A method and system for monitoring faults of numerically controlled machine tools

Abstract

The present invention discloses a method and system for monitoring faults of numerically controlled machine tools, and relates to the technical field of data processing. The method comprises: obtaining the operating data of each sub-device during the operation of the numerically controlled machine tool; performing correlation mining on the operating data of each sub-device to obtain the operating correlation relationship of the sub-device; establishing the fault characteristics of the numerically controlled machine tool and the fault-parameter fluctuation correlation relationship of the operating parameters of each sub-device; fusing the sub-device operating correlation relationship with the fault-parameter fluctuation correlation relationship to construct a fault diagnosis engine; inputting the operating data of each sub-device collected in real time into the fault diagnosis engine as an index to obtain a fault identification result, and the fault identification result includes a fault characteristic and a fault probability. The technical problems of low fault diagnosis efficiency and insufficient accuracy in the prior art are solved, and the technical effect of improving the efficiency and accuracy of fault diagnosis is achieved by constructing a fault diagnosis engine and using the operating data collected in real time as an index input for fault diagnosis.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a method and system for monitoring faults of numerically controlled machine tools. Background Art

[0002] With the rapid development of modern manufacturing industry, CNC machine tools, as the core equipment of precision machining, are crucial to ensuring production efficiency and product quality through their stability and reliability. However, during the operation of CNC machine tools, various faults are inevitable due to the influence of various internal and external factors, such as wear, overload, electrical failure, etc. If these faults are not discovered and handled in time, they will not only cause production interruptions and affect production progress, but may also cause greater damage to the machine tool itself and increase maintenance costs. Traditional CNC machine tool fault monitoring methods mainly rely on manual inspections and regular maintenance, which is not only inefficient, but also often difficult to detect potential faults in time. Summary of the invention

[0003] The present application provides a method and system for monitoring faults of numerically controlled machine tools, which solve the technical problems of low fault diagnosis efficiency and insufficient accuracy in the prior art.

[0004] In view of the above problems, the present application provides a method and system for monitoring faults of CNC machine tools.

[0005] In a first aspect of the present application, a method for monitoring faults of a numerically controlled machine tool is provided, the method comprising:

[0006] The operation data of each sub-device during the operation of the CNC machine tool is obtained; the operation data of each sub-device is associated and mined to obtain the sub-device operation association relationship; the fault characteristics of the CNC machine tool and the fault-parameter fluctuation association relationship of the operation parameters of each sub-device are established; the sub-device operation association relationship is fused with the fault-parameter fluctuation association relationship to construct a fault diagnosis engine, the fault diagnosis engine uses the equipment operation data as an index, and uses the fault identification module established by the fused association relationship as a classifier; the operation data of each sub-device collected in real time is input into the fault diagnosis engine as an index to obtain a fault identification result, the fault identification result includes a fault characteristic and a fault probability.

[0007] A second aspect of the present application provides a CNC machine tool fault monitoring system, the system comprising:

[0008] Data acquisition unit: obtains the operation data of each sub-device during the operation of the CNC machine tool; association mining unit: performs association mining on the operation data of each sub-device to obtain the sub-device operation association relationship; association establishment unit: establishes the fault characteristics of the CNC machine tool and the fault-parameter fluctuation association relationship of the operation parameters of each sub-device; fault diagnosis engine construction unit: fuses the sub-device operation association relationship with the fault-parameter fluctuation association relationship to construct a fault diagnosis engine, the fault diagnosis engine uses the equipment operation data as an index, and uses the fault identification module established by the fused association relationship as a classifier; fault diagnosis unit: inputs the real-time collected operation data of each sub-device as an index into the fault diagnosis engine to obtain a fault identification result, the fault identification result includes a fault characteristic and a fault probability.

[0009] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0010] First, the operation data of each sub-device during the operation of the CNC machine tool is obtained. Then, the operation data of each sub-device is associated and mined to obtain the sub-device operation association relationship. Further, the fault characteristics of the CNC machine tool and the fault-parameter fluctuation association relationship of the operation parameters of each sub-device are established. Then, the sub-device operation association relationship and the fault-parameter fluctuation association relationship are fused to build a fault diagnosis engine. The fault diagnosis engine uses the equipment operation data as an index and the fault identification module established by the fused association relationship as a classifier. Finally, the real-time collected operation data of each sub-device is input as an index into the fault diagnosis engine to obtain the fault identification result, which includes the fault characteristics and the fault probability. The technical problems of low fault diagnosis efficiency and insufficient accuracy in the prior art are solved. By building a fault diagnosis engine and using the real-time collected operation data as an index input for fault diagnosis, the technical effect of improving the efficiency and accuracy of fault diagnosis is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0012] Figure 1 A schematic diagram of a method for monitoring a fault of a CNC machine tool provided in an embodiment of the present application;

[0013] Figure 2 A schematic diagram of the structure of a CNC machine tool fault monitoring system provided in an embodiment of the present application.

[0014] Explanation of the reference numerals: data acquisition unit 11 , association mining unit 12 , association relationship establishment unit 13 , fault diagnosis engine construction unit 14 , fault diagnosis unit 15 . DETAILED DESCRIPTION

[0015] The present application solves the technical problems of low efficiency and insufficient accuracy of fault diagnosis in the prior art by providing a method and system for monitoring faults of CNC machine tools.

[0016] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0017] It should be noted that the terms "including" and "having" are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules that are not explicitly listed or are inherent to these processes, methods, products or devices.

[0018] Embodiment 1, as Figure 1 As shown, the present application provides a method for monitoring faults of a CNC machine tool, wherein the method comprises:

[0019] Obtain the operating data of each sub-equipment during the operation of the CNC machine tool.

[0020] During the operation of the CNC machine tool, the operating data of each sub-device (such as the spindle, feed system, cooling system, etc.) is collected in real time through sensors or monitoring devices installed on each sub-device, including but not limited to key parameters such as temperature, vibration, speed, and current.

[0021] The operation data of each sub-device is subjected to association mining to obtain the sub-device operation association relationship.

[0022] Conduct in-depth analysis of the collected operating data of each sub-device to explore the operating correlation between different sub-devices. By analyzing these relationships, we can reveal the potential collaboration mode and operating rules between devices and provide a basis for fault diagnosis. For example, the vibration of the spindle may be related to the motion state of the feed system, or the temperature of the cooling system may be related to the load of the spindle.

[0023] Furthermore, performing association mining on the operation data of each sub-device to obtain the sub-device operation association relationship includes:

[0024] Perform data alignment on the operating data of each sub-device; calculate parameter changes based on the operating data of each sub-device to obtain the operating parameter fluctuation time series; perform synchronization analysis on the operating parameter fluctuation time series of each sub-device according to the data alignment relationship, and calculate the synchronization correlation coefficient between each sub-device according to the synchronization; obtain the sub-device operation correlation relationship based on the synchronization correlation coefficient.

[0025] Specifically, the collected sub-device operating data are aligned according to timestamps to ensure the time synchronization of the data; the change in the time series of the operating parameters of each sub-device (such as current, vibration, temperature, etc.) is calculated to generate the operating parameter fluctuation time series of each sub-device.

[0026] Using the aligned data, analyze the synchronization of the operating parameter fluctuation time series between each sub-device. Optionally, use the dynamic time warping algorithm (DTW) to calculate the similarity between the fluctuation time series of two sub-devices. First, construct a distance matrix for the two fluctuation time series, where each element of the matrix represents the difference between the two time series data at the corresponding time point; then, recursively calculate the cumulative distance matrix, and use the formula D(i,j)=d(i,j)+min{D(i-1,j),D(i,j-1),D(i-1,j-1)} to find the minimum matching cost from the starting point to the end point, where D(i,j) represents the minimum cumulative distance from the starting point to the current point, and d(i,j)=|ΔP 1,i -ΔP 2,j |, d(i,j) represents the difference (absolute value) between the fluctuation sequence of sub-device 1 at point i and the fluctuation sequence of sub-device 2 at point j, ΔP 1,i and ΔP 2,j Denote the i-th and j-th data points of sub-device 1 and sub-device 2 in the fluctuation time series respectively, min{D(i-1,j),D(i,j-1),D(i-1,j-1)} denotes the minimum cost of the three possible paths to the current point, D(i-1,j) denotes that sub-device 2 keeps the current time point and sub-device 1 moves forward, D(i,j-1) denotes that sub-device 1 keeps the current time point and sub-device 2 moves forward, and D(i-1,j-1) denotes that both sub-device 1 and sub-device 2 move forward; trace back from the end point (n,m) of the cumulative distance matrix to the starting point (1,1), and find a path that minimizes the cumulative distance of the points passed on the path. This path is the optimal matching path of the two time series data. According to the cumulative distance D(n,m) of the optimal matching path, the similarity (i.e., synchronization correlation coefficient) of the two time series is defined, and the similarity = 1 / [1+D(n,m)]. The closer the similarity is to 1, the higher the synchronization of the two time series.

