CN117991752B - Equipment fault prediction system and method based on digital twin and Internet of things technology - Google Patents
Equipment fault prediction system and method based on digital twin and Internet of things technology Download PDFInfo
- Publication number
- CN117991752B CN117991752B CN202410124646.4A CN202410124646A CN117991752B CN 117991752 B CN117991752 B CN 117991752B CN 202410124646 A CN202410124646 A CN 202410124646A CN 117991752 B CN117991752 B CN 117991752B
- Authority
- CN
- China
- Prior art keywords
- training
- numerical control
- control machine
- operation parameter
- machine tool
- 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
- 238000000034 method Methods 0.000 title claims abstract 3
- 230000002159 abnormal effect Effects 0.000 claims abstract 4
- 238000012544 monitoring process Methods 0.000 claims abstract 3
- 238000012549 training Methods 0.000 claims 37
- 239000013598 vector Substances 0.000 claims 36
- 238000010586 diagram Methods 0.000 claims 6
- 238000006243 chemical reaction Methods 0.000 claims 4
- 238000000605 extraction Methods 0.000 claims 4
- 238000005259 measurement Methods 0.000 claims 4
- 238000013527 convolutional neural network Methods 0.000 claims 3
- 238000001514 detection method Methods 0.000 claims 2
- 239000011159 matrix material Substances 0.000 claims 2
- 230000009466 transformation Effects 0.000 claims 2
- 238000003062 neural network model Methods 0.000 claims 1
- 238000012545 processing Methods 0.000 claims 1
- 238000012423 maintenance Methods 0.000 abstract 2
- 238000013473 artificial intelligence Methods 0.000 abstract 1
- 238000013135 deep learning Methods 0.000 abstract 1
- 230000003449 preventive effect Effects 0.000 abstract 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B23/00—Testing or monitoring of control systems or parts thereof
- G05B23/02—Electric testing or monitoring
- G05B23/0205—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults
- G05B23/0218—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterised by the fault detection method dealing with either existing or incipient faults
- G05B23/0243—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterised by the fault detection method dealing with either existing or incipient faults model based detection method, e.g. first-principles knowledge model
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B2219/00—Program-control systems
- G05B2219/20—Pc systems
- G05B2219/24—Pc safety
- G05B2219/24065—Real time diagnostics
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
The application relates to the technical field of intelligent prediction, in particular to a device fault prediction system and a device fault prediction method based on digital twin and Internet of things technology, which utilize an artificial intelligence technology based on deep learning to monitor and analyze a spindle rotation speed value, a spindle temperature value, a feeding speed value, an operation power value and a vibration signal when a numerical control machine tool operates, and capturing time sequence characteristic expression of each monitoring parameter, and judging whether the running state of the numerical control machine tool is abnormal or not based on time sequence related characteristics of each monitoring parameter. Thus, potential faults of the numerical control machine tool can be predicted in advance, so that appropriate maintenance and preventive measures can be taken, and the downtime and maintenance cost are reduced.
Description
Technical Field
The application relates to the technical field of intelligent prediction, in particular to a device fault prediction system and method based on digital twinning and the Internet of things technology.
Background
A numerical control machine (Computer Numerical Control Machine Tool, abbreviated as CNC machine) is an automated device that controls the movement and process of the machine through a computer control system. The tool can accurately control the tool to perform machining operations such as cutting, drilling, milling, grinding and the like on a workpiece according to the pre-programmed instructions. CNC machines are typically composed of a machine body, a numerical control system, an execution system, and auxiliary equipment. The numerical control system controls parameters such as a motion track, a cutting speed, a feeding speed and the like of the machine tool through a pre-written program, so that accurate machining operation is realized. The execution system is responsible for converting the instructions of the numerical control system into the actual movements of the machine tool.
With the continuous development of technology, a numerical control machine tool has become one of core equipment in the modern manufacturing industry. However, various faults may occur in the operation process of the numerical control machine tool, resulting in production line stoppage and production delay, thereby affecting production planning and product delivery time. Therefore, the machine tool is usually required to be inspected and maintained by a worker at regular intervals. However, when the manual periodic detection method faces to complex and variable running environments, misjudgment and missed judgment can occur, and the fault of the equipment cannot be accurately predicted. Moreover, the method cannot monitor and pre-warn the running state of the equipment in real time.
Therefore, a system and method for predicting equipment failure based on digital twinning and internet of things technology is desired.
Disclosure of Invention
The present application has been made to solve the above-mentioned technical problems. The embodiment of the application provides a device fault prediction system and a device fault prediction method based on digital twin and Internet of things technologies, which utilize an artificial intelligence technology based on deep learning to monitor and analyze a spindle rotation speed value, a spindle temperature value, a feed speed value, an operation power value and a vibration signal when a numerical control machine tool is operated, capture time sequence characteristic expression of various monitoring parameters, and judge whether the operation state of the numerical control machine tool is abnormal or not based on time sequence association characteristics of various monitoring parameters. Thus, potential faults of the numerical control machine tool can be predicted in advance, so that appropriate maintenance and preventive measures can be taken, and the downtime and maintenance cost are reduced.
Accordingly, according to one aspect of the present application, there is provided a device fault prediction system based on digital twinning and internet of things technology, comprising:
The numerical control machine tool operation monitoring module is used for acquiring a plurality of pieces of operation parameter data of a plurality of preset time points in a preset time period and vibration signals of the preset time period when the numerical control machine tool is operated, wherein the plurality of pieces of operation parameter data comprise a spindle rotating speed value, a spindle temperature value, a feeding speed value and an operation power value;
the operation parameter time sequence coding module is used for respectively performing time sequence coding on the operation parameter data to obtain operation parameter feature vectors of a plurality of numerical control machine tools;
the vibration characteristic extraction module is used for extracting the time-frequency characteristic of the vibration signal in the preset time period to obtain a vibration signal time-frequency characteristic vector;
the fault detection module is used for determining whether the running state of the numerical control machine tool is abnormal or not based on global correlation characteristics between the characteristic vectors of the running parameters of the numerical control machine tool and the time-frequency characteristic vectors of the vibration signals.
