Device state detection method and apparatus, device, and storage medium

Abstract

The application discloses a kind of equipment state detection method, device, equipment and storage medium.The method comprises: the original noise of target equipment is collected, and the processing period corresponding to the original noise is determined according to trigger signal;Determine the noise characteristic signal of the original noise in each processing period;According to the comparison result of each noise characteristic signal and preset processing object template signal, the equipment state is determined.The embodiment of the application, by comparing the comparison result of each noise characteristic signal and preset processing template signal, the accurate detection of machine tool type equipment state is realized, the reliability of equipment production is improved, and the production quality of product is enhanced.

Description

Technical Field

[0001] The present invention relates to the field of automation control technology, and in particular to a method, apparatus, device and storage medium for detecting equipment status. Background Technology

[0002] Machine tools are essential equipment in industrial production lines. The processing quality of machine tools directly determines the quality of the products. Monitoring the status of machine tools and detecting anomalies are technologies urgently needed in industry. They play a crucial role in improving production efficiency, enhancing product quality, and achieving automated management of production lines.

[0003] Currently, monitoring machine tool equipment faces the following challenges: First, machine tools are diverse, and there is a lack of universal monitoring methods. Existing machine tool monitoring methods can only achieve "one model per machine," requiring secondary development for specific equipment and scenarios. The effectiveness of this secondary development directly affects the accuracy of machine tool status monitoring and also requires significant manpower and resources. Second, because the processing objects of machine tools are usually not unique, machine tool parameters frequently change, making it impossible to accurately detect machine tool status. This leads to uncontrollable product processing quality and reduces the overall production quality of machine tools. Therefore, there is an urgent need for a method for monitoring the status of machine tools. Summary of the Invention

[0004] This invention provides a method, apparatus, electronic device, and storage medium for monitoring equipment status, so as to achieve accurate detection of the status of machine tool equipment.

[0005] In a first aspect, embodiments of the present invention provide a device status monitoring method, comprising:

[0006] The raw noise of the target device is collected, and the processing cycle corresponding to the raw noise is determined based on the trigger signal;

[0007] Determine the noise characteristic signal of the original noise within each processing cycle;

[0008] The equipment status is determined based on the comparison results between each noise characteristic signal and the preset processing object template signal.

[0009] Secondly, embodiments of the present invention also provide a device for monitoring device status, the device comprising:

[0010] The cycle determination module is used to collect the original noise of the target device and determine the processing cycle corresponding to the original noise based on the trigger signal.

[0011] The feature signal module is used to determine the noise feature signal of the original noise in each processing cycle;

[0012] The status detection module is used to determine the equipment status based on the comparison results between each noise characteristic signal and the preset processing object template signal.

[0013] Thirdly, embodiments of the present invention also provide a computer device, comprising:

[0014] One or more processors;

[0015] Storage device for storing one or more programs;

[0016] When the one or more programs are executed by one or more processors, the one or more processors implement the device state detection method as described in the first aspect above.

[0017] Fourthly, embodiments of the present invention also provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the device detection method as described in the first aspect above.

[0018] In the technical solution provided by this invention, the original noise of the target equipment is first collected, and then the processing cycle corresponding to the original noise is determined according to the trigger signal. Then, the noise characteristic signals of the original noise are determined within each processing cycle. The equipment status is determined based on the comparison results between each noise characteristic signal and a preset processing template signal. By comparing each noise characteristic signal with the preset processing template signal, accurate detection of the machine tool equipment status is achieved, improving the reliability of equipment production and enhancing product quality. Attached Figure Description

[0019] Figure 1 This is a flowchart of a device testing method provided in Embodiment 1 of the present invention;

[0020] Figure 2 This is a flowchart of a device testing method provided in Embodiment 2 of the present invention;

[0021] Figure 2a This is a schematic diagram of a trigger signal template provided in Embodiment 2 of the present invention;

[0022] Figure 2b This is a schematic diagram of the initial positioning effect of a production area provided in Embodiment 2 of the present invention;

[0023] Figure 2c This is a schematic diagram of template signals for different processing objects provided in Embodiment 2 of the present invention;

[0024] Figure 2d This is a schematic diagram illustrating an example of a finely categorized anchor point and credential interval provided in Embodiment 2 of the present invention;

[0025] Figure 2eThis is a schematic diagram illustrating the effect of a device testing method provided in Embodiment 2 of the present invention;

[0026] Figure 2f This is a schematic diagram illustrating the effect of a device testing method provided in Embodiment 2 of the present invention;

[0027] Figure 2g This is a schematic diagram illustrating the effect of real-time monitoring of processing progress and prediction of the next data point provided in Embodiment 2 of the present invention;

[0028] Figure 3 This is a schematic diagram of the structure of a device testing apparatus provided in Embodiment 3 of the present invention;

[0029] Figure 4 This is a schematic diagram of the structure of a computer device provided in Embodiment 4 of the present invention. Detailed Implementation

[0030] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and not intended to limit it. Furthermore, it should be noted that, for ease of description, only the parts relevant to the invention are shown in the drawings, not all structures. Moreover, unless otherwise specified, the embodiments and features described herein can be combined with each other.

[0031] Example 1

[0032] Figure 1 This is a flowchart of an equipment detection method provided in Embodiment 1 of the present invention. This embodiment can be applied to the situation of monitoring the status of machine tool equipment in a production environment. The method can be executed by an equipment detection device, which can be implemented in hardware and / or software and is usually configured in a computer device.

[0033] The method specifically includes the following steps:

[0034] S101. Collect the original noise of the target device and determine the processing cycle corresponding to the original noise based on the trigger signal.

[0035] The target device can be any device that needs to be monitored. The trigger signal can be a signal reflecting the periodic working pattern of the target device, indicating its working status. This trigger information can be a manually issued signal, such as a start-up or shutdown signal for the target device, or a signal indicating that the target device is performing an operation, such as a preparation signal for processing or a cleaning signal.

[0036] For example, when the target device is in the start-up state, the machine tool will have a cleaning process after each processing, which is represented by two high pulses in the characteristic signal. This signal has high recognition and appears stably. The two high pulse signals that appear in the cleaning process can be used as the trigger signal in the implementation of the present invention.

[0037] In this embodiment, the processing cycle can be the cycle in which the target equipment produces one product. Since the products produced within the same production cycle are different, the noise level of the target equipment will also vary. For example, machine tool equipment can produce multiple products, and the production time of each product can be considered as a processing cycle.

[0038] S102. Determine the noise characteristic signal of the original noise in each processing cycle.

[0039] Among them, the noise characteristic signal can be a characteristic signal extracted from the original noise. The extracted characteristic signal can make the noise characteristics generated when the target equipment is in operation clearer.

[0040] In this embodiment, the method for extracting noise feature signals may include extracting specific frequency band components of the acoustic signal from the original noise and calculating their mean or standard deviation. The calculation result can be used as the noise feature signal. The method for extracting noise feature signals is not limited to this; for example, it may also include using a neural network model to extract noise feature signals from the original noise.

[0041] S103. Determine the equipment status based on the comparison results between each noise characteristic signal and the preset processing object template signal.

[0042] The object being processed can be a part of the target equipment. For example, it could be a coupling of the target equipment, or a gear of the target equipment, etc., which is not limited in this embodiment. The preset processing object template signal refers to various sound signals during the production of the processing object. For example, the preset processing object template signal can be the sound signal reflecting the equipment's production of the processing object. The preset processing object template signal can be all the sound signals within a complete processing cycle of the equipment, or it can be the sound signals of the equipment processing within one or more specific frequency bands.

[0043] In this embodiment, the comparison method between each noise feature signal and the preset processing object template signal can be a comparison of the intersection-exchange ratio (IoE) of each noise feature signal and the preset processing object template signal; it can also be a comparison of the frequency values ​​obtained by performing Fourier transform on each noise feature signal and the preset processing object template signal in the time domain system to obtain frequency values, and then comparing the obtained low-frequency and high-frequency values; or it can be a comparison of the obtained noise feature signals and the preset processing object template signal to obtain graphic signals, and then comparing them based on the degree of matching correlation of the obtained graphic signals.