[0027] The synchronization correlation coefficients between the sub-devices are represented by a graph structure. For example, the sub-devices are used as nodes and the synchronization correlation coefficients are used as edge weights to intuitively display the correlation strength between the devices.

[0028] Furthermore, aligning the operation data of each sub-device includes:

[0029] Obtain the workpiece processing path and collaborative processing relationship; based on the workpiece processing path and collaborative processing relationship, perform a timing node analysis of the operation impact of each device with the earliest operation time as the starting point, and determine the response time difference of each device parameter; based on the response time difference of each device parameter, make corrections according to the collection timestamp of the operation data of each sub-device to determine the timing alignment relationship; align the operation data of each sub-device based on the timing alignment relationship.

[0030] Based on the operation tasks of CNC machine tools, the processing path of the workpiece and the collaborative processing relationship between each sub-device are extracted. The processing path reflects the motion trajectory of the workpiece on the machine tool, while the collaborative processing relationship describes the collaborative mode of the sub-devices at different processing stages, such as the synchronous work of the spindle and the feed system. According to the obtained workpiece processing path and collaborative processing relationship, the operation impact of each sub-device is analyzed with the earliest operation time point as a reference. By monitoring the change trend of the operation data of each device, the response time difference of its operation parameters to the collaborative processing is analyzed. For example, the startup of a certain device may cause other devices to respond within a specific time. Combined with the results of the operation impact timing node analysis, the response time difference of the operation parameters of each sub-device is calculated. It can be determined by comparing the timestamps of the changes in the key parameters of each sub-device (such as temperature, current, vibration, etc.). For example, the startup of sub-device A may cause sub-device B to vibrate after several milliseconds. This time difference is the response time difference. According to the response time difference determined above, the operating data timestamps of each sub-device are corrected to ensure that the data can be arranged according to a unified time axis. For example, the data timestamps of the sub-devices that respond later are shifted forward by the corresponding response time difference to align to the earliest operating time point. According to the corrected timestamps, a timing alignment relationship is established to ensure that the operating data of all sub-devices are consistent on a unified time axis. Finally, based on this alignment relationship, the operating data of each sub-device is aligned to form a time-synchronized data set, providing an accurate data basis for subsequent analysis.

[0031] Furthermore, the parameter variation is calculated according to the operation data of each sub-device to obtain the operation parameter fluctuation time series, including:

[0032] Obtain target control data and monitoring operation data of each sub-device; perform difference calculation based on the target control data and monitoring operation data to obtain parameter difference; perform adjacent time series comparison based on the parameter difference to obtain parameter change; integrate based on the time series relationship of the parameter change to obtain the operation parameter fluctuation time series.

[0033] Extract the target control data (such as speed setting value, power target value, etc.) and actual monitoring operation data (such as real-time speed, power output, etc.) of each sub-device from the system. The target control data represents the expected value of the device under the ideal operating state, while the monitoring operation data reflects the real-time state in actual operation. Based on the monitoring operation data and target control data of the sub-device, calculate their difference point by point to obtain the parameter difference, which is used to quantify the deviation between the device operating state and the target state. For each sub-device parameter difference sequence, calculate the change between adjacent time points (i.e., the parameter difference between the current time point and the previous time point) to obtain the parameter change. Integrate the parameter change of each sub-device in chronological order to generate the operating parameter fluctuation time series of the device.

