by the processor realizes the steps in the method according to any of claims 1-6.
9. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method according to any of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202510379438.3A CN119915503B (en) | 2025-03-28 | 2025-03-28 | Linear guide rail fault diagnosis method and system based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202510379438.3A CN119915503B (en) | 2025-03-28 | 2025-03-28 | Linear guide rail fault diagnosis method and system based on deep learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN119915503A CN119915503A (en) | 2025-05-02 |
| CN119915503B true CN119915503B (en) | 2025-08-01 |
Family
ID=95508086
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202510379438.3A Active CN119915503B (en) | 2025-03-28 | 2025-03-28 | Linear guide rail fault diagnosis method and system based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN119915503B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118410279A (en) * | 2024-05-08 | 2024-07-30 | 内蒙古电力(集团)有限责任公司内蒙古电力科学研究院分公司 | Deep learning-based turbine unit fault processing method and system |
| CN118408583A (en) * | 2024-05-08 | 2024-07-30 | 苏州申恩电子科技有限公司 | Encoder fault diagnosis method and system |
| CN119558827A (en) * | 2024-11-27 | 2025-03-04 | 新疆准能投资有限公司 | Equipment failure level assessment method, device, equipment, medium and program product |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103927927A (en) * | 2014-03-14 | 2014-07-16 | 陈志平 | Application teaching method based on linear rolling guide rail sample |
| CN115594043A (en) * | 2022-08-25 | 2023-01-13 | 北方工业大学（Cn） | Wheelchair lift platform fault monitoring and state diagnostic |
| CN115682898A (en) * | 2022-10-11 | 2023-02-03 | 上海诺倬力机电科技有限公司 | A five-axis machining center linear guide fault detection method and system |
| CN116451876B (en) * | 2023-06-15 | 2023-09-22 | 国网江西省电力有限公司信息通信分公司 | A distribution network fault prediction and proactive maintenance system based on artificial intelligence |
-
2025
- 2025-03-28 CN CN202510379438.3A patent/CN119915503B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118410279A (en) * | 2024-05-08 | 2024-07-30 | 内蒙古电力(集团)有限责任公司内蒙古电力科学研究院分公司 | Deep learning-based turbine unit fault processing method and system |
| CN118408583A (en) * | 2024-05-08 | 2024-07-30 | 苏州申恩电子科技有限公司 | Encoder fault diagnosis method and system |
| CN119558827A (en) * | 2024-11-27 | 2025-03-04 | 新疆准能投资有限公司 | Equipment failure level assessment method, device, equipment, medium and program product |
Also Published As
| Publication number | Publication date |
|---|---|
| CN119915503A (en) | 2025-05-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10963797B2 (en) | System for analyzing machine data | |
| CN118313519B (en) | Electromechanical full life cycle prediction modeling method and system | |
| CN119482974B (en) | An operation method and system based on electric power measurement and control instrument | |
| CN119577659B (en) | A method, system, medium and program product for generating an operation and maintenance mode analysis model | |
| CN117636073A (en) | Concrete defect detection method, device and storage medium | |
| CN117331790A (en) | Machine room fault detection method and device for data center | |
| CN119850195B (en) | Operation and maintenance management method and system based on digital twin | |
| CN118965114A (en) | Pumped storage unit fault detection method and detection device | |
| CN119728397B (en) | Network fault prediction method and system | |
| CN119218846B (en) | Elevator operation fault detection method and system | |
| CN119494754A (en) | A method and system for monitoring and warning abnormal environment in a smart power plant | |
| CN117574303B (en) | Construction condition monitoring and early warning method, device, equipment and storage medium | |
| CN119915503B (en) | Linear guide rail fault diagnosis method and system based on deep learning | |
| CN119229323A (en) | High-precision abnormality perception operation and maintenance method and system based on drone inspection | |
| JP2015072644A (en) | Interactive method of predicting facility failure | |
| CN118096112A (en) | Intelligent building facility operation and maintenance method, system, equipment and medium | |
| US20240035445A1 (en) | Systems and methods for estimating future risk of failure of a wind turbine component using machine learning | |
| CN116840658A (en) | Fault prediction and diagnosis method for chip firmware | |
| CN118861944B (en) | A system construction method combining Internet of Things and virtual reality technology | |
| CN118710471B (en) | Water resource dynamic core quantity method and system based on big data analysis | |
| CN119962406B (en) | A maintenance equipment information prediction method and device based on digital twin model | |
| CN120338765A (en) | A method for optimizing assessment of military equipment maintenance and support equipment | |
| CN119357037A (en) | Automated testing method, device, electronic equipment and storage medium | |
| CN119150021A (en) | Method, system, equipment and medium for constructing bolt fault diagnosis model | |
| CN120087759A (en) | An intelligent management platform for construction engineering data information |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |