Numerical control machine tool equipment state real-time monitoring system and method based on multiple sensors

Abstract

The invention provides a numerical control machine tool equipment state real-time monitoring system and method based on multiple sensors, and relates to the field of numerical control machine tool intelligent monitoring. Temperature data collected by temperature sensors deployed at multiple positions at multiple time points of a numerical control machine tool are analyzed to know the overall temperature condition of the numerical control machine tool; meanwhile, vibration signals and sound signals collected by the vibration sensor and the sound sensor during operation of the numerical control machine tool are analyzed to know the working states of all parts of the numerical control machine tool, and whether the numerical control machine tool equipment is abnormal or not is judged by combining and considering the overall temperature condition of the numerical control machine tool and the working states of all the parts. Therefore, the state of the numerical control machine tool can be comprehensively and automatically monitored, and the monitoring accuracy and reliability are improved.

Description

Technical Field

[0001] The present application relates to the field of intelligent monitoring of CNC machine tools, and more specifically, to a system and method for real-time monitoring of the equipment status of CNC machine tools based on multiple sensors. Background Art

[0002] CNC machine tool equipment is an automated machine tool equipped with a program control system. It can automatically process the parts to be processed according to a pre-programmed processing program.

[0003] The status of CNC machine tools directly affects the accuracy and quality of machined parts. When a failure occurs during operation, the machined parts will not meet the standards, resulting in waste and economic losses. If the CNC machine tool continues to operate under failure conditions, there will be serious production safety risks. When monitoring the status of CNC machine tools, there are often many limitations when relying solely on traditional technical means, such as limited monitoring point settings, a single data collection method, and a relatively rough data analysis method, or relying solely on manual collection of equipment operation information. Manual collection is not only inefficient, but also easily interfered by subjective factors, which greatly reduces the accuracy and reliability of the data.

[0004] Therefore, an optimized multi-sensor based real-time monitoring solution for CNC machine tool equipment status is needed. Summary of the invention

[0005] In order to solve the above technical problems, the present application is proposed. The embodiments of the present application provide a multi-sensor based CNC machine tool equipment status real-time monitoring system and method.

[0006] According to one aspect of the present application, a multi-sensor based real-time monitoring system for CNC machine tool equipment status is provided, which includes:

[0007] A data collection module related to the status monitoring of a CNC machine tool is used to obtain the temperature values ​​collected by temperature sensors deployed at multiple locations of the CNC machine tool at multiple predetermined time points within a predetermined time period, as well as the vibration signals and sound signals collected by vibration sensors and sound sensors when the CNC machine tool is running;

[0008] A CNC machine tool state monitoring related data encoding module is used to perform feature encoding on the temperature values ​​collected by the temperature sensors at the multiple positions at multiple predetermined time points within a predetermined time period and the vibration signal and the sound signal collected when the CNC machine tool is running to obtain a CNC machine tool equipment state monitoring feature matrix;

[0009] The CNC machine tool state monitoring result generation module is used to determine whether the CNC machine tool has an abnormality based on the CNC machine tool equipment state monitoring feature matrix.

[0010] According to another aspect of the present application, a method for real-time monitoring of CNC machine tool equipment status based on multiple sensors is provided, comprising:

[0011] Acquire temperature values ​​collected by temperature sensors deployed at multiple locations of the CNC machine tool at multiple predetermined time points within a predetermined time period, and vibration signals and sound signals collected by vibration sensors and sound sensors when the CNC machine tool is running;

[0012] Performing feature coding on the temperature values ​​collected by the temperature sensors at the multiple locations at multiple predetermined time points within a predetermined time period and the vibration signal and the sound signal collected when the CNC machine tool is running to obtain a CNC machine tool equipment status monitoring feature matrix;

[0013] Based on the CNC machine tool equipment status monitoring characteristic matrix, it is determined whether the CNC machine tool has an abnormality.

[0014] Compared with the prior art, the multi-sensor based real-time monitoring system and method for CNC machine tool equipment status provided by the present application analyzes the temperature data collected by temperature sensors deployed at multiple locations of the CNC machine tool at multiple time points to understand the overall temperature of the CNC machine tool, and analyzes the vibration signals and sound signals collected by the vibration sensor and the sound sensor during the operation of the CNC machine tool to understand the working status of each component of the CNC machine tool, and considers the overall temperature of the CNC machine tool and the working status of each component to determine whether the CNC machine tool equipment is abnormal. In this way, comprehensive and automated monitoring of the CNC machine tool status can be achieved, improving the accuracy and reliability of monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0016] Figure 1 The system block diagram of the multi-sensor based real-time monitoring system for the status of CNC machine tools according to an embodiment of the present application.

[0017] Figure 2 It is a block diagram of a CNC machine tool status monitoring related data encoding module in a CNC machine tool equipment status real-time monitoring system based on multiple sensors according to an embodiment of the present application.

[0018] Figure 3It is a block diagram of a temperature data feature extraction unit in a multi-sensor based real-time monitoring system for CNC machine tool equipment status according to an embodiment of the present application.

[0019] Figure 4 It is a block diagram of a CNC machine tool status monitoring result generation module in a CNC machine tool equipment status real-time monitoring system based on multiple sensors according to an embodiment of the present application.

[0020] Figure 5 It is a flowchart of a method for real-time monitoring of CNC machine tool equipment status based on multiple sensors according to an embodiment of the present application. DETAILED DESCRIPTION

[0021] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0022] As the cornerstone of modern manufacturing, CNC machine tools have a high degree of automation and can automatically complete complex processing tasks according to pre-programmed programs. The precise operation of such machine tools is crucial to ensuring product quality. However, once a CNC machine tool fails during operation, it will not only cause the precision of the produced parts to decrease and fail to meet quality standards, but may also cause serious safety hazards, waste resources and economic losses.

[0023] In the actual production process, condition monitoring of CNC machine tools is an essential task. Traditional monitoring methods, such as setting limited monitoring points, adopting a single data collection method, and using relatively primitive data analysis technology, often have many shortcomings. These methods may not be able to fully capture the subtle changes in the operation of the machine tool, and thus fail to detect and deal with potential problems in a timely manner.

[0024] In addition, over-reliance on manual collection of equipment status information is not only inefficient, but also easily affected by the operator's subjective judgment, which may reduce the accuracy and reliability of the data. This limitation is particularly obvious in a fast-paced production environment and needs to be overcome by more advanced technical means. Therefore, an optimized multi-sensor-based real-time monitoring solution for CNC machine tool equipment status is expected.

[0025] In recent years, deep learning and neural networks have been widely used in computer vision, natural language processing, text signal processing and other fields. In addition, deep learning and neural networks have also shown a level close to or even beyond that of humans in image classification, object detection, semantic segmentation, text translation and other fields. The development of deep learning and neural networks has provided new solutions and solutions for real-time monitoring of CNC machine tool equipment status.

[0026] Figure 1 : is a system block diagram of a multi-sensor based CNC machine tool equipment status real-time monitoring system according to an embodiment of the present application. Figure 1 As shown, in the multi-sensor-based real-time monitoring system 100 for the equipment status of CNC machine tools, it includes: a CNC machine tool status monitoring related data collection module 110, which is used to obtain the temperature values ​​collected by the temperature sensors deployed at multiple positions of the CNC machine tool at multiple predetermined time points within a predetermined time period, and the vibration signals and sound signals collected by the vibration sensors and sound sensors when the CNC machine tool is running; a CNC machine tool status monitoring related data encoding module 120, which is used to feature encode the temperature values ​​collected by the temperature sensors at multiple positions at multiple predetermined time points within a predetermined time period, and the vibration signals and sound signals collected when the CNC machine tool is running to obtain a CNC machine tool equipment status monitoring feature matrix; a CNC machine tool status monitoring result generation module 130, which is used to determine whether the CNC machine tool has an abnormality based on the CNC machine tool equipment status monitoring feature matrix.

[0027] Specifically, in the technical solution of the present application, first, the temperature values ​​collected by temperature sensors deployed at multiple positions of the CNC machine tool at multiple predetermined time points within a predetermined time period and the vibration signals and sound signals collected by the vibration sensors and sound sensors when the CNC machine tool is running are obtained.

[0028] It should be understood that during the operation of a CNC machine tool, its various components (such as motors, bearings, gearboxes, cutting areas, etc.) will generate heat due to friction, current flow, etc. The distribution and changes of these heat can reflect the working status and potential problems inside the machine tool. Due to the complex structure of the machine tool, the working conditions and thermal characteristics of different components are different, so it is often difficult to fully reflect the overall status of the machine tool through temperature data at a single location. By deploying temperature sensors at multiple key locations, more comprehensive temperature distribution information can be obtained, thereby more accurately assessing the health of the machine tool. Specifically, temperature data is one of the important indicators for judging whether a CNC machine tool has an abnormality. When a component of the machine tool is abnormal (such as bearing wear, motor overheating, cutting area overheating, etc.), the temperature of the component and its surrounding area tends to change. By analyzing the change trend, fluctuation range, and correlation with temperature data at other locations of these temperature data, potential problems of the machine tool, such as overheating and cooling system failure, can be discovered in a timely manner.

[0029] Vibration is an inevitable physical phenomenon during the operation of CNC machine tools. It contains rich information about the motion status of various components inside the machine tool. When a component of the machine tool is abnormal (such as bearing damage, gear wear, imbalance, etc.), its vibration characteristics often change significantly. Therefore, by monitoring the vibration signal of the machine tool during operation, these abnormalities can be discovered and diagnosed in a timely manner. Specifically, the vibration signal contains parameter information such as the vibration frequency, amplitude, and phase of each component of the machine tool. These parameters can reflect the dynamic characteristics and working status of the internal structure of the machine tool. For example, when the bearing is damaged, a specific vibration frequency and amplitude change will occur; imbalance will cause the phase shift of the vibration signal, etc. By analyzing these vibration signals, information related to the machine tool failure can be obtained.

[0030] Sound is another important physical phenomenon generated during the operation of CNC machine tools. It also contains rich information about the working status of various components inside the machine tool. When a part of the machine tool is abnormal (such as poor gear meshing, increased bearing noise, abnormal cutting noise, etc.), its sound characteristics will also change accordingly. The sound signal contains parameter information such as the sound frequency and loudness of each part of the machine tool when it is working. These parameters can reflect the acoustic characteristics and working status of the internal structure of the machine tool. For example, poor gear meshing will produce specific noise frequencies and timbres; increased bearing noise will cause changes in sound loudness, etc. By analyzing these sound signals, information related to machine tool abnormalities can also be obtained.

[0031] In general, the operating status of the CNC machine tool can be obtained by analyzing the temperature data of each position of the CNC machine tool, as well as the vibration signal and sound signal during CNC operation.

[0032] Figure 2 FIG. 1 is a block diagram of a data encoding module for CNC machine tool status monitoring in a multi-sensor based real-time monitoring system for CNC machine tool equipment status according to an embodiment of the present application. Figure 2 As shown, the CNC machine tool state monitoring related data encoding module 120 includes: a temperature data feature extraction unit 121, which is used to extract features from the temperature values ​​collected by the temperature sensors at the multiple positions at multiple predetermined time points within a predetermined time period to obtain a topological global temperature feature matrix; a vibration data feature extraction unit 122, which is used to extract features from the vibration signal to obtain a vibration signal waveform feature matrix; a sound data feature extraction unit 123, which is used to extract features from the sound signal to obtain a sound signal waveform feature matrix; a multi-source data feature fusion unit 124, which is used to fuse the topological global temperature feature matrix, the vibration signal waveform feature matrix and the sound signal waveform feature matrix to obtain the CNC machine tool equipment state monitoring feature matrix.

[0033] Figure 3FIG. 1 is a block diagram of a temperature data feature extraction unit in a multi-sensor based CNC machine tool equipment status real-time monitoring system according to an embodiment of the present application. Figure 3 As shown, the temperature data feature extraction unit 121 includes: a temperature data structuring subunit 1211, which is used to arrange the temperature values ​​collected by the temperature sensors at the multiple locations at multiple predetermined time points within a predetermined time period according to the time dimension to obtain multiple temperature time series input vectors; a temperature data time series feature generation subunit 1212, which is used to pass the multiple temperature time series input vectors through a temperature time series feature encoder to obtain multiple temperature time series feature vectors; a feature vector two-dimensional arrangement subunit 1213, which is used to arrange the multiple temperature time series feature vectors in two dimensions to obtain a global temperature time series feature matrix; a spatial topology The matrix construction subunit 1214 is used to construct the spatial topological matrix of the multiple temperature sensors, the value of each non-diagonal position in the spatial topological matrix is ​​the distance between the corresponding two temperature sensors, and the value of each diagonal position in the spatial topological matrix is ​​zero; the spatial topological feature extraction subunit 1215 is used to pass the spatial topological matrix through a spatial topological feature encoder to obtain a spatial topological feature matrix; the topological temperature feature generation subunit 1216 is used to pass the global temperature time series feature matrix and the spatial topological matrix through a graph neural network model to obtain the topological global temperature feature matrix.

[0034] During the operation of the CNC machine tool, the temperature data changes dynamically over time. In order to clearly reflect the temperature change trend of each part of the machine tool over time and better understand the operating status of the CNC machine tool, in the technical solution of the present application, the temperature values ​​collected by the temperature sensors at multiple positions at multiple predetermined time points within a predetermined time period are arranged according to the time dimension to obtain multiple temperature time series input vectors. The temperature data arranged according to the time dimension can maintain the time sequence and continuity of the data, avoiding the decrease in monitoring accuracy due to data confusion or loss. At the same time, the orderly data arrangement also helps the subsequent data processing and analysis work, and improves the accuracy and reliability of the overall monitoring system.

[0035] Each temperature time series input vector not only contains the specific temperature values ​​at each time point, but also implies dynamic information such as the temperature change trend, fluctuation range, and periodicity over time. When the machine tool is operating normally, the temperature of each component should be kept within a certain range and show a relatively stable change trend. However, once the machine tool is abnormal, such as motor overheating, bearing wear, cooling system failure, etc., these failures often cause abnormal changes in the temperature of the corresponding components. Therefore, it is necessary to analyze the temperature time series input vector to timely discover temperature changes, thereby characterizing the abnormal operation state of the CNC machine tool. Specifically, the multiple temperature time series input vectors are passed through the temperature time series feature encoder to obtain multiple temperature time series feature vectors. In this application, the temperature time series feature encoder adopts a convolutional neural network model including a one-dimensional convolution layer and a fully connected layer. The one-dimensional convolution layer can automatically extract local features from the temperature time series input vector, which are usually closely related to the temperature change trend, fluctuation pattern, etc.; the fully connected layer is located after the convolution layer, and is used to map the feature vector extracted by the convolution layer to the final output space. In the fully connected layer, all neurons are connected to the neurons in the previous layer to achieve feature integration. In general, using a convolutional neural network model containing a one-dimensional convolutional layer and a fully connected layer as a temperature time series feature encoder can automatically extract rich feature information from the temperature time series input vector, thereby better understanding the state of the CNC machine tool during operation.

[0036] During the operation of CNC machine tools, temperature sensors are deployed at multiple locations to monitor the temperature status of different components. The temperature sensor at each location will collect temperature data. By analyzing the data collected by the temperature sensor at each location, a temperature time series feature vector can be obtained, which reflects the temperature change characteristics of each location. In order to integrate the temperature information from multiple locations of the CNC machine tool, it is helpful to understand the temperature distribution of the machine tool from a global perspective. It is necessary to arrange the multiple temperature time series feature vectors in two dimensions to obtain a global temperature time series feature matrix.

[0037] The complex structure of a CNC machine tool determines that the position distribution of its various components in space is diverse. Temperature sensors are arranged at different positions of the machine tool to monitor the temperature status of each component. When performing CNC machine tool status monitoring, in addition to paying attention to the time variation of temperature data, it is also necessary to consider the distribution and association of temperature data in space. Therefore, in the technical solution of the present application, a spatial topological matrix of the multiple temperature sensors is constructed, and the value of each position on the non-diagonal position in the spatial topological matrix is ​​the distance between the corresponding two temperature sensors, and the value of each position on the diagonal in the spatial topological matrix is ​​zero.

[0038] Although the spatial topology matrix intuitively represents the spatial position relationship between the temperature sensors, the information contained therein may be relatively primitive and redundant. In order to automatically extract the temperature spatial features that are more critical to judging the operating state of the CNC machine tool, it is necessary to pass the spatial topology matrix through the spatial topology feature encoder to obtain the spatial topology feature matrix. Through the processing of the spatial topology feature encoder, features such as the spatial distribution pattern of temperature changes and the temperature correlation of different regions can be obtained. In this application, the spatial topology feature encoder adopts a convolutional neural network model using a two-dimensional convolution kernel. Those of ordinary skill in the art should know that the neural network (CNN) using a two-dimensional convolution kernel performs well when processing data with a spatial structure (such as images, grid data, etc.). The spatial topology matrix is ​​essentially a two-dimensional grid data, which contains the spatial position relationship between the temperature sensors. Therefore, the CNN using a two-dimensional convolution kernel can capture the feature information in this spatial structure well.

[0039] Next, the global temperature time series feature matrix and the spatial topology matrix are passed through the graph neural network model to obtain the topological global temperature feature matrix. It should be understood that the global temperature time series feature matrix contains the temperature state information of the CNC machine tool at different time points, while the spatial topology matrix reflects the spatial position relationship between the temperature sensors. Combining the two, the temperature changes of the machine tool in both time and space dimensions can be fully captured. The graph neural network model can simultaneously process the characteristics of the nodes (sensors) (temperature time series features) and the connection relationship between the nodes (spatial topology), thereby integrating time and space information to provide more comprehensive data support for subsequent abnormal state detection. Graph Neural Network (GNN) is a neural network model specifically used to process graph structure data. It expands the traditional neural network model so that it can process nodes and edges in the graph structure, thereby capturing the complex relationship and information flow between nodes. The basic idea of ​​the graph neural network is to update the representation of the node through the node information propagated in the graph, thereby realizing the representation learning of the entire graph.

[0040] The vibration signal contains rich information about the operating status of the machine tool, including load, wear and tear. However, the original vibration signal usually contains a lot of noise and redundant information, and it is difficult to accurately determine whether the machine tool is abnormal by direct analysis. By processing the waveform of the vibration signal, the key feature information contained in the vibration signal can be extracted, that is, the waveform of the vibration signal is passed through the waveform feature encoder to obtain the vibration signal waveform feature matrix. By extracting the feature information in the waveform, it can be determined whether the machine tool is abnormal. For example, when the machine tool has problems such as imbalance, looseness or wear, the waveform of its vibration signal will change, which is manifested as irregular waveform, increased amplitude, etc. In this application, the waveform feature encoder adopts a convolutional neural network model using convolution kernels of different sizes, wherein the size of the first convolution kernel is larger than the size of the second convolution kernel. The first convolution kernel is larger in size and can capture the global features and larger scale changes in the vibration signal waveform. The second convolution kernel is smaller in size and can further extract finer and more local features based on the features extracted by the first convolution kernel. This multi-scale feature extraction method can more comprehensively reflect the characteristics of vibration signals and improve the richness and accuracy of feature information.

[0041] The sound signal contains various acoustic features of the machine tool during operation, such as the friction sound of mechanical parts, the impact sound, the hum of the motor, etc. These sound signals can reflect the key information such as the operating status, load condition, and degree of wear of the machine tool. In order to extract key features from this information to characterize whether the CNC machine tool is operating normally, in the technical solution of the present application, the waveform of the sound signal is passed through the waveform feature encoder to obtain the waveform feature matrix of the sound signal. It should be understood that the first convolution kernel in the waveform feature encoder is larger in size and can capture global features and larger acoustic events; while the second convolution kernel is smaller in size, it can extract local features and detail information at a finer scale. Specifically, the first convolution kernel first performs coarse-grained feature extraction on the original sound signal to remove noise and redundant information; the second convolution kernel performs more refined feature learning on this basis. This hierarchical feature learning method helps the model to gradually and deeply understand the essential characteristics of the sound signal and improve the accuracy and effectiveness of feature extraction.

[0042] Considering that the topological global temperature characteristic matrix reflects the temperature conditions of multiple positions of the CNC machine tool, the vibration signal waveform characteristic matrix reflects the wear, looseness, imbalance and other conditions of the mechanical parts of the CNC machine tool, and the sound signal waveform characteristic matrix reflects the friction, impact and other conditions of the mechanical parts of the CNC machine tool, these three characteristic matrices comprehensively describe the machine tool status from different dimensions. In order to more comprehensively reflect the overall operating status of the CNC machine tool, the topological global temperature characteristic matrix, the vibration signal waveform characteristic matrix and the sound signal waveform characteristic matrix are fused in the technical solution of the present application to obtain the CNC machine tool equipment status monitoring characteristic matrix. The fused characteristic matrix contains richer information, so that more factors can be considered when judging whether the machine tool is abnormal, thereby improving the accuracy and reliability of the judgment. And because multiple characteristic matrices are fused, the system's sensitivity to single sensor failure or signal interference is reduced, and the robustness and stability of the system are enhanced.

[0043] Figure 4 FIG. 1 is a block diagram of a CNC machine tool status monitoring result generation module in a CNC machine tool equipment status real-time monitoring system based on multiple sensors according to an embodiment of the present application. Figure 4 As shown, the CNC machine tool state monitoring result generation module 130 includes: a feature matrix expansion unit 131, used to expand the CNC machine tool equipment state monitoring feature matrix to obtain a CNC machine tool equipment state monitoring feature vector; a feature vector optimization unit 132, used to perform acceptable group distribution constraints on the CNC machine tool equipment state monitoring vector based on statistical characteristics to obtain an optimized CNC machine tool equipment state monitoring vector; a feature vector classification unit 133, used to pass the optimized CNC machine tool equipment state monitoring feature vector through a CNC machine tool state classifier to obtain a classification result, and the classification result is used to indicate whether the CNC machine tool has an abnormality.

[0044] Considering that the CNC machine tool equipment state monitoring feature matrix has a high dimension, which will lead to an increase in computational complexity, in order to reduce the computational complexity and improve the computational performance of the model, the CNC machine tool equipment state monitoring feature matrix needs to be expanded in the technical solution of the present application to obtain a CNC machine tool equipment state monitoring feature vector. By expanding the two-dimensional feature matrix to obtain a one-dimensional feature vector, the feature data is reduced in dimension while retaining the key information in the original feature matrix, and the classification results will not be affected.

[0045] In particular, the state monitoring of CNC machine tools is a complex process involving data collected by multiple sensors and complex data processing processes. In particular, temperature, vibration and sound sensors collect different types of data respectively, which may generate noise and abnormal values ​​due to factors such as the accuracy of the sensor, environmental factors, and slight changes in the equipment itself. In addition, the temperature time series feature encoder and the waveform feature encoder need to convert the original data into feature vectors. This conversion process may cause disturbances in the eigenvalues ​​due to the limitations of the algorithm and nonlinear characteristics. That is, in the technical solution of the present application, considering that the neural network model performs feature mining on the input CNC machine tool equipment state data based on the network parameter matrix, if the inherent implicit association between the network output feature and the network parameter matrix can be further considered, the prior information of the neural network model can be further used to modulate the characteristic manifold of the CNC machine tool equipment state monitoring feature vector to reduce the possibility of special instance disturbances of the eigenvalues. Therefore, the CNC machine tool equipment state monitoring vector is subjected to acceptable constraints based on statistical characteristics to obtain an optimized CNC machine tool equipment state monitoring feature vector.

[0046] Specifically, in an embodiment of the present application, the CNC machine tool equipment state monitoring vector is subjected to acceptable constraints on the population distribution based on statistical characteristics to obtain an optimized CNC machine tool equipment state monitoring vector, including: calculating the mean and variance of the CNC machine tool equipment state monitoring vector; calculating the difference between the eigenvalue of a predetermined position of the CNC machine tool equipment state monitoring vector and the mean of the CNC machine tool equipment state monitoring vector to obtain a first intermediate calculated value; multiplying the network parameter matrix by the CNC machine tool equipment state monitoring vector and then subtracting the network parameter matrix from the CNC machine tool equipment state monitoring vector by position to obtain a first intermediate vector, and calculating the Frobenius norm of the first intermediate vector to obtain a first norm value; adding the first norm value to the matrix; The variance of the CNC machine tool equipment state monitoring vector is calculated and then a square root operation is performed to obtain a second intermediate calculation value; the first intermediate calculation value is divided by the second intermediate calculation value to obtain a third intermediate calculation value; the squares of each position of the CNC machine tool equipment state monitoring vector are calculated and then accumulated to obtain a first sum value; the square root of the first sum value is divided by the scale of the network parameter matrix and then the natural exponential function value is calculated to obtain a fourth intermediate calculation value, wherein the scale of the network parameter matrix is ​​equal to the product between the width value and the height value of the network parameter matrix; the fourth intermediate calculation value is multiplied by the weighted hyperparameter and then added to the third intermediate calculation value to obtain the eigenvalue of the predetermined position of the optimized CNC machine tool equipment state monitoring vector.

[0047] In an embodiment of the present application, another expression of subjecting the CNC machine tool equipment state monitoring vector to acceptable group distribution constraints based on statistical characteristics to obtain the optimized CNC machine tool equipment state monitoring vector may be: processing the CNC machine tool equipment state monitoring vector with the following formula to obtain the optimized CNC machine tool equipment state monitoring vector; wherein the formula is:

[0048]

[0049] Where V is the CNC machine tool equipment state monitoring vector, M is the network parameter matrix, β represents the weighted hyperparameter, v i is the eigenvalue of the ith position of the CNC machine tool equipment state monitoring vector, N is the scale of the network parameter matrix, the scale of the network parameter matrix is ​​equal to the product of the width value and the height value of the network parameter matrix, L is the length of the CNC machine tool equipment state monitoring vector, ||·|| F represents the F norm of the feature vector, μ represents the mean value of the CNC machine tool equipment state monitoring vector, σ 2 represents the variance of the CNC machine tool equipment state monitoring vector, exp(·) represents an exponential function with the natural constant e as the base, v' i represents the eigenvalue of the i-th position of the optimized CNC machine tool equipment state monitoring vector, Represents matrix multiplication.

[0050] It should be understood that in the technical solution of the present application, considering that the neural network model performs feature mining on the input data based on the network parameter matrix, if the inherent implicit association between the network output feature and the network parameter matrix can be further considered, the prior information of the neural network model can be further used to modulate the characteristic manifold of the CNC machine tool equipment state monitoring vector to reduce the possibility of the occurrence of special instance disturbances of its characteristic values. Based on this, in a preferred embodiment, the CNC machine tool equipment state monitoring vector is subjected to a group distribution acceptability constraint based on statistical features, which first maps the CNC machine tool equipment state monitoring vector to the high-dimensional feature space where the network parameter matrix is ​​located, and uses the compatibility perception factor between the mapping result and the original input as the inherent implicit association descriptor between the network output feature and the network parameter matrix. Then, the statistical features of the feature group of the CNC machine tool equipment state monitoring vector are used as the inherent modal information of the CNC machine tool equipment state monitoring vector, and the inherent implicit association descriptor is used as the instance disturbance differential adjustment variable, and the attribute values ​​of each position of the CNC machine tool equipment state monitoring vector are dynamically normalized. At the same time, the class-standardized modulus data of the CNC machine tool equipment state monitoring vector is used as a bias item to constrain the acceptable range of the group distribution of the CNC machine tool equipment state monitoring vector. In this way, the attribute influence of the interference distribution of abnormal samples can be suppressed as much as possible, thereby improving the expression accuracy of the CNC machine tool equipment state monitoring vector.

[0051] Finally, the optimized CNC machine tool equipment state monitoring feature vector is passed through the CNC machine tool state classifier to obtain a classification result, and the classification result is used to indicate whether the CNC machine tool has an abnormality. Specifically, the CNC machine tool state classifier is a model based on a machine learning algorithm, and its main function is to take the equipment state monitoring feature vector as input, and output a classification result through complex internal calculations and analysis. This classification result is used to indicate whether the CNC machine tool currently has an abnormality. In this way, real-time monitoring and abnormality detection of the CNC machine tool state can be achieved, providing a strong guarantee for the smooth progress of the production process.

[0052] In summary, the multi-sensor-based real-time monitoring system 100 for the status of CNC machine tools based on the embodiment of the present application is explained, which analyzes the temperature data collected by the temperature sensors deployed at multiple locations of the CNC machine tools at multiple time points to understand the overall temperature of the CNC machine tools, and analyzes the vibration signals and sound signals collected by the vibration sensors and sound sensors during the operation of the CNC machine tools to understand the working status of each component of the CNC machine tools, and combines the overall temperature of the CNC machine tools and the working status of each component to determine whether the CNC machine tools have abnormalities. In this way, comprehensive and automated monitoring of the status of CNC machine tools can be achieved, improving the accuracy and reliability of monitoring.

[0053] As described above, the multi-sensor-based real-time monitoring system 100 for the status of CNC machine tools according to the embodiment of the present application can be implemented in various terminal devices, such as a server for real-time monitoring of the status of CNC machine tools based on multi-sensors, etc. In one example, the multi-sensor-based real-time monitoring system 100 for the status of CNC machine tools according to the embodiment of the present application can be integrated into the terminal device as a software module and / or a hardware module. For example, the multi-sensor-based real-time monitoring system 100 for the status of CNC machine tools can be a software module in the operating system of the terminal device, or can be an application developed for the terminal device; of course, the multi-sensor-based real-time monitoring system 100 for the status of CNC machine tools can also be one of the many hardware modules of the terminal device.

[0054] Alternatively, in another example, the multi-sensor based real-time monitoring system 100 for the status of CNC machine tools and the terminal device may be separate devices, and the multi-sensor based real-time monitoring system 100 for the status of CNC machine tools may be connected to the terminal device via a wired and / or wireless network and transmit interactive information in accordance with an agreed data format.

[0055] Figure 5 Flow chart of a method for real-time monitoring of CNC machine tool equipment status based on multiple sensors according to an embodiment of the present application. Figure 5 As shown, in the real-time monitoring method of CNC machine tool equipment status based on multiple sensors, the method includes: S110, obtaining temperature values ​​collected by temperature sensors deployed at multiple positions of the CNC machine tool at multiple predetermined time points within a predetermined time period, and vibration signals and sound signals collected by vibration sensors and sound sensors when the CNC machine tool is running; S120, feature encoding the temperature values ​​collected by the temperature sensors at multiple positions at multiple predetermined time points within a predetermined time period, and the vibration signals and sound signals collected when the CNC machine tool is running to obtain a CNC machine tool equipment status monitoring feature matrix; S130, judging whether the CNC machine tool has an abnormality based on the CNC machine tool equipment status monitoring feature matrix.

[0056] Here, those skilled in the art can understand that the specific operations of each step in the above-mentioned multi-sensor based real-time monitoring method for CNC machine tool equipment status have been referred to above. Figures 1 to 4 The description of the real-time monitoring system for the equipment status of CNC machine tools based on multiple sensors has been introduced in detail, and therefore, its repeated description will be omitted.

[0057] In summary, a multi-sensor-based real-time monitoring method for the status of CNC machine tools based on the embodiment of the present application is explained, which analyzes the temperature data collected by temperature sensors deployed at multiple locations of the CNC machine tools at multiple time points to understand the overall temperature of the CNC machine tools, and analyzes the vibration signals and sound signals collected by the vibration sensors and sound sensors during the operation of the CNC machine tools to understand the working status of each component of the CNC machine tools, and considers the overall temperature of the CNC machine tools and the working status of each component to determine whether the CNC machine tools have abnormalities. In this way, comprehensive and automated monitoring of the status of CNC machine tools can be achieved, improving the accuracy and reliability of monitoring.

Claims

1. A multi-sensor based real-time monitoring system for CNC machine tool equipment status, characterized in that: include: A data collection module related to the status monitoring of a CNC machine tool is used to obtain the temperature values ​​collected by temperature sensors deployed at multiple locations of the CNC machine tool at multiple predetermined time points within a predetermined time period, as well as the vibration signals and sound signals collected by vibration sensors and sound sensors when the CNC machine tool is running; A CNC machine tool state monitoring related data encoding module is used to perform feature encoding on the temperature values ​​collected by the temperature sensors at the multiple positions at multiple predetermined time points within a predetermined time period and the vibration signal and the sound signal collected when the CNC machine tool is running to obtain a CNC machine tool equipment state monitoring feature matrix; The CNC machine tool state monitoring result generation module is used to determine whether the CNC machine tool has an abnormality based on the CNC machine tool equipment state monitoring feature matrix.

2. The multi-sensor based real-time monitoring system for CNC machine tool equipment status according to claim 1 is characterized in that: The CNC machine tool state monitoring related data encoding module includes: A temperature data feature extraction unit, used for performing feature extraction on the temperature values ​​collected by the temperature sensors at the multiple locations at multiple predetermined time points within a predetermined time period to obtain a topological global temperature feature matrix; A vibration data feature extraction unit, used for extracting features from the vibration signal to obtain a vibration signal waveform feature matrix; A sound data feature extraction unit, used for extracting features from the sound signal to obtain a sound signal waveform feature matrix; A multi-source data feature fusion unit is used to fuse the topological global temperature feature matrix, the vibration signal waveform feature matrix and the sound signal waveform feature matrix to obtain the CNC machine tool equipment status monitoring feature matrix.