[0034] Furthermore, obtaining the sub-device operation association relationship according to the synchronization association coefficient includes:

[0035] According to the synchronization association coefficient, the synchronization relationship sub-devices and their association coefficients are extracted; each sub-device is used as a graph node, and the node connection is established according to the synchronization relationship sub-device to construct an edge, and the association coefficient is set as the edge weight; for each node, the number of connections of the node is calculated to construct a degree matrix; according to the edges and edge weights, an adjacency matrix is ​​constructed; the difference calculation is performed through the degree matrix and the adjacency matrix to construct a Laplace matrix; the eigenvalues ​​and eigenvectors of the Laplace matrix are solved to extract the association relationship between the sub-devices, and the corresponding eigenvalues ​​are used to represent the association strength between the nodes to obtain the sub-device operation association relationship.

[0036] Preferably, according to the synchronization correlation coefficient, sub-device pairs with significant synchronization relationships are screened out, and the corresponding correlation coefficients are extracted. For example, sub-device pairs with synchronization correlation coefficients greater than a set threshold are considered to have significant correlation relationships; each sub-device is used as a node of the graph, and connections (edges) are established between sub-device pairs with synchronization relationships, and the synchronization correlation coefficient is set as the weight of the edge, thus forming a weighted undirected graph, in which the nodes represent sub-devices and the edge weights reflect the synchronization correlation strength; for each node, the number of connections with other nodes is counted to generate a degree matrix D, which is a diagonal matrix, and the diagonal elements of the matrix D ii Represents the degree of node i (i.e., the number of connections); constructs the adjacency matrix A according to the connection relationship and edge weight of the edge in the graph. The adjacency matrix A i，jRepresents the edge weight between node i and node j. If i and j are not connected, then A i，j =0; the Laplace matrix L is calculated through the degree matrix D and the adjacency matrix A. The Laplace matrix is ​​the basic feature description of the graph and is used to represent the association characteristics between nodes; the eigenvalues ​​and eigenvectors of the Laplace matrix L are solved. The eigenvalues ​​reflect the association strength between the nodes in the graph, while the eigenvectors describe the complex association patterns between sub-devices; according to the eigenvalues ​​and eigenvectors of the Laplace matrix, the association relationship between sub-devices is analyzed, and the eigenvectors corresponding to smaller eigenvalues ​​can reveal the strong association between nodes; the eigenvalues ​​are used as indicators to quantify the synchronization association strength between sub-devices, and the results are used as the final output of the sub-device operation association relationship.

[0037] Establish the fault-parameter fluctuation correlation between the fault characteristics of CNC machine tools and the operating parameters of each sub-equipment.

[0038] According to the known fault characteristics of CNC machine tools (such as mechanical faults, electrical faults, other faults, etc.), combined with the operating parameters of sub-equipment, the correlation between fault characteristics and parameter fluctuations is established.

[0039] Furthermore, the fault-parameter fluctuation correlation between the fault characteristics of CNC machine tools and the operating parameters of each sub-equipment is established, including:

[0040] A numerical control machine tool fault database is constructed, which includes fault types, fault probabilities and corresponding fault characteristics and equipment operation parameter characteristics; the fault types are used as primary classification characteristics and the fault probabilities are used as secondary classification characteristics to classify and identify the numerical control machine tool fault database; similar fault data are extracted based on the classification identification to construct similar fault data sets, and the correlation relationship between fault characteristics and equipment operation parameter characteristics is fitted for the similar fault data sets to obtain the fault-parameter fluctuation correlation relationship, wherein the fault-parameter fluctuation correlation relationship has a classification label.

[0041] Specifically, historical operation data and fault records are sorted and collected to build a CNC machine tool fault database, which includes fault types (such as spindle overheating, tool wear, motor overload, etc.), fault probabilities (frequency or probability of faults based on historical data) and corresponding fault characteristics (core indicators reflecting the fault status, such as abnormal vibration amplitude, excessive temperature, etc.), equipment operation parameter characteristics (fault-related sub-equipment operation data fluctuation patterns, such as current, voltage, speed, displacement, etc.); fault type is used as the first-level classification feature, for example, mechanical failure and electrical failure, and fault probability is used as the second-level classification feature to further subdivide the classification of high probability, medium probability and low probability faults in each type of fault to form a clear grouping structure, for example, first-level classification: "spindle failure", second-level classification: high probability "spindle overheating", medium probability "spindle wear", low probability "spindle jam". According to the classification identification, data with the same classification label is extracted from the fault database to construct a similar fault data set; through statistical regression, machine learning or association rule mining and other methods, the correlation between fault characteristics and equipment operating parameter characteristics is fitted, and the correlation between parameter fluctuation patterns and fault types is identified. Optionally, a linear or nonlinear regression relationship between operating parameter changes (such as temperature rise, current fluctuations) and fault characteristics (such as spindle overheating) is established; the fitting results are output as a fault-parameter fluctuation correlation relationship with classification labels, and the corresponding relationship between each fault type and related operating parameters and their fluctuation patterns is clarified.

[0042] The sub-device operation association relationship and the fault-parameter fluctuation association relationship are fused to construct a fault diagnosis engine, wherein the fault diagnosis engine uses the equipment operation data as an index and uses a fault identification module established by the fused association relationship as a classifier.

[0043] The sub-device operation association relationship and the fault-parameter fluctuation association relationship are uniformly modeled, and a comprehensive relationship model is formed by analyzing the correlation between the collaborative characteristics of equipment operation and the fluctuation of fault characteristic parameters; the sub-device operation association relationship provides the collaborative working mode and synchronization characteristics between devices, while the fault-parameter fluctuation association relationship describes the connection between equipment operation parameters and specific fault types; through the matching of association matrix and classification labels, the data structures of the two are unified into a fusion model. The equipment operation data is used as the input index of the fault diagnosis engine, and the fused association relationship model is dynamically called according to the real-time collected operation data; the fault identification module is set as a classifier in the fault diagnosis engine, and the fault identification module maps the real-time data to the corresponding fault characteristic category based on the fused association relationship.

[0044] Furthermore, the sub-device operation association relationship is integrated with the fault-parameter fluctuation association relationship to construct a fault diagnosis engine, including:

[0045] According to the fault-parameter fluctuation correlation relationship, fault identification fitting is performed to establish a first fault identification module, the output result of the first fault identification module is the fault diagnosis result corresponding to the operating parameter characteristics of each sub-device; based on the sub-device operation correlation relationship, the associated fault characteristics in the fault-parameter fluctuation correlation relationship are located, and the sub-device operation correlation relationship is used to perform operation correlation relationship analysis on the operation data of each sub-device, and then through the determined sub-device operation correlation relationship, the fault characteristics are enhanced and identified according to the associated fault characteristics to construct a second fault identification module, and the second fault identification module is a fault diagnosis result determined based on the operation correlation characteristics between the sub-devices; the identification weights of the first fault identification module and the second fault identification module are configured, and the first fault identification module and the second fault identification module are merged based on the identification weights to obtain the fault identification module; the fault identification module is used as a classifier, and the operation data of each sub-device is used as an input index constraint to construct the fault diagnosis engine.

[0046] Based on the fault-parameter fluctuation correlation, the collected operation data is fitted for fault identification to form a first fault identification module. The first fault identification module outputs the preliminary fault diagnosis results corresponding to each sub-device operation parameter by analyzing the correlation between the operation parameter characteristics (such as temperature, vibration, current, etc.) of each sub-device and the fault characteristics. Combined with the sub-device operation correlation, the fault characteristics related to the cooperative work of the sub-devices in the fault-parameter fluctuation correlation are located, and the cooperative characteristics between the sub-devices are modeled through the operation correlation analysis. For example, by analyzing the synchronization and delay characteristics between the sub-devices, the recognition ability of the associated fault characteristics is further enhanced; using this correlation information, the second fault identification module is constructed, and the second fault identification module provides more comprehensive fault diagnosis results based on the operation correlation characteristics between the sub-devices. Weights are assigned to the diagnosis results of the first fault identification module and the second fault identification module. The setting of the identification weight can be determined based on factors such as the recognition accuracy of the module and the matching adaptability of the specific fault; through weighted fusion, the diagnosis results of the two modules are integrated into a unified fault identification module to achieve multi-dimensional diagnosis capabilities. The fused fault identification module is used as the core classifier of the fault diagnosis engine, and the real-time collected sub-device operation data is input as the index constraint to drive the fault diagnosis engine to perform dynamic analysis. The engine realizes accurate identification and positioning of the fault type by combining the operation parameter feature diagnosis results of the first fault identification module and the operation correlation characteristic diagnosis results of the second fault identification module.

[0047] The real-time collected operation data of each sub-device is input into the fault diagnosis engine as an index to obtain a fault identification result, which includes a fault feature and a fault probability.