In the above device fault prediction system based on digital twin and internet of things, the operation parameter time sequence coding module includes: the operation parameter time sequence arrangement unit is used for arranging a plurality of operation parameter data of a plurality of preset time points according to time sequence respectively to obtain a plurality of numerical control machine tool operation parameter vectors; the time sequence feature extraction unit is used for enabling the operation parameter vectors of the numerical control machine to pass through the operation parameter time sequence encoder based on the one-dimensional convolution layer respectively to obtain the operation parameter feature vectors of the numerical control machine.
In the above device fault prediction system based on digital twin and internet of things, the vibration feature extraction module includes: the time-frequency diagram conversion unit is used for calculating a SIFT conversion time-frequency diagram of the vibration signal; and the time-frequency characteristic extraction unit is used for enabling the SIFT transformation time-frequency diagram to pass through a vibration characteristic extractor based on a convolutional neural network model so as to obtain the vibration signal time-frequency characteristic vector.
In the above device fault prediction system based on digital twinning and internet of things, the fault detection module includes: the global association unit is used for carrying out global context association coding on the plurality of numerical control machine tool operation parameter vectors and the vibration signal time-frequency feature vector so as to obtain a numerical control machine tool operation state global association feature vector; the classification unit is used for enabling the global association feature vector of the running state of the numerical control machine to pass through a classifier to obtain a classification result, and the classification result is used for indicating whether the running state of the numerical control machine is abnormal or not.
In the above device failure prediction system based on digital twin and internet of things, the global association unit is configured to: and the plurality of numerical control machine tool operation parameter vectors and the vibration signal time-frequency characteristic vector pass through a context encoder based on a converter to obtain the numerical control machine tool operation state global correlation characteristic vector.
In the above device fault prediction system based on digital twinning and internet of things, the classification unit is configured to: processing the global associated feature vector of the running state of the numerical control machine by using the classifier according to the following classification formula to obtain the classification result; wherein, the classification formula is:
O＝softmax{(Wn,Bn):…:(W1,B1)|X}
Wherein, W 1 to W n are weight matrixes, B 1 to B n are bias vectors, X is a global association feature vector of the running state of the numerical control machine tool, softmax represents a normalized exponential function, and O represents the classification result.
The equipment fault prediction system based on the digital twin and the Internet of things technology further comprises a training module for training the one-dimensional convolutional layer-based operation parameter time sequence encoder, the convolutional neural network model-based vibration feature extractor, the converter-based context encoder and the classifier.
In the above device fault prediction system based on digital twinning and internet of things, the training module includes: the training data acquisition unit is used for acquiring training data, wherein the training data comprises a plurality of pieces of training operation parameter data of a plurality of preset time points in a preset time period and training vibration signals of the preset time period when the numerical control machine tool is operated, and the plurality of pieces of training operation parameter data comprise a training spindle rotating speed value, a training spindle temperature value, a training feeding speed value and a training operation power value; the training operation parameter arrangement unit is used for arranging a plurality of training operation parameter data of the plurality of preset time points into a plurality of training numerical control machine tool operation parameter vectors according to a time sequence respectively; the training operation parameter time sequence coding unit is used for enabling the operation parameter vectors of the plurality of training numerical control machines to respectively pass through the operation parameter time sequence coder based on the one-dimensional convolution layer so as to obtain a plurality of training numerical control machine operation parameter feature vectors; the training vibration signal time-frequency diagram conversion unit is used for calculating a training SIFT conversion time-frequency diagram of the training vibration signal; the training vibration signal time-frequency characteristic extraction unit is used for enabling the training SIFT transformation time-frequency diagram to pass through the vibration characteristic extractor based on the convolutional neural network model so as to obtain a training vibration signal time-frequency characteristic vector; the training data global association unit is used for enabling the plurality of training numerical control machine tool operation parameter vectors and the training vibration signal time-frequency feature vectors to pass through the context encoder based on the converter so as to obtain training numerical control machine tool operation state global association feature vectors; the classification loss unit is used for enabling the running state global associated feature vector of the training numerical control machine to pass through the classifier to obtain a classification loss function value; the loss compensation unit is used for calculating a high-dimensional space internal element correlation derivative measurement coefficient of the training numerical control machine tool running state global correlation feature vector to serve as a compensation loss function value; and the training unit is used for training the one-dimensional convolutional layer-based operation parameter time sequence coder, the convolutional neural network model-based vibration feature extractor, the converter-based context coder and the classifier by taking the weighted sum of the classification loss function value and the compensation loss function value as the loss function value.
In the above device failure prediction system based on digital twinning and internet of things, the loss compensation unit is configured to: calculating a correlation derivative measurement coefficient of an element in a high-dimensional space of the global correlation feature vector of the running state of the training numerical control machine tool by using the following correlation derivative measurement formula as a compensation loss function value; wherein, the correlation derivative measurement formula is:
V c represents the global associated feature vector of the running state of the training numerical control machine, f i represents the feature value of the global associated feature vector of the running state of the training numerical control machine, p represents the probability value obtained by the pre-classifier of the global associated feature vector of the running state of the training numerical control machine, I F represents the Frobenius norm of the vector, and Loss represents the correlation derivative measurement coefficient of the element in the high-dimensional space.
According to another aspect of the present application, there is provided a device failure prediction method based on digital twinning and internet of things technology, comprising:
Acquiring a plurality of pieces of operation parameter data of a plurality of preset time points in a preset time period and vibration signals of the preset time period when the numerical control machine tool operates, wherein the plurality of pieces of operation parameter data comprise a spindle rotating speed value, a spindle temperature value, a feeding speed value and an operation power value;
respectively carrying out time sequence coding on the multiple operation parameter data to obtain a plurality of operation parameter feature vectors of the numerical control machine tool;
Extracting the time-frequency characteristic of the vibration signal in the preset time period to obtain a time-frequency characteristic vector of the vibration signal;
And determining whether the running state of the numerical control machine is abnormal or not based on global correlation features between the running parameter feature vectors of the numerical control machine and the time-frequency feature vectors of the vibration signals.
Compared with the prior art, the equipment fault prediction system and method based on the digital twin and Internet of things technology provided by the application utilize an artificial intelligence technology based on deep learning to monitor and analyze a spindle rotation speed value, a spindle temperature value, a feeding speed value, an operation power value and a vibration signal when a numerical control machine tool is operated, capture time sequence characteristic expression of each monitoring parameter, and judge whether the operation state of the numerical control machine tool is abnormal or not based on time sequence association characteristics of each monitoring parameter. Thus, potential faults of the numerical control machine tool can be predicted in advance, so that appropriate maintenance and preventive measures can be taken, and the downtime and maintenance cost are reduced.
Drawings
The above and other objects, features and advantages of the present application will become more apparent by describing embodiments of the present application in more detail with reference to the attached drawings. The accompanying drawings are included to provide a further understanding of embodiments of the application and are incorporated in and constitute a part of this specification, illustrate the application and together with the embodiments of the application, and not constitute a limitation to the application. In the drawings, like reference numerals generally refer to like parts or steps.
Fig. 1 is a block diagram of a device fault prediction system based on digital twinning and internet of things according to an embodiment of the present application.
Fig. 2 is a schematic architecture diagram of a device failure prediction system based on digital twinning and internet of things according to an embodiment of the present application.
FIG. 3 is a block diagram of an operational parameter timing encoding module in a device fault prediction system based on digital twinning and Internet of things technology according to an embodiment of the present application.
Fig. 4 is a block diagram of a vibration feature extraction module in a device fault prediction system based on digital twinning and internet of things according to an embodiment of the present application.
Fig. 5 is a block diagram of a fault detection module in a device fault prediction system based on digital twinning and internet of things according to an embodiment of the present application.
Fig. 6 is a block diagram of a training module in a device failure prediction system based on digital twinning and internet of things according to an embodiment of the present application.
Fig. 7 is a flowchart of a device failure prediction method based on digital twinning and internet of things according to an embodiment of the present application.
Detailed Description
Hereinafter, exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. It should be apparent that the described embodiments are only some embodiments of the present application and not all embodiments of the present application, and it should be understood that the present application is not limited by the example embodiments described herein.
Fig. 1 is a block diagram of a device fault prediction system based on digital twinning and internet of things according to an embodiment of the present application. Fig. 2 is a schematic architecture diagram of a device failure prediction system based on digital twinning and internet of things according to an embodiment of the present application. As shown in fig. 1 and 2, a device failure prediction system 100 based on digital twin and internet of things according to an embodiment of the present application includes: the numerical control machine tool operation monitoring module 110 is configured to obtain a plurality of pieces of operation parameter data of a plurality of predetermined time points in a predetermined time period and vibration signals of the predetermined time period when the numerical control machine tool is operated, where the plurality of pieces of operation parameter data include a spindle rotation speed value, a spindle temperature value, a feed speed value, and an operation power value; the operation parameter time sequence coding module 120 is configured to perform time sequence coding on the plurality of operation parameter data respectively to obtain a plurality of operation parameter feature vectors of the numerical control machine tool; a vibration feature extraction module 130, configured to extract a time-frequency feature of the vibration signal in the predetermined period of time to obtain a time-frequency feature vector of the vibration signal; the fault detection module 140 is configured to determine whether an abnormality exists in an operation state of the numerically-controlled machine tool based on global correlation features between the feature vectors of the operation parameters of the plurality of numerically-controlled machine tools and the time-frequency feature vector of the vibration signal.
As described in the background art above, various faults may occur in the operation process of the numerical control machine tool, which may cause production line stoppage and production delay, thereby affecting production schedule and product delivery time. Therefore, the machine tool is usually required to be inspected and maintained by a worker at regular intervals. However, when the manual periodic detection method faces to complex and variable running environments, misjudgment and missed judgment can occur, and the fault of the equipment cannot be accurately predicted. Moreover, the manual detection method cannot monitor and pre-warn the running state of the equipment in real time.
Aiming at the technical problems, the technical concept of the application is to monitor and analyze the spindle rotation speed value, the spindle temperature value, the feeding speed value, the running power value and the vibration signal when the numerical control machine tool runs by utilizing an artificial intelligence technology based on deep learning, capture the time sequence characteristic expression of each monitoring parameter, and judge whether the running state of the numerical control machine tool is abnormal or not based on the time sequence related characteristic of each monitoring parameter. Thus, potential faults of the numerical control machine tool can be predicted in advance, so that appropriate maintenance and preventive measures can be taken, and the downtime and maintenance cost are reduced.
In the above-mentioned equipment failure prediction system 100 based on digital twin and internet of things, the operation monitoring module 110 of the numerically-controlled machine tool is configured to obtain a plurality of pieces of operation parameter data of a plurality of predetermined time points and vibration signals of the predetermined time period during operation of the numerically-controlled machine tool, where the plurality of pieces of operation parameter data include a spindle rotation speed value, a spindle temperature value, a feed speed value, and an operation power value. It should be understood that spindle speed is an important operating parameter in numerically controlled machine tools that reflects the rotational speed of the spindle. Abnormal spindle speeds may mean spindle bearing wear, poor lubrication, or other mechanical problems. The spindle temperature is the working temperature of the index control machine tool spindle. Excessive spindle temperatures may indicate cooling system failure, poor lubrication, or cutting fluid supply problems. Potential fault conditions can be found in time by monitoring the temperature of the main shaft, and damage of the machine tool due to overheating is avoided. The feed rate is the speed of movement of the workpiece relative to the tool during machining in a numerically controlled machine tool. Abnormal feed rates may lead to reduced cutting quality, workpiece damage or tool wear. The operating power is the electrical energy that the machine tool is controlled to consume during operation. Abnormal operating power may indicate that the machine is experiencing load imbalance, motor failure, or other electrical problems. By monitoring the change of the operating power, the abnormal condition of the electrical system can be found in time, and the machine tool is prevented from being stopped or damaged due to electrical faults. Vibration signals are a common type of signal that monitors the condition of a machine tool and can reflect the stability of the machine tool structure and the dynamics during cutting. By analyzing the vibration signal, the information such as the vibration frequency, the vibration amplitude, the vibration mode and the like of the machine tool can be detected, so that whether the machine tool has abnormality such as looseness, abrasion, cutter collision and the like is judged. That is, by monitoring a plurality of operation parameters and vibration signals during the operation of the numerical control machine tool, comprehensive machine tool state information can be obtained, an accurate fault prediction system can be constructed, and real-time monitoring and abnormality detection of the machine tool operation state can be realized.
In the above-mentioned equipment failure prediction system 100 based on digital twin and internet of things, the operation parameter time sequence encoding module 120 is configured to perform time sequence encoding on the multiple operation parameter data respectively to obtain multiple operation parameter feature vectors of the numerical control machine tool. Specifically, fig. 3 is a block diagram of an operation parameter timing encoding module in an equipment failure prediction system based on digital twin and internet of things technology according to an embodiment of the present application. As shown in fig. 3, the operation parameter timing encoding module 120 includes: an operation parameter time sequence arrangement unit 121, configured to arrange a plurality of operation parameter data of the plurality of predetermined time points in time sequence to obtain a plurality of operation parameter vectors of the numerical control machine tool; the time sequence feature extraction unit 122 is configured to obtain the plurality of operation parameter feature vectors of the numerically-controlled machine tool by passing the plurality of operation parameter vectors of the numerically-controlled machine tool through operation parameter time sequence encoders based on the one-dimensional convolution layer.
Specifically, the operation parameter time sequence arrangement unit 121 is configured to arrange the plurality of pieces of operation parameter data at the plurality of predetermined time points in time sequence to obtain a plurality of operation parameter vectors of the numerically-controlled machine tool. It will be appreciated that the operating parameters of the machine tool typically have a time-series correlation, i.e. the parameter value at the current time may be related to the parameter value at the previous time or before. By arranging the operation parameter data of the predetermined time points according to time sequence, the time sequence correlation characteristic can be maintained, so that the model can learn a dynamic change mode among parameter values, further capture a change trend, periodic change or other time sequence modes of the parameter values, and further identify an abnormal mode related to machine tool faults. For example, if a certain parameter value exhibits an abnormal fluctuation or mutation over time, it may mean that there is a fault or abnormal condition in the machine tool. That is, by arranging each item of operation parameter data in time series, it is possible to facilitate capturing of time series characteristics of the machine tool operation state in the subsequent data analysis process, thereby providing more accurate input data for the subsequent failure prediction and abnormality detection.
Specifically, the timing feature extraction unit 122 is configured to pass the operation parameter vectors of the plurality of numerically-controlled machine tools through a one-dimensional convolution layer-based operation parameter timing encoder to obtain the operation parameter feature vectors of the plurality of numerically-controlled machine tools, respectively. It should be appreciated that one-dimensional convolutional layers are one type of convolutional neural network layer commonly used in deep learning, and are commonly used to process data having a time-series structure. In the technical scheme of the application, the operation parameter vector of the numerical control machine tool is input into the operation parameter time sequence encoder based on the one-dimensional convolution layer, and the operation parameter time sequence encoder can capture the local mode in the operation parameter vector of the numerical control machine tool by utilizing the local perception capability of the one-dimensional convolution layer. Specifically, the one-dimensional convolution layer performs local feature extraction on different parts of the operation parameter vector of the numerical control machine by sliding a learnable convolution kernel (filter) on the operation parameter vector of the numerical control machine, and learns local dependency relations and time sequence modes among parameter values, so that the time sequence relations and change modes among the parameter values are better captured, and feature representation with more information is provided for subsequent fault prediction and anomaly detection.
In the above-mentioned equipment failure prediction system 100 based on digital twinning and internet of things, the vibration feature extraction module 130 is configured to extract a time-frequency feature of the vibration signal in the predetermined period of time to obtain a vibration signal time-frequency feature vector. Specifically, fig. 4 is a block diagram of a vibration feature extraction module in a device failure prediction system based on digital twinning and internet of things according to an embodiment of the present application. As shown in fig. 4, the vibration feature extraction module 130 includes: a time-frequency diagram conversion unit 131, configured to calculate a SIFT conversion time-frequency diagram of the vibration signal; and the time-frequency characteristic extraction unit 132 is used for passing the SIFT transformation time-frequency chart through a vibration characteristic extractor based on a convolutional neural network model to obtain the vibration signal time-frequency characteristic vector.
Specifically, the time-frequency diagram converting unit 131 is configured to calculate a SIFT conversion time-frequency diagram of the vibration signal. It should be appreciated that different mechanical faults typically produce specific patterns or features in the time and frequency domains of the vibration signal, such as variations in spectral peaks, shifts in frequency components, etc. Therefore, in order to more fully describe the characteristics of the vibration signal, the time-frequency characteristic representation of the vibration signal is further extracted. Because the SIFT transformation analyzes the signal at different scales and frequencies, the local structure and frequency distribution of the signal can be captured. Therefore, by calculating the SIFT conversion time-frequency diagram of the vibration signal, a group of characteristic points with better robustness and invariance can be obtained and used for describing the remarkable time-frequency characteristics of the vibration signal, so that the time-frequency characteristics of the vibration signal are more comprehensively represented, the running state of a machine tool is understood, and valuable input data are provided for subsequent fault diagnosis and prediction.
Specifically, the time-frequency feature extraction unit 132 is configured to pass the SIFT transformed time-frequency graph through a vibration feature extractor based on a convolutional neural network model to obtain the vibration signal time-frequency feature vector. It should be understood that, considering that the SIFT transform time-frequency diagram has a relatively high dimension and a complex structure, directly using the SIFT transform time-frequency diagram as the time-frequency characteristic representation of the vibration signal may cause the characteristic dimension to be too high, which increases the computational complexity. Thus, the SIFT transformed time-frequency graph is converted into a more compact and informative feature vector representation by using a convolutional neural network based feature extractor. Here, the convolutional neural network can learn important time-frequency characteristics in the SIFT transformation time-frequency diagram through convolution and pooling operation, filter irrelevant redundant characteristics, extract more abstract and more representative characteristic representation, thereby better representing the essence of vibration signals, improving the expressive power and the distinguishing power of the characteristics, reducing the feature dimension, and reducing the storage space and the calculation cost.
In a specific example of the present application, the time-frequency feature extraction unit 132 is configured to: and respectively carrying out convolution processing based on a two-dimensional convolution kernel, global average pooling processing and nonlinear activation processing based on each feature matrix of the channel dimension on input data in forward transmission of layers by using each layer of the vibration feature extractor based on the convolution neural network model so as to output the vibration signal time-frequency feature vector by the last layer of the vibration feature extractor based on the convolution neural network model.