[0044] In the technical solution provided by this embodiment of the invention, the original noise of the target equipment is collected, and the processing cycle corresponding to the original noise is determined according to the trigger signal. Then, the noise characteristic signal of the original noise is determined within each processing cycle. Finally, the equipment status is determined based on the comparison result of each noise characteristic signal and the preset processing object template signal. This embodiment of the technical solution accurately detects the status of the machine tool equipment by comparing the noise characteristic signal with the preset processing object template signal, promptly identifies faults, improves the reliability of equipment production, and enhances the production quality of the product.

[0045] Example 2

[0046] Figure 2 This is a flowchart of a device testing method provided in Embodiment 2 of the present invention. This embodiment further refines the methods described in the above embodiments. Specifically, it may include the following steps:

[0047] S201. Collect the original noise at the machine tool spindle of the target equipment.

[0048] Here, the raw noise refers to the noise collected by a microphone placed near the target device during the operation of its machine tool spindle. The collected raw noise can be bearing noise, impact noise, or friction noise, and this embodiment of the invention does not impose any limitations on it.

[0049] Specifically, the S201 can collect the raw noise of the target device when it is in the startup state.

[0050] S202. Divide the original noise into at least one processing cycle according to the trigger signal, wherein the trigger signal includes the equipment flushing pulse signal.

[0051] The equipment flushing pulse signal is a discrete signal with a certain periodicity. It can represent information, serve as a carrier wave, or act as a clock signal for various digital circuits and high-performance chips. The time interval between equipment flushing pulse signals is called the period. For example, when the target equipment is in the startup state, the machine tool performs a cleaning process after each processing cycle, which is represented by two high-frequency pulses in the characteristic signal. These two high-frequency pulse signals generated during the cleaning process can be called the equipment flushing pulse signal.

[0052] In this embodiment, the acquired raw noise can be divided into one or more processing cycles based on the trigger signal. For example, the acquired raw noise can be divided into one or more processing cycles based on a manually issued signal, i.e., a start signal or a stop signal of the target device; or it can be divided into one or more processing cycles based on a signal indicating that the target device is performing an operation, i.e., a preparation signal or a cleaning signal of the target device.

[0053] It should be noted that when detecting the trigger signal, the analysis of the previously processed sound signal needs to begin after each processed sound signal is identified. Therefore, a signal that appears in each cycle of the target device needs to be set as the trigger signal for identification. By using the target detection rule based on the cross-union ratio, each trigger signal can be captured, and subsequent steps can be executed after the trigger signal is detected.

[0054] S203. Use preset processing operations to process the original noise in each processing cycle to obtain noise characteristic signals.

[0055] Specifically, the noise feature signal is obtained by: processing the original noise within one or more processing cycles using preprocessing operations, then extracting the original noise feature values, and finally obtaining the noise feature signal.

[0056] Optionally, the preset processing operation includes at least one of the following: the mean of the high-frequency component, the standard deviation of the high-frequency component, the mean of the low-frequency component, and the standard deviation of both high and low-frequency components. Here, the preprocessing operation refers to an operation that performs certain processing beforehand.

[0057] In this embodiment, at least one of the following is used to process the original noise in one or more processing cycles: the mean of high-frequency components, the standard deviation of high-frequency components, the mean of low-frequency components, and the standard deviation of high and low-frequency components. First, the original noise feature values ​​are extracted, and then the noise feature signal under the original noise feature values ​​is obtained.

[0058] The mean refers to the average, a measure of central tendency in a set of data. The standard deviation is the square root of the arithmetic mean of the squared deviations of the standard values ​​of all units in a population from their mean; it reflects the dispersion among individuals within the group. A frequency band refers to a specific range of frequencies. The mean of the high-frequency component is the mean of the sample frequency distribution within the high-frequency range. The standard deviation of the high-frequency component is the standard deviation of the sample frequency distribution within the high-frequency range. The mean of the low-frequency component is the mean of the sample frequency distribution within the low-frequency range. The standard deviation of both high and low-frequency components refers to the standard deviation of the sample frequency distribution within both high and low-frequency ranges.

[0059] S204. Obtain the processing object of the target device and determine the coarse classification of the object corresponding to the processing object.

[0060] The processing object can be a part of the target equipment, such as a coupling or a gear; this embodiment does not impose any limitations. Coarse object classification refers to dividing the processing objects into broad categories. Processing objects that are similar in terms of processing object template signals (i.e., processing flow) and have only minor differences in details are grouped into the same category.

[0061] Specifically, when monitoring target equipment, the processing object currently being processed by the equipment can be obtained. This can be detected through image sensors or actively input by the user when starting the target equipment to process the object. The processing object can specifically be identification information or a product name. A coarse classification can be established based on the object to which the processing object belongs. This coarse classification can include one or more different processing objects. It is understood that machine tools whose processing objects belong to the same coarse classification will have similar noise characteristic signals.

[0062] S205. Obtain the preset processing object template signal within the object coarse classification.

[0063] Specifically, within one or more object coarse classifications, the preset processing object template signal is obtained.

[0064] It should be noted that the method for obtaining the preset processing object template signal within the object coarse classification is as follows: after the object coarse classification, the preset processing object template signals with high similarity will be in the same large category, and the preset processing object template signal can be obtained directly.

[0065] S206. Determine the crossover ratio between the noise characteristic signal and the preset processing object template signal.

[0066] The intersection-to-union ratio (IoU) is a metric used to describe the degree of overlap between the generated noise feature signal and the preset processing object template signal; that is, the ratio of their intersection to their union. The accuracy of the detection can be evaluated by calculating the IoU value; the higher the IoU value, the more accurate the measurement result.

[0067] Specifically, the cross-intersection over union (CUI) ratio is determined based on the noise characteristic signal and the preset processing object template signal to determine whether the target equipment is in normal working condition. The specific steps are as follows: based on the processing object of the target equipment, determine the corresponding object coarse category; then, obtain the preset processing object template signal within the object coarse category; determine the CUI ratio between the noise characteristic signal and the preset processing object template signal at this time; and finally, compare the CUI ratio with a set threshold.

[0068] S207. Compare the crossover ratio with the threshold. If the crossover ratio is less than the threshold, the target device is determined to be in a fault state. If the crossover ratio is greater than or equal to the threshold, the target device is determined to be in a normal state.

[0069] Specifically, based on the comparison of the cross-intersection ratio (CIP) of the noise characteristic signal and the preset processing target template signal with a threshold, if the CIP is greater than or equal to the threshold, the target device is determined to be in a fault state; if the CIP is less than the threshold, the target device is determined to be in a normal state.

[0070] For example, given the known classification of objects processed by the target equipment, a template signal for the corresponding object category can be obtained from the target detection parameters. This template signal is then compared with the noise characteristic signal within the target equipment's processing cycle to analyze the current processing progress. Based on the preset template signal, the approximate range of the noise characteristic signal for the next stage can be predicted. Comparing the predicted result with the noise characteristic signal generated by subsequent equipment reveals a significant deviation, indicating an anomaly in the target equipment's production process.

[0071] In this embodiment of the invention, raw noise collected during the device's operation is used to obtain noise feature signals based on the collected raw signals. By acquiring the processing objects of the target device, a coarse classification of the corresponding objects is determined to obtain a preset processing object template signal. Finally, the intersection-exchange ratio (IoE) of the noise feature signals and the preset processing object template signal is compared with a threshold to determine whether the device is in a normal state. This embodiment of the invention, by classifying the processing objects and, given the known classification, achieves real-time monitoring of processing progress and can capture abnormal behaviors during processing, further improving the accuracy of target device detection.

[0072] Furthermore, based on the above embodiments of the invention, the equipment status detection method further includes: acquiring noise data of at least one processing object; and extracting the noise feature signals of each noise data as a preset processing object template signal corresponding to the processing object.

[0073] It's important to know that methods for acquiring noise data of the machined object include: when the target equipment is in the startup state, noise can be generated at its machine tool spindle, couplings, etc. Extracting this noise data allows the extracted sound signals to be used as the noise data of the machined object. The noise data collected from the machined object can be one or more samples.

[0074] Specifically, noise feature signals of noise data from one or more processing objects are extracted and used as preset processing object template signals corresponding to the processing objects.

[0075] In this embodiment, the method for extracting noise feature signals from noise data includes: extracting feature values ​​of one or more noise data, wherein the extracted feature values ​​can be one of the mean of high-frequency components, the standard deviation of high-frequency components, the mean of low-frequency components, and the standard deviation of high and low-frequency components.

