Machining fault detection method and equipment

Abstract

The embodiment of the invention relates to the technical field of machining, in particular to a machining fault detection method and device, and the method comprises the steps: obtaining a target signal of a machine tool under a target working condition; through a preset sliding window, the target signal is sampled multiple times, multiple detection signals are determined, and the total length of the multiple detection signals is larger than the length of the target signal; the characteristic value of each detection signal is compared with a preset threshold value boundary to determine whether a machine tool machining fault exists or not, and the threshold value boundary is determined according to a fault-free signal of the machine tool under the target working condition. Based on the method, after the target signal needing to be detected is determined, multiple times of sampling are carried out through the sliding window, so that a plurality of detection signals used for characteristic value comparison are extracted from limited data duration, and the effect of utilizing the data duration to the maximum extent is achieved; the influence on the detection result due to insufficient effective data in the prior art is effectively avoided.

Description

Technical Field

[0001] The embodiments of the present application relate to the technical field of machining, and in particular, to a machining fault detection method and device. Background Art

[0002] In the process of machining such as turning, milling, boring, etc., it is inevitable to encounter various faults, such as tool breakage and chipping, which affect the machining efficiency and quality. Usually, during the machining process, it is necessary to detect the state of the machine tool in real time to give an early warning in time when a machine tool fault occurs and avoid unnecessary losses.

[0003] Currently, the machining fault detection scheme mainly controls through eigenvalue boundaries. However, due to the large number of factors affecting the eigenvalues during machining, it is difficult to explain the eigenvalue fluctuations. Therefore, high-resolution data is required to ensure the reliability of the detection results, resulting in poor detection effects when detecting faults with a small amount of data. In most cases, the effective duration of the data related to the machining process is often insufficient, thus affecting the machining fault detection effect. Summary of the Invention

[0004] An object of the embodiments of the present application is to provide a machining fault detection method to solve the technical problem that the detection effect is affected due to the insufficient effective duration of relevant data when detecting machining faults in related technologies.

[0005] In a first aspect, the embodiments of the present application provide a machining fault detection method, including: obtaining a target signal of a machine tool under a target working condition; performing multiple samplings on the target signal through a preset sliding window to determine a plurality of detection signals, and the total length of the plurality of detection signals is greater than the length of the target signal; comparing the eigenvalue of each detection signal with a preset threshold boundary to determine whether there is a machining fault of the machine tool, and the threshold boundary is determined according to the fault-free signal of the machine tool under the target working condition.

[0006] In combination with the first aspect, in a possible implementation manner, the obtaining the target signal of the machine tool under the target working condition includes: obtaining the machining signal of the machine tool; determining the start machining time and the end machining time of the target working condition; determining the effective machining time between the start machining time and the end machining time according to a preset time interval; identifying the machining signal of the machine tool within the effective machining time, and determining a target time period in which the signal amplitude is within a preset amplitude range; extracting the machining signal within the target time period as the target signal.

[0007] In combination with the first aspect, in a possible implementation, performing data sampling operations on the target signal through a preset sliding window to determine a plurality of detection signals includes: determining the sampling length and sampling interval of the sliding window; extracting the detection signals from the target signal in sequence according to the sampling length and sampling interval of the sliding window, and the length of each detection signal is the sampling length of the sliding window.

[0008] In combination with the first aspect, in a possible implementation, the extracting the detection signals from the target signal in sequence according to the sampling length and sampling interval includes: sequentially determining a plurality of start times and a plurality of end times, where each start time corresponds to an end time, the interval between adjacent start times is the sampling interval, and the interval between each start time and the corresponding end time is the sampling length; extracting the detection signals from the target signal in sequence according to each start time and the corresponding end time.

[0009] In combination with the first aspect, in a possible implementation, comparing the eigenvalue of each detection signal with a preset threshold boundary to determine whether there is a machine tool processing fault includes: when each detection signal is determined, inputting the detection signal into a preset first model, where the first model is used to extract the eigenvalue of the detection signal and compare it with the threshold boundary; determining whether the eigenvalue is less than the threshold boundary according to the output of the first model; if the eigenvalue is less than the threshold boundary, it is determined that there is a machine tool processing fault; if the eigenvalue is greater than the threshold boundary, it is determined that there is no machine tool processing fault.

[0010] In combination with the first aspect, in a possible implementation, before inputting the detection signal into the first model when each detection signal is determined, it further includes: obtaining the historical processing signal of the machine tool; creating a data set according to the historical processing signal; training a preset initial model through the data set to obtain the first model.

[0011] In combination with the first aspect, in a possible implementation, the data set includes a training set and a test set, and the training the preset initial model through the data set to obtain the first model includes: training the initial model according to the training set to obtain a target model; evaluating the target model through the test set and a preset loss function to determine the accuracy of the target model; if the accuracy is lower than a preset accuracy threshold, updating the hyperparameters of the target model to perform multiple iterative trainings on the target model until the accuracy of the target model is higher than the accuracy threshold to obtain the first model.

[0012] In combination with the first aspect, in a possible implementation, before comparing the eigenvalue of each of the detection signals with a preset threshold boundary to determine whether there is a machining fault in the machine tool, it further includes: obtaining a fault-free signal of the machine tool under the target working condition, where the fault-free signal includes a plurality of reference signal segments; performing data sampling operations on each of the reference signal segments according to a preset sampling rate to obtain calibrated signal segments, and the number of the calibrated signal segments is equal to the number of the target data segments; calculating the eigenvalue of each of the calibrated signal segments to determine the threshold boundary.

[0013] In combination with the first aspect, in a possible implementation, the calculating the eigenvalue of each of the calibrated data segments to determine the threshold boundary includes: determining the waveform of each of the calibrated data segments; performing waveform extraction in the waveform of each of the calibrated data segments according to a preset waveform sampling period and waveform sampling length to determine the waveform subset of the calibrated data segment; calculating the eigenvalue of each of the calibrated data segments according to the waveform subset of each of the calibrated data segments; and determining the threshold boundary according to the eigenvalue.

[0014] In a second aspect, an embodiment of the present application further provides a machining fault detection device, including:

[0015] a data acquisition module, configured to acquire a target signal of a machine tool under a target working condition;

[0016] a data sampling module, configured to perform multiple samplings on the target signal through a preset sliding window to determine a plurality of detection signals, and the total length of the plurality of detection signals is greater than the length of the target signal;

[0017] a data comparison module, configured to compare the eigenvalue of each of the detection signals with a preset threshold boundary to determine whether there is a machining fault in the machine tool, and the threshold boundary is determined according to the fault-free signal of the machine tool under the target working condition.

[0018] In a third aspect, an embodiment of the present application further provides a computer device, including a memory and a processor, the memory is connected to the processor, and the processor is configured to execute one or more computer programs stored in the memory. When the processor executes the one or more computer programs, the computer device implements the method as described in the first aspect.

[0019] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, where the computer-readable storage medium stores a computer program, the computer program includes program instructions, and when the program instructions are executed by a processor, the processor executes the method as described in the first aspect.

[0020] The embodiments of the present application can achieve the following technical effects:

[0021] When the method proposed in the embodiment of the present application is used for machining fault detection, after obtaining the target signal of the machine tool under the target working condition, multiple samplings are performed through a preset sliding window, so as to determine multiple detection signals with a total length greater than the length of the original target signal. By comparing the eigenvalue of each detection signal with the threshold boundary, it is finally determined whether there is a machining fault. Compared with the related technology, the method proposed in the embodiment of the present application, after determining the target signal to be detected, performs multiple samplings through a sliding window, so as to extract multiple detection signals for eigenvalue comparison from a limited data duration, achieving the effect of maximizing the utilization of the data duration and effectively avoiding the influence on the detection result caused by insufficient effective data in the related technology. Description of the Drawings

[0022] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for the description of the embodiments of the present application will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0023] Figure 1 It is a schematic structural diagram of a machining fault detection system provided by an embodiment of the present application;

[0024] Figure 2 It is a schematic flow diagram of a machining fault detection method provided by an embodiment of the present application;

[0025] Figure 3 It is a schematic structural diagram of a machining fault detection device provided by an embodiment of the present application;

[0026] Figure 4 It is a schematic structural diagram of a computer device provided by an embodiment of the present application. Detailed Embodiments

[0027] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the drawings and embodiments. It should be understood that the specific embodiments described here are only used to explain the present application and are not used to limit the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present application.

[0028] It should be noted that if there is no conflict, the various features in the embodiments of the present application can be combined with each other, and all are within the protection scope of the present application. In addition, although functional modules are divided in the device schematic diagram and the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order from the module division in the device or the flowchart. Furthermore, the terms "first", "second", "third", etc. used in the present application do not limit the data and execution order, but only distinguish the same items or similar items with basically the same functions and roles.

[0029] To better explain the present application, a machining fault detection system is first proposed.

[0030] As Figure 1 shown, the machining fault detection system 100 includes a plurality of numerical control machine tools 11 and a plurality of host computers 12. Among them, the numerical control machine tool 11 is used to machine workpieces, for example, various machining methods such as milling, drilling, reaming, boring, tapping or turning are implemented on various workpieces.

[0031] In some embodiments, the numerical control machine tool 11 includes any suitable type of machine tool, such as a vertical numerical control machine tool or a horizontal numerical control machine tool.

[0032] The host computer 12 is used to collect various machining signals of the numerical control machine tool 11 during workpiece machining. For example, the host computer 12 can collect the machining signals of the numerical control machine tool 11 during workpiece machining (such as turning, milling, boring), including the physical state, mechanical effects, cutting parameters, and lubrication and cooling conditions of the numerical control machine tool 11 during operation. For another example, the host computer 12 can also collect specific signals generated by the numerical control machine tool 11 when machining workpieces under specific working conditions, such as machining signals of the numerical control machine tool 11 under specific conditions and environments.

[0033] It can be understood that those skilled in the art can configure the hardware architecture and software design of the host computer 12 according to the machining signals required by specific business needs. The specific composition of the host computer 12 is not limited here. For example, if the configured process data includes power signals, the host computer 12 should be configured with current sensors and voltage sensors; if the configured process data includes three-axis acceleration signals, the host computer 12 should be configured with three-axis acceleration sensors, and the three-axis acceleration sensors can be installed in the machining area.

[0034] It can be understood that the numerical control machine tool 11 or the host computer 12 involved in this embodiment can be configured into any suitable architecture.

[0035] Based on the above Figure 1 shown machining fault detection system, a machining fault detection method is proposed, which can be applied to Figure 1 the host computer shown. Specifically, please refer to Figure 2 ,Figure 2 The following is a schematic flow chart of a machining fault detection method, including:

[0036] Step S10: Obtain the target signal of the machine tool under the target working condition;

[0037] Those skilled in the art can understand that the working condition refers to the set of various conditions and environments in which the machine tool is in the actual machining process, and is used to indicate the physical state, mechanical effects, cutting parameters, and lubrication and cooling conditions of the machine tool during operation. Specifically, the operating state of the machine tool is affected by factors such as cutting force, cutting speed, and feed speed. These factors further cause a number of tools provided on the machine tool to work according to a predetermined strategy, generating vibrations, deformations, and impact forces, etc., and ultimately determining the machining result of the machine tool.

[0038] In this embodiment, the target working condition refers to a specific working condition pre-determined during the machining process of the machine tool, and this specific working condition is used to constrain the machine tool to work under specific conditions and environments to complete the expected workpiece machining task. Correspondingly, the target signal refers to the machining signal when the machine tool is in its target working condition, and is used to indicate the actual working state of the lathe.

[0039] In some embodiments, after determining the target working condition, sensors are installed on the machine tool to collect the target signal under the target working condition. The sensors installed on the machine tool transmit the collected machining signals to the host computer in the form of wireless or wired communication according to the signal acquisition time indicated by the target working condition, and the host computer stores the machining signals classified according to different working conditions. Among them, the sensors can include acceleration sensors (such as piezoelectric acceleration sensors) for collecting the three-axis acceleration during machine tool machining, or can also include power sensors for collecting power parameters during machine tool machining (such as voltage, current, or actual power when the machine tool is working).

[0040] In other embodiments, before determining the target working condition, all machining signals generated during the entire machining process of the machine tool are collected through the cooperation of multiple sensors. These machining signals are all transmitted to the host computer, so that the host computer classifies them according to the time corresponding to the signals, and extracts the machining signals corresponding to the target working condition from all the machining signals as the target signal. For example, data processing software such as MATLAB can be used to analyze the time series corresponding to the machining signals, so as to extract the machining signals corresponding to different working conditions.

[0041] Step S20: Perform multiple samplings on the target signal through a preset sliding window to determine multiple detection signals, and the total length of the multiple detection signals is greater than the length of the target signal;

[0042] It is easy to understand that the sliding window is used to extract continuous signal streams. By reasonably setting the parameters and sliding rules of the sliding window, the subsequence data generated by it can meet the expectations and achieve the effect of specific data extraction.

[0043] In this embodiment, the preset sliding window essentially indicates a data extraction method in which, before implementation, the implementer sets reasonable parameters and sliding rules according to the characteristics of the target signal. Since the effective duration of the target signal is short, the effect of directly performing feature extraction or fault detection on it is relatively poor and cannot accurately reflect the true working state of the machine tool. Therefore, in this embodiment, multiple samplings are performed through the sliding window. Since the total length of multiple detection signals is greater than the length of the target signal, that is, this embodiment is equivalent to multiplexing the target signal multiple times, extending the length of each detection signal used for detection, and effectively improving the utilization rate of the target signal under the condition of limited data volume, achieving the purpose of improving the detection effect.

[0044] Step S30, compare the eigenvalue of each said detection signal with a preset threshold boundary to determine whether there is a machining fault in the machine tool, and the threshold boundary is determined according to the fault-free signal of the machine tool under the target working condition.

[0045] It should be noted that the eigenvalue of the detection signal refers to a numerical value or parameter used to characterize the nature and characteristics of the detection signal, representing the key information of the detection signal when used for machining fault detection. In this embodiment, the specific type of the eigenvalue is not limited. Since the actual application scenario and the implementer's definition of the target working condition are different, the eigenvalue can indicate the different natures of the detection signal from multiple dimensions. However, after the target working condition is determined, the eigenvalue should be able to reflect the information most relevant to the machining fault under this target working condition.

[0046] By way of example and not limitation, this embodiment gives the following common eigenvalue forms: time-domain eigenvalues, including the mean, variance, peak value, pulse width, etc. of the signal, used to describe the changes and waveform characteristics of the signal in time; frequency-domain eigenvalues, including the spectral distribution, frequency components, spectral peak values, etc. of the signal, used to describe the characteristics and frequency composition of the signal in frequency; statistical eigenvalues, including the autocorrelation function, cross-correlation function, power spectral density, etc. of the signal, used to describe the statistical properties and correlations of the signal; energy eigenvalues, including the energy distribution, energy density, energy spectrum, etc. of the signal, used to describe the distribution of the energy of the signal at different frequencies or time periods; peak eigenvalues, including the peak amplitude, the time or frequency at which the peak appears, etc. of the signal, used to describe the position and amplitude of the extreme points or peaks of the signal.

[0047] For different eigenvalue forms, different threshold boundaries can be set accordingly. In most cases, it is considered that a machine tool fault occurs when the eigenvalue is less than the threshold boundary, and there is no machine tool fault when the eigenvalue is greater than the threshold boundary.

[0048] Further, in the above embodiment, obtaining the target signal of the machine tool under the target working condition includes: obtaining the processing signal of the machine tool; determining the start processing time and the end processing time of the target working condition; determining the effective processing time according to a preset time interval between the start processing time and the end processing time; identifying the machine tool processing signal within the effective processing time, and determining the target time period where the signal amplitude is within the preset amplitude range; extracting the processing signal within the target time period as the target signal.

[0049] Among them, the processing signal of the machine tool refers to all processing signals of the machine tool under all possible working conditions, including the target signal under the set target working condition. The start processing time and the end processing time of the target working condition refer to the time period corresponding to the target working condition. The preset time interval is a fixed duration used to determine the effective processing time of the machine tool during actual processing near the time points corresponding to the start processing time and the end processing time. The preset amplitude range refers to a fixed waveform amplitude range.

[0050] It is easy to understand that in the actual machine tool processing scenario, the start processing time and the end processing time of the target working condition are not actually equal to the effective processing time of the machine tool for processing the workpiece. The processing signals generated during the time when the tool enters and exits and when the tool contacts the workpiece but has not started processing should not be regarded as the effective processing time for processing the workpiece. Therefore, the detection of machining faults should avoid the time period when the tool enters and exits.

[0051] In this embodiment, by setting a time interval, the time periods irrelevant to workpiece processing are excluded between the start processing time and the end processing time. For example, assume that the start processing time of the machine tool is t1, the end processing time is t2, and the preset time interval is ts. Then, it can be determined that t1 + ts is the start of the effective processing time, and t2 - ts is the end of the effective processing time, that is, the effective processing time is obtained.

[0052] In some embodiments, the machine tool processing signal within the effective processing time is collected by setting the start acquisition time and the end sampling time. Assume that the start sampling time is ta seconds after the start processing time, and the end sampling time is tb seconds after the start processing time, where ta < tb ≤ ta - tb to ensure that the tool is completely processing within the material.

[0053] Further, in the above embodiment, the step of performing data sampling on the target signal through a preset sliding window to determine a plurality of detection signals includes: determining the sampling length and sampling interval of the sliding window; extracting detection signals from the target signal in sequence according to the sampling length and sampling interval of the sliding window, and the length of each detection signal is the sampling length of the sliding window. Among them, the step of extracting detection signals from the target signal in sequence according to the sampling length and sampling interval includes: sequentially determining a plurality of start times and a plurality of end times, where each start time corresponds to an end time, the interval between adjacent start times is the sampling interval, and the interval between each start time and the corresponding end time is the sampling length; extracting detection signals from the target signal in sequence according to each start time and the corresponding end time.

[0054] As a feasible implementation, a queue can be defined to store the obtained target signal. When the length of the target signal is greater than a predetermined time (i.e., the above sampling length, assumed to be 1 second), the eigenvalue is calculated and compared with a preset threshold boundary. Then, the target signal for x milliseconds is dequeued, and a new target signal for x milliseconds is enqueued, repeating this process to form a sliding window with a sampling length of 1 second and a sampling interval of x milliseconds. Let the current time be T, the window to the left is T - 1 second, and to the right is T. And the eigenvalue is calculated and compared with the preset threshold boundary every x milliseconds. If the eigenvalue is less than the threshold boundary, it is considered that a fault has occurred; if it is greater than the threshold boundary, it is considered that there is no fault.

[0055] Further, in the above embodiment, comparing the eigenvalue of each detection signal with a preset threshold boundary to determine whether there is a machine tool processing fault includes: when each detection signal is determined, inputting the detection signal into a first model, where the first model is used to extract the eigenvalue of the detection signal and compare it with the threshold boundary; determining whether the eigenvalue is less than the threshold boundary according to the output of the first model; if the eigenvalue is less than the threshold boundary, it is determined that there is a machine tool processing fault; if the eigenvalue is greater than the threshold boundary, it is determined that there is no machine tool processing fault.

[0056] Among them, the first model refers to a large model for signal recognition and feature comparison. As an example, its specific form can be a large model based on deep learning, including convolutional neural network (CNN), recurrent neural network (RNN), and support vector machine (SVM), etc. The convolutional neural network extracts local features of the input detection signal, and combines the pooling layer and the fully connected layer for feature extraction and classification. The recurrent neural network is mainly used to capture the dependence relationship between signals and time series and processes time series data in specific signal recognition. The support vector machine, as a traditional machine learning model, is used to perform binary classification on the detection signal after analysis and processing to determine whether the eigenvalue is greater than the threshold boundary and output the judgment result.

[0057] In some embodiments, it is also necessary to train the initial model before using the first model, including: obtaining the historical processing signals of the machine tool; creating a data set according to the historical processing signals; and training a preset initial model through the data set to obtain the first model.

[0058] Specifically, the step of training the preset initial model according to the data set includes: training the initial model according to the training set to obtain a target model; evaluating the target model through the test set and a preset loss function to determine the accuracy rate of the target model; if the accuracy rate is lower than the preset accuracy rate threshold, updating the hyperparameters of the target model to perform multiple iterative trainings on the target model until the accuracy rate of the target model is higher than the accuracy rate threshold to obtain the first model.

[0059] Among them, the training set includes a large amount of already labeled data, and each group of data contains corresponding labels or output values. The initial model learns the patterns and correlations of each signal in the data in a supervised manner, so as to make reasonable inferences when given an input. Correspondingly, the sample set contains a large amount of unlabeled data for evaluating the performance of the model. In this embodiment, the loss function is used to measure the difference or error between the output result of the model and the actual observed value. As a feasible implementation manner, the Hinge function can be used in this embodiment to evaluate the output of the SVM classification model in the large model, and it evaluates the performance of the large model by measuring the margin example between the output result and the true label.

[0060] Further, in the above embodiment, before comparing the eigenvalue of each detection signal with the preset threshold boundary, it is also necessary to determine the threshold boundary, specifically including: obtaining the fault-free signal of the machine tool under the target working condition, and the fault-free signal includes multiple reference signal segments; performing data sampling operations on each reference signal segment according to the preset sampling rate to obtain a calibrated signal segment, and the number of calibrated signal segments is equal to the number of target data segments; calculating the eigenvalue of each calibrated signal segment to determine the threshold boundary.

[0061] Among them, the step of calculating the eigenvalue of each calibrated data segment includes: determining the waveform of each calibrated data segment; performing waveform extraction in the waveform of each calibrated data segment according to the preset waveform sampling period and waveform sampling length to determine the waveform subset of the calibrated data segment; calculating the eigenvalue of each calibrated data segment according to the waveform subset of each calibrated data segment; and determining the threshold boundary according to the eigenvalue.