In the above-mentioned equipment fault prediction system 100 based on digital twinning and internet of things, the fault detection module 140 is configured to determine whether an abnormal state exists in the operation state of the numerically-controlled machine tool based on global correlation features between the feature vectors of the operation parameters of the plurality of numerically-controlled machine tools and the time-frequency feature vector of the vibration signal. Specifically, fig. 5 is a block diagram of a fault detection module in a device fault prediction system based on digital twinning and internet of things according to an embodiment of the present application. As shown in fig. 5, the fault detection module 140 includes: the global association unit 141 is configured to perform global context association coding on the plurality of operation parameter vectors of the numerically-controlled machine tool and the vibration signal time-frequency feature vector to obtain a global association feature vector of an operation state of the numerically-controlled machine tool; and the classification unit 142 is configured to pass the global associated feature vector of the operation state of the numerically-controlled machine tool through a classifier to obtain a classification result, where the classification result is used to indicate whether the operation state of the numerically-controlled machine tool is abnormal.
Specifically, the global association unit 141 is configured to perform global context-based association encoding on the plurality of operation parameter vectors of the numerically-controlled machine tool and the vibration signal time-frequency feature vector to obtain a globally associated feature vector of an operation state of the numerically-controlled machine tool. It should be appreciated that the operating state of the numerically controlled machine tool may be composed of information of a plurality of aspects, including the plurality of operating parameters and the vibration signal. By fusing these different types of operating characteristics, a more comprehensive, accurate representation of machine tool state may be provided. In a specific example of the present application, the encoding method for performing global context-based association encoding on the plurality of numerically-controlled machine tool operation parameter vectors and the vibration signal time-frequency feature vector to obtain a numerically-controlled machine tool operation state global association feature vector is that the plurality of numerically-controlled machine tool operation parameter vectors and the vibration signal time-frequency feature vector pass through a context encoder based on a converter to obtain the numerically-controlled machine tool operation state global association feature vector. Here, by modeling the plurality of numerical control machine tool operation parameter vectors and the vibration signal time-frequency feature vector by using a context encoder based on a converter, the association and the dependency between different operation parameter features can be learned by using the strong long-distance dependence modeling capability of the converter model, and the context and the dependency in input data are captured, so that the global association feature representation of the numerical control machine tool operation state with more expression capability is obtained, and the operation state of the machine tool can be described more accurately.
Specifically, the classification unit 142 is configured to pass the global related feature vector of the operation state of the numerically-controlled machine tool through a classifier to obtain a classification result, where the classification result is used to indicate whether the operation state of the numerically-controlled machine tool is abnormal. It should be understood that the classifier is trained to learn the characteristic mode difference between the normal running state and the abnormal running state of the numerically-controlled machine tool, so that the running state of the machine tool can be accurately detected abnormally. That is, by inputting the global association feature vector of the running state of the numerically-controlled machine tool into the classifier after training, the classifier can classify the global association feature vector of the running state of the numerically-controlled machine tool by utilizing association rules between different running state features and running state type labels of the numerically-controlled machine tool learned in the training process, so as to obtain a classification result for indicating whether the running state of the numerically-controlled machine tool is abnormal or not. Therefore, when the classifier judges that the running state of the machine tool is abnormal, a corresponding alarm system or maintenance flow can be automatically triggered, so that measures can be taken in time to conduct fault investigation and repair, the fault processing efficiency and response speed are improved, and the downtime of the production line and production loss are reduced.
In a specific example of the present application, the classifying unit 142 is configured to: processing the global associated feature vector of the running state of the numerical control machine by using the classifier according to the following classification formula to obtain the classification result; wherein, the classification formula is:
O＝softmax{(Wn,Bn):…:(W1,B1)|X}
Wherein, W 1 to W n are weight matrixes, B 1 to B n are bias vectors, X is a global association feature vector of the running state of the numerical control machine tool, softmax represents a normalized exponential function, and O represents the classification result.
It should be appreciated that the one-dimensional convolutional layer based operational parameter timing encoder, the convolutional neural network model based vibration feature extractor, the converter based context encoder, and the classifier need to be trained prior to utilizing the neural network model described above. That is, in the device fault prediction system based on the digital twin and the internet of things technology of the present application, a training module for training the one-dimensional convolutional layer-based operation parameter timing encoder, the convolutional neural network model-based vibration feature extractor, the converter-based context encoder, and the classifier is further included.
Fig. 6 is a block diagram of a training module in a device failure prediction system based on digital twinning and internet of things according to an embodiment of the present application. In the above-mentioned equipment failure prediction system based on digital twinning and internet of things, the training module 200 includes: a training data obtaining unit 210, configured to obtain training data, where the training data includes a plurality of pieces of training operation parameter data of a plurality of predetermined time points in a predetermined time period and training vibration signals of the predetermined time period when the numerically controlled machine tool is operated, and the plurality of pieces of training operation parameter data include a training spindle rotation speed value, a training spindle temperature value, a training feeding speed value, and a training operation power value; a training operation parameter arrangement unit 220, configured to arrange the plurality of training operation parameter data at the plurality of predetermined time points into a plurality of training numerical control machine operation parameter vectors according to a time sequence, respectively; a training operation parameter time sequence encoding unit 230, configured to pass the operation parameter vectors of the plurality of training numerically-controlled machine tools through the operation parameter time sequence encoder based on the one-dimensional convolution layer to obtain a plurality of operation parameter feature vectors of the training numerically-controlled machine tools; a training vibration signal time-frequency diagram conversion unit 240, configured to calculate a training SIFT conversion time-frequency diagram of the training vibration signal; the training vibration signal time-frequency feature extraction unit 250 is configured to pass the training SIFT transformation time-frequency graph through the vibration feature extractor based on the convolutional neural network model to obtain a training vibration signal time-frequency feature vector; the training data global association unit 260 is configured to pass the plurality of training numerically-controlled machine tool operation parameter vectors and the training vibration signal time-frequency feature vector through the context encoder based on the converter to obtain a training numerically-controlled machine tool operation state global association feature vector; the classification loss unit 270 is configured to pass the global associated feature vector of the running state of the training numerical control machine through the classifier to obtain a classification loss function value; the loss compensation unit 280 is configured to calculate a correlation derivative measurement coefficient of an element in a high-dimensional space of the global correlation feature vector of the running state of the training numerical control machine as a compensation loss function value; a training unit 290 for training the one-dimensional convolutional layer-based operation parameter timing encoder, the convolutional neural network model-based vibration feature extractor, the converter-based context encoder, and the classifier with a weighted sum of the classification loss function value and the compensation loss function value as a loss function value.