3. The multi-sensor based real-time monitoring system for CNC machine tool equipment status according to claim 2 is characterized in that: The temperature data feature extraction unit comprises: a temperature data structuring subunit, configured to arrange the temperature values ​​collected by the temperature sensors at the plurality of locations at a plurality of predetermined time points within a predetermined time period according to a time dimension to obtain a plurality of temperature time series input vectors; a temperature data time series feature generating subunit, configured to pass the plurality of temperature time series input vectors through a temperature time series feature encoder to obtain a plurality of temperature time series feature vectors; A feature vector two-dimensional arrangement subunit, used for two-dimensionally arranging the plurality of temperature time series feature vectors to obtain a global temperature time series feature matrix; A spatial topology matrix construction subunit, used to construct a spatial topology matrix of the plurality of temperature sensors, wherein the value of each non-diagonal position in the spatial topology matrix is ​​the distance between the corresponding two temperature sensors, and the value of each diagonal position in the spatial topology matrix is ​​zero; A spatial topological feature extraction subunit, used for passing the spatial topological matrix through a spatial topological feature encoder to obtain a spatial topological feature matrix; The topological temperature feature generating subunit is used to obtain the topological global temperature feature matrix by passing the global temperature time series feature matrix and the spatial topological matrix through a graph neural network model.

4. The multi-sensor based real-time monitoring system for CNC machine tool equipment status according to claim 3 is characterized in that: The vibration data feature extraction unit is used to: pass the waveform of the vibration signal through a waveform feature encoder to obtain the vibration signal waveform feature matrix.

5. The multi-sensor based real-time monitoring system for CNC machine tool equipment status according to claim 4 is characterized in that: The sound data feature extraction unit is used to: pass the waveform of the sound signal through the waveform feature encoder to obtain the sound signal waveform feature matrix.

6. The multi-sensor based real-time monitoring system for CNC machine tool equipment status according to claim 5 is characterized in that: The temperature time series feature encoder is a convolutional neural network model including a one-dimensional convolutional layer and a fully connected layer, and the spatial topology feature encoder is a convolutional neural network model using a two-dimensional convolutional kernel.

7. The multi-sensor based real-time monitoring system for CNC machine tool equipment status according to claim 6 is characterized in that: The waveform feature encoder is a convolutional neural network model using convolution kernels of different sizes, wherein the size of the first convolution kernel is larger than the size of the second convolution kernel.

8. The multi-sensor based real-time monitoring system for CNC machine tool equipment status according to claim 7 is characterized in that: The CNC machine tool state monitoring result generation module comprises: A feature matrix expansion unit, used for expanding the CNC machine tool equipment state monitoring feature matrix to obtain a CNC machine tool equipment state monitoring feature vector; A feature vector optimization unit, used for performing group distribution acceptability constraints on the numerical control machine tool equipment state monitoring vector based on statistical features to obtain an optimized numerical control machine tool equipment state monitoring vector; The feature vector classification unit is used to pass the optimized CNC machine tool equipment status monitoring feature vector through a CNC machine tool status classifier to obtain a classification result, and the classification result is used to indicate whether the CNC machine tool has an abnormality.

9. The multi-sensor based real-time monitoring system for CNC machine tool equipment status according to claim 8 is characterized in that: The feature vector optimization unit is used for: Calculating the mean and variance of the numerical control machine tool equipment state monitoring vector; Calculating the difference between the characteristic value of the predetermined position of the numerical control machine tool equipment state monitoring vector and the mean value of the numerical control machine tool equipment state monitoring vector to obtain a first intermediate calculation value; Multiplying the network parameter matrix by the numerical control machine tool equipment state monitoring vector and then subtracting the matrix from the numerical control machine tool equipment state monitoring vector by position to obtain a first intermediate vector, and calculating the Frobenius norm of the first intermediate vector to obtain a first norm value; Adding the first norm value to the variance of the numerical control machine tool equipment state monitoring vector and then performing a square root operation to obtain a second intermediate calculation value; dividing the first intermediate calculated value by the second intermediate calculated value to obtain a third intermediate calculated value; Calculating the square of each position of the numerical control machine tool equipment state monitoring vector and then accumulating and summing them to obtain a first sum value; Calculate a natural exponential function value after taking the square root of the first sum value and dividing it by the scale of the network parameter matrix to obtain a fourth intermediate calculation value, wherein the scale of the network parameter matrix is ​​equal to the product of the width value of the network parameter matrix and the height value thereof; The fourth intermediate calculation value is multiplied by the weighted hyperparameter and then added to the third intermediate calculation value to obtain the characteristic value of the predetermined position of the optimized CNC machine tool equipment state monitoring vector.

10. A method for real-time monitoring of CNC machine tool equipment status based on multiple sensors, characterized in that: include: Acquire temperature values ​​collected by temperature sensors deployed at multiple locations of the CNC machine tool at multiple predetermined time points within a predetermined time period, and vibration signals and sound signals collected by vibration sensors and sound sensors when the CNC machine tool is running; Performing feature coding on the temperature values ​​collected by the temperature sensors at the multiple locations at multiple predetermined time points within a predetermined time period and the vibration signal and the sound signal collected when the CNC machine tool is running to obtain a CNC machine tool equipment status monitoring feature matrix; Based on the CNC machine tool equipment status monitoring characteristic matrix, it is determined whether the CNC machine tool has an abnormality.

Images