[0062] It should be noted that the fault-free signal refers to the machining signal when the machine tool does not have machining faults under the target working conditions. The reference signal segment refers to a segment of the fault-free signal. In order to maximize the utilization of data and ensure the accuracy of the threshold boundary, in this embodiment, based on a preset sampling rate, n segments of signals (n is the number of acquisition segments, preferably, n > 7) are sampled, so as to obtain n calibrated signal segments. All n calibrated signal segments are used to calculate the threshold boundary, thereby ensuring the reliability of the threshold boundary.

[0063] In summary, after obtaining the target signal of the machine tool under the target working conditions, this application performs multiple samplings through a preset sliding window, thereby determining multiple detection signals with a total length greater than the length of the original target signal. By comparing the characteristic values of each detection signal with the threshold boundary, it is finally determined whether there is a machining fault. Compared with the related art, the method proposed in the embodiment of this application, after determining the target signal to be detected, performs multiple samplings through a sliding window, thereby extracting multiple detection signals for eigenvalue comparison from a limited data duration, achieving the effect of maximizing the utilization of the data duration, and effectively avoiding the influence on the detection result caused by insufficient effective data in the related art.

[0064] As another aspect of the embodiment of this application, the embodiment of this application provides a machining fault detection device. Among them, the machining fault detection device can be a software module. The software module includes several instructions, which are stored in the memory. The processor can access this memory and call the instructions for execution to complete the machining fault detection method described in each of the above embodiments.

[0065] In some embodiments, the machining fault detection device can also be built by hardware devices. For example, the machining fault detection device can be built by one or more than two chips. Each chip can work in coordination with each other to complete the machining fault detection method described in each of the above embodiments. For another example, the machining fault detection device can also be built by various logic devices, such as being built by a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), a single-chip microcomputer, an ARM (Acorn RISC Machine), or other programmable logic devices, discrete gate or transistor logic, discrete hardware components, or any combination of these components.

[0066] Specifically, please refer to Figure 3 , Figure 3 As shown in the structural schematic diagram of a machining fault detection device. As shown in the figure, the device includes:

[0067] A data acquisition module 310, configured to acquire a target signal of the machine tool under the target working conditions;

[0068] A data sampling module 320, configured to perform multiple samplings on the target signal through a preset sliding window to determine a plurality of detection signals, wherein the total length of the plurality of detection signals is greater than the length of the target signal;

[0069] A data comparison module 330, configured to compare the eigenvalue of each detection signal with a preset threshold boundary to determine whether there is a machine tool processing fault, and the threshold boundary is determined according to the fault-free signal of the machine tool under the target working condition.

[0070] As a feasible implementation manner, when the data acquisition module 310 is configured to acquire the target signal of the machine tool under the target working condition, it is specifically configured to: acquire the processing signal of the machine tool; determine the start processing time and the end processing time of the target working condition; determine the effective processing time between the start processing time and the end processing time according to a preset time interval; identify the machine tool processing signal within the effective processing time, and determine the target time period with the signal amplitude within a preset amplitude range; extract the processing signal within the target time period as the target signal.

[0071] As a feasible implementation manner, when the data sampling module 320 is configured to perform data sampling operations on the target signal through a preset sliding window to determine a plurality of detection signals, it is specifically configured to: determine the sampling length and the sampling interval of the sliding window; extract the detection signals in the target signal in sequence according to the sampling length and the sampling interval of the sliding window, and the length of each detection signal is the sampling length of the sliding window.

[0072] As a feasible implementation manner, when the data sampling module 320 is configured to extract the detection signals in the target signal in sequence according to the sampling length and the sampling interval, it is specifically configured to: sequentially determine a plurality of start times and a plurality of end times, wherein each start time corresponds to an end time, the interval between adjacent start times is the sampling interval, and the interval between each start time and the corresponding end time is the sampling length; extract the detection signals in the target signal in sequence according to each start time and the corresponding end time.

[0073] As a feasible implementation manner, when the data comparison module 320 is used to compare the eigenvalue of each of the detection signals with a preset threshold boundary to determine whether there is a machining fault of the machine tool, it is specifically used for: when determining a detection signal each time, inputting the detection signal into a first model, where the first model is used to extract the eigenvalue of the detection signal and compare it with the threshold boundary; determining whether the eigenvalue is less than the threshold boundary according to the output of the first model; if the eigenvalue is less than the threshold boundary, determining that there is a machining fault of the machine tool; if the eigenvalue is greater than the threshold boundary, determining that there is no machining fault of the machine tool.