In the technical scheme of the application, the global associated feature vector of the running state of the training numerical control machine is obtained by carrying out feature fusion on a plurality of running parameter feature vectors of the training numerical control machine and a training vibration signal time-frequency feature map. However, since the correlation of the internal elements of the training numerical control machine tool operation state global correlation feature vector is poor, the following problem may be caused, firstly, the internal elements with poor correlation may cause the unstable structure of the overall feature distribution of the training numerical control machine tool operation state global correlation feature vector. In classification tasks, stable feature distribution is important for model training and inference. If the structure of the feature vector is unstable, the model can hardly learn effective feature representation, thereby affecting the accuracy of the classification result. Secondly, due to poor correlation of internal elements, the situation that a local structure collapses may exist in the global correlation feature vector of the running state of the training numerical control machine. Local structural collapse refers to the fact that the relationships between certain elements in the feature vector become blurred or lost, resulting in a model that cannot accurately capture these relationships. This can lead to confusion or errors in the classifier in making the classification decisions, thereby affecting the accuracy of the classification results. That is, the poor correlation of the internal elements of the global correlation feature vector of the running state of the training numerical control machine may affect the structural stability of the overall feature distribution of the training numerical control machine, and may cause the collapse of the local structure, thereby affecting the accuracy of the classification result obtained by the classifier. In order to solve the problem, the high-dimensional space internal element correlation derivative measurement coefficient of the overall correlation feature vector of the running state of the training numerical control machine is calculated to be used as a compensation loss function value so as to improve the classification adaptability of the feature vector.
Specifically, the loss compensation unit 280 is configured to: calculating a correlation derivative measurement coefficient of an element in a high-dimensional space of the global correlation feature vector of the running state of the training numerical control machine tool by using the following correlation derivative measurement formula as a compensation loss function value; wherein, the correlation derivative measurement formula is:
V c represents the global associated feature vector of the running state of the training numerical control machine, f i represents the feature value of the global associated feature vector of the running state of the training numerical control machine, p represents the probability value obtained by the pre-classifier of the global associated feature vector of the running state of the training numerical control machine, I F represents the Frobenius norm of the vector, and Loss represents the correlation derivative measurement coefficient of the element in the high-dimensional space.
That is, considering that the correlation of the internal elements of the global correlation feature vector of the running state of the training numerical control machine is poor, the correlation not only affects the structural stability of the overall feature distribution of the global correlation feature vector of the running state of the training numerical control machine, but also causes the local structural collapse of the global correlation feature vector of the running state of the training numerical control machine, and affects the accuracy of the classification result obtained by the classifier. According to the technical scheme, the high-dimensional space internal element correlation derivative measurement coefficient of the overall correlation feature vector of the running state of the training numerical control machine tool is calculated to serve as a compensation loss function value, the confidence coefficient of the internal element in the feature vector to the target class probability label domain can be effectively measured by constructing a compensation loss function based on scattering response of the feature value position relative to the label probability value, so that the feature value of each position in the feature vector can keep consistency with the label probability value in the sub-dimension of the feature vector, distribution adjustment of the internal element of the feature vector is achieved, the requirement of the target class probability label domain is met, and classification adaptability of the feature vector is improved.
In summary, the device fault prediction system based on the digital twin and internet of things technology according to the embodiment of the application is explained, which monitors and analyzes a spindle rotation speed value, a spindle temperature value, a feed speed value, an operation power value and a vibration signal when a numerical control machine tool is operated by using an artificial intelligence technology based on deep learning, captures time sequence characteristic expression of each monitoring parameter, and judges whether the operation state of the numerical control machine tool is abnormal based on time sequence association characteristics of each monitoring parameter. Thus, potential faults of the numerical control machine tool can be predicted in advance, so that appropriate maintenance and preventive measures can be taken, and the downtime and maintenance cost are reduced.
Fig. 7 is a flowchart of a device failure prediction method based on digital twinning and internet of things according to an embodiment of the present application. As shown in fig. 7, a device failure prediction method based on digital twin and internet of things according to an embodiment of the present application includes the steps of: s110, acquiring a plurality of pieces of operation parameter data of a plurality of preset time points and vibration signals of a preset time period in the preset time period when the numerical control machine tool operates, wherein the plurality of pieces of operation parameter data comprise a spindle rotating speed value, a spindle temperature value, a feeding speed value and an operation power value; s120, respectively performing time sequence coding on the plurality of operation parameter data to obtain a plurality of numerical control machine operation parameter feature vectors; s130, extracting time-frequency characteristics of the vibration signal in the preset time period to obtain a time-frequency characteristic vector of the vibration signal; and S140, determining whether the running state of the numerical control machine tool is abnormal or not based on the global correlation characteristic between the running parameter characteristic vectors of the numerical control machine tool and the vibration signal time-frequency characteristic vector.
Here, it will be understood by those skilled in the art that the specific operations of the respective steps in the above-described digital twinning and internet of things based device failure prediction method have been described in detail in the above description of the digital twinning and internet of things based device failure prediction system with reference to fig. 1 to 6, and thus, repetitive descriptions thereof will be omitted.
In the foregoing embodiments, the descriptions of the embodiments are emphasized, and in part, not described or illustrated in any particular embodiment, reference is made to the related descriptions of other embodiments. In the several embodiments provided by the present invention, it should be understood that the disclosed apparatus, system and method may be implemented in other manners. For example, the system embodiments described above are merely illustrative, and for example, the module division is merely a logical function division, and other manners of division may be implemented in practice. The modules described as separate components may or may not be physically separate, and components shown as modules may or may not be physical units, may be located in one place, or may be distributed over multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional module in the embodiments of the present invention may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units can be realized in a form of hardware or a form of hardware and a form of software functional modules.