[0048] The key operating parameters of each sub-device (such as temperature, vibration, current, speed, etc.) are monitored in real time by sensors, and the collected data is preprocessed to ensure its integrity and consistency; the preprocessed data is input into the fault diagnosis engine in a unified time series as an index constraint for the engine analysis. The fault diagnosis engine uses the built-in first fault identification module and second fault identification module to perform comprehensive analysis based on the operating parameter characteristics and the operating correlation relationship of the sub-devices, and dynamically generates fault diagnosis results. The fault identification results include specific fault characteristics (such as spindle overheating, feed system jamming, etc.) and fault probability (confidence expressed as a probability value), so as to accurately identify potential fault types and their impact range.

[0049] In summary, the embodiments of the present application have at least the following technical effects:

[0050] First, the operation data of each sub-device during the operation of the CNC machine tool is obtained. Then, the operation data of each sub-device is associated and mined to obtain the sub-device operation association relationship. Further, the fault characteristics of the CNC machine tool and the fault-parameter fluctuation association relationship of the operation parameters of each sub-device are established. Then, the sub-device operation association relationship and the fault-parameter fluctuation association relationship are fused to build a fault diagnosis engine. The fault diagnosis engine uses the equipment operation data as an index and the fault identification module established by the fused association relationship as a classifier. Finally, the real-time collected operation data of each sub-device is input as an index into the fault diagnosis engine to obtain the fault identification result, which includes the fault characteristics and the fault probability. The technical problems of low fault diagnosis efficiency and insufficient accuracy in the prior art are solved. By building a fault diagnosis engine and using the real-time collected operation data as an index input for fault diagnosis, the technical effect of improving the efficiency and accuracy of fault diagnosis is achieved.

[0051] Embodiment 2, based on the same inventive concept as a method for monitoring a fault of a numerically controlled machine tool in the above embodiment, Figure 2 As shown, the present application provides a CNC machine tool fault monitoring system, wherein the system comprises:

[0052] A data acquisition unit 11 is used to obtain the operation data of each sub-device during the operation of the CNC machine tool; an association mining unit 12 is used to perform association mining on the operation data of each sub-device to obtain the operation association relationship of the sub-device; an association establishment unit 13 is used to establish the fault characteristics of the CNC machine tool and the fault-parameter fluctuation association relationship of the operation parameters of each sub-device; a fault diagnosis engine construction unit 14 is used to fuse the sub-device operation association relationship with the fault-parameter fluctuation association relationship to construct a fault diagnosis engine, wherein the fault diagnosis engine uses the equipment operation data as an index and uses the fault identification module established by the fused association relationship as a classifier; a fault diagnosis unit 15 is used to input the operation data of each sub-device collected in real time into the fault diagnosis engine as an index to obtain a fault identification result, wherein the fault identification result includes a fault characteristic and a fault probability.

[0053] Furthermore, the association mining unit 12 is used to perform the following method:

[0054] Perform data alignment on the operating data of each sub-device; calculate parameter changes based on the operating data of each sub-device to obtain the operating parameter fluctuation time series; perform synchronization analysis on the operating parameter fluctuation time series of each sub-device according to the data alignment relationship, and calculate the synchronization correlation coefficient between each sub-device according to the synchronization; obtain the sub-device operation correlation relationship based on the synchronization correlation coefficient.

[0055] Furthermore, the association mining unit 12 is used to perform the following method:

[0056] Obtain the workpiece processing path and collaborative processing relationship; based on the workpiece processing path and collaborative processing relationship, perform a timing node analysis of the operation impact of each device with the earliest operation time as the starting point, and determine the response time difference of each device parameter; based on the response time difference of each device parameter, make corrections according to the collection timestamp of the operation data of each sub-device to determine the timing alignment relationship; align the operation data of each sub-device based on the timing alignment relationship.

[0057] Furthermore, the association mining unit 12 is used to perform the following method:

[0058] Obtain target control data and monitoring operation data of each sub-device; perform difference calculation based on the target control data and monitoring operation data to obtain parameter difference; perform adjacent time series comparison based on the parameter difference to obtain parameter change; integrate based on the time series relationship of the parameter change to obtain the operation parameter fluctuation time series.