[0074] As a feasible implementation manner, when the data comparison module 320 is used to input the detection signal into the first model each time when determining a detection signal, it is further used for: acquiring the historical machining signals of the machine tool; creating a data set according to the historical machining signals; training a preset initial model through the data set to obtain the first model.

[0075] As a feasible implementation manner, when the data comparison module 320 is used to train a preset initial model through the data set to obtain the first model, it is specifically used for: training the initial model according to the training set to obtain a target model; evaluating the target model through the test set and a preset loss function to determine the accuracy rate of the target model; if the accuracy rate is lower than a preset accuracy rate threshold, updating the hyperparameters of the target model to perform multiple iterative trainings on the target model until the accuracy rate of the target model is higher than the accuracy rate threshold to obtain the first model.

[0076] As a feasible implementation manner, when the data comparison module 330 is used to compare the eigenvalue of each of the detection signals with a preset threshold boundary to determine whether there is a machining fault of the machine tool, it is further used for: acquiring a fault-free signal of the machine tool under the target working condition, where the fault-free signal includes a plurality of reference signal segments; performing data sampling operations on each of the reference signal segments according to a preset sampling rate to obtain calibrated signal segments, and the number of the calibrated signal segments is equal to the number of the target data segments; calculating the eigenvalue of each of the calibrated signal segments to determine the threshold boundary.

[0077] As a feasible implementation manner, when the data comparison module 330 is used to calculate the eigenvalue of each of the calibration data segments to determine the threshold boundary, it is specifically configured to: determine the waveform of each of the calibration data segments; perform waveform extraction in the waveform of each of the calibration data segments according to a preset waveform sampling period and waveform sampling length to determine the waveform subset of the calibration data segment; calculate the eigenvalue of each of the calibration data segments according to the waveform subset of each of the calibration data segments; and determine the threshold boundary according to the eigenvalue.

[0078] It should be noted that the above machining fault detection device can execute the machining fault detection method provided in the embodiments of the present application, and has the corresponding functional modules and beneficial effects for executing the method. For the technical details not described in detail in the embodiments of the machining fault detection device, reference can be made to the machining fault detection method provided in the embodiments of the present application.

[0079] See Figure 4 , Figure 4 is a schematic structural diagram of a computer device provided in an embodiment of the present application. The computer device 40 includes one or more processors 41 and a memory 42. The memory 42 is connected to one or more processors 41, for example, connected to the processor 41 through a bus.

[0080] The processor 41 is configured to support the computer device to execute the corresponding functions in the method in the above method embodiments. The processor 41 may be a central processing unit (CPU), a network processor (NP), a hardware chip, or any combination thereof. The above hardware chip may be an application specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The above PLD may be a complex programmable logic device (CPLD), a field-programmable gate array (FPGA), a generic array logic (GAL), or any combination thereof.

[0081] The memory 42 is used to store program codes and the like. The memory 42 may include a volatile memory (VM), such as a random access memory (RAM); the memory 42 may also include a non-volatile memory (NVM), such as a read-only memory (ROM), a flash memory, a hard disk drive (HDD), or a solid-state drive (SSD); the memory 42 may further include a combination of the above types of memories.

[0082] The memory 42 can be used to store non-volatile software programs, non-volatile computer-executable programs, and modules, such as the program instructions / modules corresponding to the machining fault detection method in the embodiments of the present application. The processor 41 executes various functional applications and data processing of the machining fault detection method and the machining fault detection device by running the non-volatile software programs, instructions, and modules stored in the memory 42, that is, realizes the functions of each module or unit of the machining fault detection method and the machining fault detection device provided in the above method embodiments.

[0083] The memory 42 may include a program storage area and a data storage area. Among them, the program storage area can store an operating system and application programs required for at least one function. The data storage area can store data created according to the use of the machining fault detection device. In some embodiments, the memory 42 may optionally include a memory remotely set relative to the processor 41, and these remote memories can be connected to the machining fault detection device through a network. Examples of the above network include but are not limited to the Internet, an enterprise intranet, a local area network, a mobile communication network, and combinations thereof.

[0084] The one or more modules are stored in the memory 42 and, when executed by the one or more processors 41, execute the machining fault detection method in any of the above method embodiments. For example, execute the method steps described in the above method embodiments and realize the functions of the modules described in the above device embodiments.

[0085] The embodiments of the present application further provide a computer-readable storage medium. The computer-readable storage medium stores a computer program, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer is caused to execute the method as described in the foregoing embodiments.

[0086] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above various methods. Among them, the storage medium can be a magnetic disk, an optical disc, a read-only memory (ROM), or a random access memory (RAM), etc.