It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential characteristics thereof. The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference signs in the claims shall not be construed as limiting the claim concerned.
Furthermore, it is evident that the word "comprising" does not exclude other elements or steps, and that the singular does not exclude a plurality. The units recited in the system claims may also be implemented by means of software or hardware.
Finally, it should be noted that the above-mentioned embodiments are merely for illustrating the technical solution of the present invention and not for limiting the same, and although the present invention has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that modifications and equivalents may be made to the technical solution of the present invention without departing from the spirit and scope of the technical solution of the present invention.
Claims (3)
1. An equipment fault prediction system based on digital twinning and the internet of things technology is characterized by comprising:
The numerical control machine tool operation monitoring module is used for acquiring a plurality of pieces of operation parameter data of a plurality of preset time points in a preset time period and vibration signals of the preset time period when the numerical control machine tool is operated, wherein the plurality of pieces of operation parameter data comprise a spindle rotating speed value, a spindle temperature value, a feeding speed value and an operation power value;
the operation parameter time sequence coding module is used for respectively performing time sequence coding on the operation parameter data to obtain operation parameter feature vectors of a plurality of numerical control machine tools;
the vibration characteristic extraction module is used for extracting the time-frequency characteristic of the vibration signal in the preset time period to obtain a vibration signal time-frequency characteristic vector;
The fault detection module is used for determining whether the running state of the numerical control machine tool is abnormal or not based on global correlation characteristics between the characteristic vectors of the running parameters of the numerical control machine tool and the time-frequency characteristic vectors of the vibration signals;
Wherein, the operation parameter time sequence coding module comprises:
the operation parameter time sequence arrangement unit is used for arranging a plurality of operation parameter data of a plurality of preset time points according to time sequence respectively to obtain a plurality of numerical control machine tool operation parameter vectors;
The time sequence feature extraction unit is used for enabling the operation parameter vectors of the plurality of numerical control machines to respectively pass through an operation parameter time sequence encoder based on a one-dimensional convolution layer so as to obtain the operation parameter feature vectors of the plurality of numerical control machines;
wherein, vibration characteristic draws the module, includes:
The time-frequency diagram conversion unit is used for calculating a SIFT conversion time-frequency diagram of the vibration signal;
The time-frequency characteristic extraction unit is used for enabling the SIFT transformation time-frequency diagram to pass through a vibration characteristic extractor based on a convolutional neural network model so as to obtain a vibration signal time-frequency characteristic vector;
Wherein, the fault detection module includes:
the global association unit is used for carrying out global context association coding on the plurality of numerical control machine tool operation parameter vectors and the vibration signal time-frequency feature vector so as to obtain a numerical control machine tool operation state global association feature vector;
the classification unit is used for enabling the overall association feature vector of the running state of the numerical control machine to pass through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the running state of the numerical control machine is abnormal or not;
wherein, global association unit is used for:
The running parameter vectors of the numerical control machine and the vibration signal time-frequency characteristic vector are passed through a context encoder based on a converter to obtain a global associated characteristic vector of the running state of the numerical control machine;
The system further comprises a training module for training an operation parameter time sequence encoder based on a one-dimensional convolution layer, a vibration feature extractor based on a convolution neural network model, a context encoder based on a converter and a classifier;
Wherein, training module includes:
The training data acquisition unit is used for acquiring training data, wherein the training data comprises a plurality of pieces of training operation parameter data of a plurality of preset time points in a preset time period and training vibration signals of the preset time period when the numerical control machine tool is operated, and the plurality of pieces of training operation parameter data comprise a training spindle rotating speed value, a training spindle temperature value, a training feeding speed value and a training operation power value;
the training operation parameter arrangement unit is used for arranging a plurality of training operation parameter data of the plurality of preset time points into a plurality of training numerical control machine tool operation parameter vectors according to a time sequence respectively;
The training operation parameter time sequence coding unit is used for enabling the operation parameter vectors of the plurality of training numerical control machines to respectively pass through the operation parameter time sequence coder based on the one-dimensional convolution layer so as to obtain a plurality of training numerical control machine operation parameter feature vectors;
the training vibration signal time-frequency diagram conversion unit is used for calculating a training SIFT conversion time-frequency diagram of the training vibration signal;
The training vibration signal time-frequency characteristic extraction unit is used for enabling the training SIFT transformation time-frequency diagram to pass through the vibration characteristic extractor based on the convolutional neural network model so as to obtain a training vibration signal time-frequency characteristic vector;
the training data global association unit is used for enabling the plurality of training numerical control machine tool operation parameter vectors and the training vibration signal time-frequency feature vectors to pass through the context encoder based on the converter so as to obtain training numerical control machine tool operation state global association feature vectors;
The classification loss unit is used for enabling the running state global associated feature vector of the training numerical control machine to pass through the classifier to obtain a classification loss function value;
the loss compensation unit is used for calculating a high-dimensional space internal element correlation derivative measurement coefficient of the training numerical control machine tool running state global correlation feature vector to serve as a compensation loss function value;
A training unit for training the one-dimensional convolutional layer-based operation parameter timing encoder, the convolutional neural network model-based vibration feature extractor, the converter-based context encoder, and the classifier with a weighted sum of the classification loss function value and the compensation loss function value as a loss function value;
wherein the loss compensation unit is configured to: calculating a correlation derivative measurement coefficient of an element in a high-dimensional space of the global correlation feature vector of the running state of the training numerical control machine tool by using the following correlation derivative measurement formula as a compensation loss function value;
Wherein, the correlation derivative measurement formula is:
wherein, Representing the running state global associated feature vector of the training numerical control machine,Representing the characteristic value of the global associated characteristic vector of the running state of the training numerical control machine,Representing the probability value obtained by the global association feature vector of the running state of the training numerical control machine through a pre-classifier,The Frobenius norm of the vector,Representing the element correlation derivative metric coefficient inside the high-dimensional space.