[0076] In this embodiment, the method of extracting the noise feature signals of each noise data as the preset processing object template signal corresponding to the processing object includes: extracting the feature values ​​of each noise data for analysis and calculating the variance and standard deviation to obtain the noise feature signals of the noise data, and directly using them as the preset processing object template signal corresponding to the processing object.

[0077] Optionally, after extracting the noise feature signals of each noise data as the preset processing object template signal corresponding to the processing object, the method further includes:

[0078] Based on the similarity of the preset processing object template signals, each preset processing object template signal is divided into at least one object coarse category; the feature intervals of each noise feature signal within the object coarse category are labeled.

[0079] It should be noted that labeling the feature intervals of each noise feature signal within the coarse classification of the object can be understood as a fine classification of the processed object. Specifically, for the time period of the performance of the subcategories within the broad classification range of the processed object, only the distinguishing local signals are extracted. These local signals are then reclassified using a target detection algorithm based on intersection-union ratio (IU). This yields the points for the fine classification of the processed object. Labeling the feature intervals reduces the amount of comparison between the noise feature signals and the preset target signals of the processed object, thereby reducing the data processing scale and improving the real-time performance of equipment status detection.

[0080] In this embodiment, objects are classified into one or more coarse categories based on the similarity of preset processing object template signals. During the coarse classification of processing objects, a target detection algorithm based on intersection-union ratio (IU) is used to broadly categorize the processing objects of the target device, selecting those with similar processing flows and only minor differences to be grouped into the same category. Since there are significant differences in processing flows between different processing object categories, the shapes of the generated noise feature signals will also differ considerably. In this case, by specifying a processing object template signal for each processing object category, the IU-based target detection algorithm can achieve coarse classification of the processing objects.

[0081] In this embodiment, when classifying the processing object template signals based on the similarity of the processing object template signals, the similarity of the processing object template signals of several processing objects within each coarse category of processing objects often only differs in minute details. These minute details account for a small proportion of the overall signal. When directly classifying the overall waveform based on the cross-parallel ratio (CPAR) target rule, the CPAR fluctuations generated by these minute details are very small and easily covered by disturbances caused by accidental factors, making it difficult to classify the minute details. Therefore, a classification method is provided for the fine classification stage, which involves labeling the feature intervals of each noise feature signal within the coarse category of the object. First, it is possible to specifically examine the time periods in which the differences between the several subcategories within the coarse category of each processing object are manifested. Only the local signals that produce the differences are extracted. These local signals are then classified using the same method as in the coarse classification of the object. The feature intervals of each noise feature signal within the coarse category of the object are then labeled, and the labeled points are the fine classification points.

[0082] For example, the annotation points that mark the feature intervals of noise feature signals within the coarse object classification can be called the anchor points of the local signals in the fine classification. The anchor points can be understood as the intervals where differences are generated after manually observing the differences between several different processed objects within the coarse object classification.

[0083] Furthermore, based on the above-mentioned invention, the equipment detection method also includes: determining the matching length between the noise characteristic signal and the preset processing object template signal within the processing cycle, and using the matching length as the processing progress of the target equipment.

[0084] The matching length can be the length of the overlapping portion between the noise feature signal and the preset processing object template signal within the processing cycle. The processing progress can be understood as the process of matching the real-time generated noise feature signal with the preset processing object template signal, and the proportion of the overlapping portion in the preset processing object template signal can be used to determine the processing progress of the target device.

[0085] In this embodiment, the matching length of the preset processing object template signal can be determined by the annotation points marked on the feature interval of the noise feature signal. The signal within a distance on both sides of the annotation point is determined by the main signal that generates the distinction.

[0086] For example, taking the monitoring of a machine tool's operating status through noise as an example, the equipment status detection method of this embodiment of the invention may include the following process: recording the noise of the machine tool's operation using a noise acquisition device installed near the machine tool's spindle, and obtaining a feature signal with a sampling rate of 1Hz after simple variance feature extraction. The sampling rate can be adjusted according to the specific application scenario. In this embodiment of the invention, there is no rapid transient process, and the duration of a single processing cycle is approximately half an hour; therefore, a sampling rate of 1Hz is sufficient.

[0087] Observation revealed that the machine tool undergoes a cleaning process after each machining operation, which manifests as two high-frequency pulses in the characteristic signal. This signal is highly recognizable and occurs stably, making it suitable as the trigger signal for the equipment detection method in this embodiment. The signal from one cleaning step is selected as a template, such as... Figure 2a As shown.