[0059] Furthermore, the association mining unit 12 is used to perform the following method:

[0060] According to the synchronization association coefficient, the synchronization relationship sub-devices and their association coefficients are extracted; each sub-device is used as a graph node, and node connections are established according to the synchronization relationship sub-devices to construct edges, and the association coefficient is set as the edge weight; for each node, the number of connections of the node is calculated to construct a degree matrix; according to the edges and edge weights, an adjacency matrix is ​​constructed; the difference calculation is performed through the degree matrix and the adjacency matrix to construct a Laplace matrix; the eigenvalues ​​and eigenvectors of the Laplace matrix are solved to extract the association relationship between the sub-devices, and the corresponding eigenvalues ​​are used to represent the association strength between the nodes to obtain the sub-device operation association relationship.

[0061] Furthermore, the association relationship establishing unit 13 is used to execute the following method:

[0062] A numerical control machine tool fault database is constructed, which includes fault types, fault probabilities and corresponding fault characteristics and equipment operation parameter characteristics; the fault types are used as primary classification characteristics and the fault probabilities are used as secondary classification characteristics to classify and identify the numerical control machine tool fault database; similar fault data are extracted based on the classification identification to construct similar fault data sets, and the correlation relationship between fault characteristics and equipment operation parameter characteristics is fitted for the similar fault data sets to obtain the fault-parameter fluctuation correlation relationship, wherein the fault-parameter fluctuation correlation relationship has a classification label.

[0063] Furthermore, the fault diagnosis engine construction unit 14 is used to execute the following method:

[0064] According to the fault-parameter fluctuation correlation relationship, fault identification fitting is performed to establish a first fault identification module, the output result of the first fault identification module is the fault diagnosis result corresponding to the operating parameter characteristics of each sub-device; based on the sub-device operation correlation relationship, the associated fault characteristics in the fault-parameter fluctuation correlation relationship are located, and the sub-device operation correlation relationship is used to perform operation correlation relationship analysis on the operation data of each sub-device, and then through the determined sub-device operation correlation relationship, the fault characteristics are enhanced and identified according to the associated fault characteristics to construct a second fault identification module, and the second fault identification module is a fault diagnosis result determined based on the operation correlation characteristics between the sub-devices; the identification weights of the first fault identification module and the second fault identification module are configured, and the first fault identification module and the second fault identification module are merged based on the identification weights to obtain the fault identification module; the fault identification module is used as a classifier, and the operation data of each sub-device is used as an input index constraint to construct the fault diagnosis engine.

[0065] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0066] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

[0067] This specification and the drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, those skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. A method for monitoring faults of a numerically controlled machine tool, characterized in that: The method comprises: Obtain the operation data of each sub-equipment during the operation of the CNC machine tool; Performing association mining on the operation data of each sub-device to obtain the sub-device operation association relationship; Establish the fault-parameter fluctuation correlation between the fault characteristics of CNC machine tools and the operating parameters of each sub-equipment; The sub-device operation association relationship is integrated with the fault-parameter fluctuation association relationship to construct a fault diagnosis engine, wherein the fault diagnosis engine uses the device operation data as an index and uses a fault identification module established by the integrated association relationship as a classifier; The real-time collected operation data of each sub-device is input into the fault diagnosis engine as an index to obtain a fault identification result, which includes a fault feature and a fault probability.

2. The method for monitoring faults of a numerically controlled machine tool according to claim 1, wherein: Performing association mining on the operation data of each sub-device to obtain the sub-device operation association relationship includes: Performing data alignment on the operation data of each sub-device; Calculate the parameter variation according to the operation data of each sub-device to obtain the operation parameter fluctuation time series; According to the data alignment relationship, the synchronization analysis of the operating parameter fluctuation sequence of each sub-device is performed, and the synchronization correlation coefficient between each sub-device is calculated according to the synchronization; The sub-device operation association relationship is obtained according to the synchronization association coefficient.

3. The method for monitoring faults of a numerically controlled machine tool according to claim 2, wherein: Performing data alignment on the operation data of each sub-device includes: Obtain workpiece processing path and collaborative processing relationship; According to the workpiece processing path and the collaborative processing relationship, the operation impact timing node analysis of each device is performed starting from the earliest operation time to determine the response time difference of each device parameter; According to the response time difference of the parameters of each device, correction is performed according to the acquisition timestamp of the operation data of each sub-device to determine the timing alignment relationship; The operation data of each sub-device is aligned based on the timing alignment relationship.