[0087] The above-disclosed are only the preferred embodiments of the present application. Of course, the scope of rights of the present application cannot be limited thereby. Therefore, equivalent changes made according to the claims of the present application still fall within the scope covered by the present application.

Claims

1. A machining fault detection method, characterized in that, Including: Obtain the target signal of the machine tool under the target working condition; Perform multiple samplings on the target signal through a preset sliding window to determine a plurality of detection signals, and the total length of the plurality of detection signals is greater than the length of the target signal; Compare the eigenvalue of each detection signal with a preset threshold boundary to determine whether there is a machining fault of the machine tool, and the threshold boundary is determined according to the fault-free signal of the machine tool under the target working condition.

2. The method according to claim 1, wherein The obtaining the target signal of the machine tool under the target working condition includes: Obtain the machining signal of the machine tool; Determine the start machining time and the end machining time of the target working condition; Determine the effective machining time between the start machining time and the end machining time according to a preset time interval; Identify the machining signal of the machine tool within the effective machining time, and determine the target time period with the signal amplitude within a preset amplitude range; Extract the machining signal within the target time period as the target signal.

3. The method according to claim 1, wherein The performing data sampling operation on the target signal through a preset sliding window to determine a plurality of detection signals includes: Determine the sampling length and the sampling interval of the sliding window; Extract the detection signals from the target signal in sequence according to the sampling length and the sampling interval of the sliding window, and the length of each detection signal is the sampling length of the sliding window.

4. The method according to claim 3, characterized in that, The extracting the detection signals from the target signal in sequence according to the sampling length and the sampling interval includes: Sequentially determine a plurality of start times and a plurality of end times, wherein each start time corresponds to an end time, the interval between adjacent start times is the sampling interval, and the interval between each start time and the corresponding end time is the sampling length; Extract the detection signals from the target signal in sequence according to each start time and the corresponding end time.

5. The method according to claim 1, wherein The comparing the eigenvalue of each detection signal with a preset threshold boundary to determine whether there is a machining fault of the machine tool includes: When each detection signal is determined, input the detection signal into a first model, and the first model is used to extract the eigenvalue of the detection signal and compare it with the threshold boundary; Determine whether the eigenvalue is less than the threshold boundary according to the output of the first model; If the eigenvalue is less than the threshold boundary, it is determined that there is a machining fault of the machine tool; If the eigenvalue is greater than the threshold boundary, it is determined that there is no machining fault of the machine tool.

6. The method according to claim 5, wherein Before inputting the detection signal into the first model for each determination of the detection signal, and the first model is used to extract the eigenvalue of the detection signal and compare it with the threshold boundary, it further includes: Obtain the historical machining signal of the machine tool; Create a data set according to the historical machining signal; Train a preset initial model through the data set to obtain the first model.

7. The method according to claim 6, wherein The data set includes a training set and a test set. The training the preset initial model through the data set to obtain the first model includes: Train the initial model according to the training set to obtain a target model; Evaluate the target model through the test set and a preset loss function to determine the accuracy of the target model; If the accuracy is lower than a preset accuracy threshold, update the hyperparameters of the target model to perform multiple iterative trainings on the target model until the accuracy of the target model is higher than the accuracy threshold to obtain the first model.

8. The method according to claim 1, wherein Before comparing the eigenvalue of each detection signal with a preset threshold boundary to determine whether there is a machine tool processing fault, it further includes: Obtain a fault-free signal of the machine tool under the target working condition, where the fault-free signal includes a plurality of reference signal segments; Perform data sampling operations on each reference signal segment according to a preset sampling rate to obtain calibrated signal segments, and the number of the calibrated signal segments is equal to the number of the target data segments; Calculate the eigenvalue of each calibrated signal segment to determine the threshold boundary.

9. The method according to claim 8, wherein The calculating the eigenvalue of each calibrated data segment to determine the threshold boundary includes: Determine the waveform of each calibrated data segment; Perform waveform extraction in the waveform of each calibrated data segment according to a preset waveform sampling period and waveform sampling length to determine the waveform subset of the calibrated data segment; Calculate the eigenvalue of each calibrated data segment according to the waveform subset of each calibrated data segment; Determine the threshold boundary according to the eigenvalue.

10. A computer device, characterized in that, It includes a memory and a processor, the memory is connected to the processor, and the processor is configured to execute one or more computer programs stored in the memory. When the processor executes the one or more computer programs, the computer device implements the method according to any one of claims 1-9.

Images