[0088] Next, the template signal can be used to scan the feature signal in real time to determine its cross-section over parallel (COP) curve with the original feature signal. When the template signal and the original feature signal overlap at a certain time point, the COP curve will show a significant peak. Therefore, by performing peak detection on the curve, the position of each trigger signal can be located. The effect is as follows... Figure 2b As shown. After this step, the machine tool's machining actions can be roughly positioned within a certain time interval before the trigger signal.

[0089] Secondly, the template cycle of 11 types of parts recently produced by the factory was determined from the characteristic signals, such as... Figure 2c As shown. Among them, from Figure 2c As can be seen, the template signals in rows 6, 7, and 8 from top to bottom are grouped into the same category due to their similar processing flow. Each time the algorithm captures a trigger signal, it sequentially scans the signal between the current trigger signal and the previous trigger signal using template signals of each category. The maximum value of the intersection-exchange ratio (IoU) curve obtained from each template signal scan is taken, and this maximum value is taken again between different templates. If this maximum value is greater than a set threshold, it is determined to be the category corresponding to that value; if it is lower than the threshold, it is determined to be an unknown part.

[0090] Then, the method for classifying the finer categories is as follows: Figure 2d As shown, since the signals within the three boxes represent the main distinguishing regions for the three types of parts, the three signals circled in black are selected as anchor points. After detecting a signal of this broad category once, the same cross-intersection-union (CUI) based detection method as before is used, but the detection target is changed to the local signals near the anchor points to determine the anchor point's location. By taking signals within a certain interval to the left or right of the anchor point, the key local signals that generate the distinction can be obtained. Performing the same template matching and threshold comparison on the local signals determines the specific sub-category corresponding to this production.

[0091] Furthermore, the effect of processing part type recognition is as follows: Figure 2e As shown, as Figure 2fAs shown, the first line represents the characteristic signal output by the sensor; the second line is the matching curve of the trigger signal, with its peak corresponding to the cleaning action, used to trigger the classification of the processing process; each peak of the third line corresponds to a trigger signal for one identification; and the fourth line represents the category of the part processed each time, with 0 indicating no algorithm triggered, 1 indicating an unknown category, and 2 and above corresponding to... Figure 2c The algorithm presents 11 categories. As can be seen from the figure, the algorithm distinguishes between different categories very accurately. Furthermore, the algorithm handles the situation where a factory mainly processes a certain type of part during a certain period, but occasionally produces other parts as well.

[0092] Finally, given the type of part being processed, the algorithm retrieves the corresponding noise feature template from the parameter library. During processing, the real-time generated noise feature signal is matched with the template signal. The current processing progress of the part can be determined from the length ratio of the best matching part in the template signal. Furthermore, based on the information in the template signal, the expected range of real-time noise features for a subsequent period can be given. This process can be achieved by simply traversing each substring of the template signal and performing intersection-comparison (OCC) matching with the existing signals. Alternatively, Markov chains, recurrent neural networks, or other methods can be used for prediction. When there is a significant deviation between the real-time signal and the predicted range, it can be considered that an anomaly has occurred in the processing. Figure 2g A diagram illustrating the real-time processing progress calculation using this method is provided, with the orange cross representing the estimated value of the subsequent signal predicted by the algorithm. This monitoring logic fails near the transition points of the template signal, but this problem can be avoided by combining it with transition point detection algorithms and changing the monitoring method to a backtracking approach after the transition in the vicinity of the template's transition points.

[0093] Example 3

[0094] Figure 3 This is a schematic diagram of a device testing apparatus provided in Embodiment 3 of the present invention. The device testing apparatus provided in this embodiment can be implemented by software and / or hardware, and can be configured in a server to implement a device testing method according to an embodiment of the present invention. Figure 3 As shown, the device may specifically include: a period determination module 301, a feature signal module 302, and a state detection module 303.

[0095] The cycle determination module 301 is used to collect the original noise of the target device and determine the processing cycle corresponding to the original noise based on the trigger signal.

[0096] The feature signal module 302 is used to determine the noise feature signal of the original noise in each of the processing cycles.