2. The device fault prediction system based on digital twinning and internet of things according to claim 1, wherein the classification unit is configured to: processing the global associated feature vector of the running state of the numerical control machine by using the classifier according to the following classification formula to obtain the classification result; wherein, the classification formula is:
wherein, To the point ofAs a matrix of weights, the weight matrix,To the point ofAs a result of the offset vector,Global associated feature vectors for the running state of the numerical control machine,Representing the normalized exponential function of the sample,Representing the classification result.
3. A device fault prediction method based on digital twin and internet of things technology, using the device fault prediction system based on digital twin and internet of things technology as claimed in claim 1, comprising:
Acquiring a plurality of pieces of operation parameter data of a plurality of preset time points in a preset time period and vibration signals of the preset time period when the numerical control machine tool operates, wherein the plurality of pieces of operation parameter data comprise a spindle rotating speed value, a spindle temperature value, a feeding speed value and an operation power value;
respectively carrying out time sequence coding on the multiple operation parameter data to obtain a plurality of operation parameter feature vectors of the numerical control machine tool;
Extracting the time-frequency characteristic of the vibration signal in the preset time period to obtain a time-frequency characteristic vector of the vibration signal;
And determining whether the running state of the numerical control machine is abnormal or not based on global correlation features between the running parameter feature vectors of the numerical control machine and the time-frequency feature vectors of the vibration signals.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410124646.4A CN117991752B (en) | 2024-01-29 | 2024-01-29 | Equipment fault prediction system and method based on digital twin and Internet of things technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410124646.4A CN117991752B (en) | 2024-01-29 | 2024-01-29 | Equipment fault prediction system and method based on digital twin and Internet of things technology |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117991752A CN117991752A (en) | 2024-05-07 |
| CN117991752B true CN117991752B (en) | 2024-09-27 |
Family
ID=90897158
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410124646.4A Active CN117991752B (en) | 2024-01-29 | 2024-01-29 | Equipment fault prediction system and method based on digital twin and Internet of things technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117991752B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118552186B (en) * | 2024-07-29 | 2024-10-15 | 国网湖北省电力有限公司 | Substation operation and maintenance management system and method based on digital twin |
| CN119067640B (en) * | 2024-11-04 | 2025-01-03 | 杭州优服云科技有限公司 | Intelligent operation and maintenance management method and system for numerical control machine tool |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116880402A (en) * | 2023-07-28 | 2023-10-13 | 江西斯源科技股份有限公司 | Intelligent factory cooperative control system and method thereof |
| CN117176270A (en) * | 2023-09-05 | 2023-12-05 | 浙江畅能数智科技有限公司 | Indoor antenna with signal monitoring function |
| CN117420792A (en) * | 2023-11-06 | 2024-01-19 | 浙江电泰阀业有限公司 | Central control system for composite processing machine tool |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4321581B2 (en) * | 2006-11-30 | 2009-08-26 | パナソニック電工株式会社 | Machine tool comprehensive monitoring device |
| CN116929815A (en) * | 2023-07-20 | 2023-10-24 | 漳州市诺兰信息科技有限公司 | Equipment working state monitoring system and method based on Internet of things |
| CN116880358A (en) * | 2023-08-11 | 2023-10-13 | 安徽省通快数控机床制造有限公司 | Intelligent monitoring system for numerical control machine tool |
-
2024
- 2024-01-29 CN CN202410124646.4A patent/CN117991752B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116880402A (en) * | 2023-07-28 | 2023-10-13 | 江西斯源科技股份有限公司 | Intelligent factory cooperative control system and method thereof |
| CN117176270A (en) * | 2023-09-05 | 2023-12-05 | 浙江畅能数智科技有限公司 | Indoor antenna with signal monitoring function |
| CN117420792A (en) * | 2023-11-06 | 2024-01-19 | 浙江电泰阀业有限公司 | Central control system for composite processing machine tool |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117991752A (en) | 2024-05-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11237539B2 (en) | System and method for operational-data-based detection of anomaly of a machine tool | |
| CN117991752B (en) | Equipment fault prediction system and method based on digital twin and Internet of things technology | |
| Schmidt et al. | Cloud-enhanced predictive maintenance | |
| Li et al. | Intelligent predictive maintenance for fault diagnosis and prognosis in machine centers: Industry 4.0 scenario | |
| CN117786373B (en) | Equipment operation diagnosis system based on big data corrugated paper processing | |
| Rao | Decision making in the manufacturing environment: using graph theory and fuzzy multiple attribute decision making methods | |
| Butler et al. | Condition monitoring of machine tool feed drives: a review | |
| US20060036403A1 (en) | Diagnostic systems and methods for predictive condition monitoring | |
| Groba et al. | Architecture of a predictive maintenance framework | |
| CN119312148B (en) | Equipment failure prediction method and related equipment based on industrial Internet of Things | |
| CN119513671A (en) | A traffic anomaly identification method for CNC machine tool protocols | |
| CN118259622A (en) | Intelligent processing control method and electronic equipment for CNC lathe | |
| CN119847066A (en) | A method and system for intelligently controlling the operation status of a precision CNC machine tool | |
| CN118818292B (en) | A servo motor operation quality detection method based on big data | |
| CN118243378A (en) | Fault diagnosis method and system for cutting speed reducer of cantilever type heading machine | |
| Alfaro-Nango et al. | Predictive maintenance algorithm based on machine learning for industrial asset | |
| Moreira et al. | A Timed Automaton Model with Timing Intervals and Outputs for Fault Diagnosis of the Drilling Process on a CNC Machine | |
| Shankar | Intelligent Fault Diagnosis in Electric Motors Using AI Techniques | |
| Juuso et al. | Combining monitoring and process data in detecting operation conditions in process industry | |
| Mukund et al. | Optimizing Predictive Maintenance in Mechanical Engineering: AI and ML for Lathe Machines | |
| Ramanathan et al. | CNC machine tool monitoring using artificial intelligence a review | |
| Huang | Online Fault Diagnosis of CNC Machine Tools Based on Deep Learning | |
| Sarode | A framework for identifying domain knowledge to be included in data driven fault diagnostics | |
| CN120234730A (en) | Predictive maintenance method for machine tools based on multi-point vibration sensor monitoring | |
| Tieng et al. | Generic tool-performance-estimation scheme for anomaly detection of equipment |
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 |