4. The method for monitoring faults of a numerically controlled machine tool according to claim 2, wherein: Calculating parameter changes according to the operation data of each sub-device to obtain the operation parameter fluctuation time series includes: Obtain target control data and monitoring operation data of each sub-device; Perform difference calculation based on the target control data and the monitoring operation data to obtain parameter difference; Perform adjacent time series comparison according to the parameter difference to obtain parameter variation; The time series relationship of the parameter changes is integrated to obtain the operating parameter fluctuation time series.

5. The method for monitoring faults of a numerically controlled machine tool according to claim 3, wherein: Obtaining the sub-device operation association relationship according to the synchronization association coefficient includes: Extracting synchronization-related sub-devices and their association coefficients according to the synchronization association coefficients; Taking each sub-device as a graph node, establishing node connections and building edges according to the synchronization relationship sub-devices, and setting the association coefficient as the edge weight; For each node, calculate the number of connections of the node and construct the degree matrix; Constructing an adjacency matrix according to the edges and edge weights; Performing difference calculations through the degree matrix and the adjacency matrix to construct a Laplace matrix; The eigenvalues ​​and eigenvectors of the Laplace matrix are solved, the association relationship between the sub-devices is extracted, the corresponding eigenvalues ​​are used to represent the association strength between the nodes, and the sub-device operation association relationship is obtained.

6. The method for monitoring faults of a numerically controlled machine tool according to claim 1, wherein: Establish the relationship between the fault characteristics of CNC machine tools and the fault-parameter fluctuation of the operating parameters of each sub-equipment, including: Build a CNC machine tool fault database, which includes fault type, fault probability and corresponding fault characteristics, and equipment operation parameter characteristics; Using the fault type as the primary classification feature and the fault probability as the secondary classification feature, classifying and identifying the numerical control machine tool fault database; Based on the classification identifier, similar fault data is extracted, a similar fault data set is constructed, and the correlation relationship between fault characteristics and equipment operation parameter characteristics is fitted for the similar fault data set to obtain the fault-parameter fluctuation correlation relationship, and the fault-parameter fluctuation correlation relationship has a classification label.

7. The method for monitoring faults of a numerically controlled machine tool according to claim 6, wherein: The sub-device operation association relationship is integrated with the fault-parameter fluctuation association relationship to construct a fault diagnosis engine, including: According to the fault-parameter fluctuation correlation relationship, fault identification fitting is performed to establish a first fault identification module, wherein the output result of the first fault identification module is the fault diagnosis result corresponding to the operating parameter characteristics of each sub-device; Locating the associated fault features in the fault-parameter fluctuation association relationship based on the sub-device operation association relationship, performing operation association analysis on the operation data of each sub-device using the sub-device operation association relationship, and then performing enhanced fault feature identification according to the associated fault features through the determined sub-device operation association relationship, and constructing a second fault identification module, wherein the second fault identification module is a fault diagnosis result determined based on the operation association features between the sub-devices; configuring identification weights of the first fault identification module and the second fault identification module, and fusing the first fault identification module with the second fault identification module based on the identification weights to obtain the fault identification module; The fault identification module is used as a classifier, and the operation data of each sub-device is used as an input index constraint to construct the fault diagnosis engine.

8. A CNC machine tool fault monitoring system, characterized in that: A method for monitoring faults of a numerically controlled machine tool according to any one of claims 1 to 7, the system comprising: Data acquisition unit: obtains the operation data of each sub-equipment during the operation of the CNC machine tool; Association mining unit: performing association mining on the operation data of each sub-device to obtain the association relationship of the sub-device operation; Correlation relationship establishment unit: establishes the fault characteristics of CNC machine tools and the fault-parameter fluctuation correlation relationship between the operating parameters of each sub-equipment; A fault diagnosis engine construction unit: the sub-device operation association relationship is integrated with the fault-parameter fluctuation association relationship to construct a fault diagnosis engine, wherein the fault diagnosis engine uses the device operation data as an index and uses a fault identification module established by the integrated association relationship as a classifier; Fault diagnosis unit: inputs the real-time collected operation data of each sub-device as an index into the fault diagnosis engine to obtain a fault identification result, which includes fault characteristics and fault probability.

Images