[0097] The status detection module 303 is used to determine the equipment status based on the comparison results between each of the noise characteristic signals and the preset processing object template signal.

[0098] In the technical solution provided by this embodiment of the invention, the original noise of the target equipment is collected by a cycle determination module, and the processing cycle corresponding to the original noise is determined according to the trigger signal. Then, a feature signal module is used to determine the noise feature signal of the original noise within each processing cycle. Finally, a status detection module determines the equipment status based on the comparison result between each noise feature signal and the preset processing object template signal. This embodiment of the technical solution accurately detects the status of the machine tool equipment by comparing the noise feature signal in the status detection module with the preset processing object template signal, promptly identifying faults, improving the reliability of equipment production, and enhancing product quality.

[0099] Optionally, based on the above embodiments, the period determination module 301 may specifically include:

[0100] The noise acquisition unit is specifically used to acquire the original noise at the machine tool spindle of the target equipment.

[0101] The processing cycle division unit is specifically used to divide the original noise into at least one processing cycle according to a trigger signal, wherein the trigger signal includes a device flushing pulse signal.

[0102] Optionally, based on the above embodiments, the feature signal module 302 may specifically include:

[0103] The feature signal acquisition unit can specifically be used to process the original noise within each processing cycle using preset processing operations to obtain noise feature signals. The preset processing operations include at least one of the following: the mean of high-frequency components, the standard deviation of high-frequency components, the mean of low-frequency components, and the standard deviations of both high and low-frequency components.

[0104] Optionally, based on the above embodiments, the state detection module 303 may specifically include:

[0105] A coarse classification determination unit is used to obtain the processing object of the target device and determine the coarse classification of the object corresponding to the processing object;

[0106] The template signal acquisition unit is used to acquire the preset processing object template signal within the object coarse classification.

[0107] The cross-intersection-to-parallel ratio (CIP) determination unit is used to determine the CIP ratio between the noise feature signal and the preset processing object template signal.

[0108] The comparison unit is used to compare the crossover ratio with a threshold. If the crossover ratio is less than the threshold, the target device is determined to be in a fault state. If the crossover ratio is greater than or equal to the threshold, the target device is determined to be in a normal state.

[0109] Optional, also includes:

[0110] The data acquisition module is used to acquire noise data of at least one processing object.

[0111] The feature signal extraction module is used to extract the noise feature signals of each noise data as a preset processing object template signal corresponding to the processing object.

[0112] Optionally, following the feature signal extraction module, the following may also be included:

[0113] The object coarse classification unit is specifically used to classify each preset processing object template signal into at least one object coarse classification according to the similarity of each preset processing object template signal.

[0114] The feature interval labeling unit is specifically used to label the feature intervals of each noise feature signal within the coarse classification of the object.

[0115] Optional, also includes:

[0116] The matching length determination module is used to determine the matching length between the noise feature signal and the preset processing object template signal within the processing cycle, and to use the matching length as the processing progress of the target device.

[0117] The equipment testing apparatus provided in this embodiment of the invention can execute the equipment testing method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of executing the method.

[0118] Example 4

[0119] Figure 4 This is a schematic diagram of the structure of a computer device provided in Embodiment 4 of the present invention, as shown below. Figure 4 As shown, the device includes a processor 401, a memory 402, an input device 403, and an output device 404; the number of processors 401 in the device can be one or more. Figure 4 Taking a processor 401 as an example; the processor 401, memory 402, input device 403, and output device 404 in the device can be connected via a bus or other means. Figure 4 Taking the example of a connection between China and Israel via a bus.

[0120] The memory 402, as a computer-readable storage medium, can be used to store software programs, computer-executable programs, and modules, such as the program instructions / modules corresponding to the device detection method in this embodiment of the invention (e.g., the period determination module 301, the feature signal module 302, and the status detection module 303 in the device detection apparatus). The processor 401 executes various functional applications and data processing of the device by running the software programs, instructions, and modules stored in the memory 402, thereby implementing the aforementioned device detection method.

[0121] The memory 402 may primarily include a program storage area and a data storage area. The program storage area may store the operating system and at least one application program required for a given function; the data storage area may store data created based on terminal usage. Furthermore, the memory 402 may include high-speed random access memory and non-volatile memory, such as at least one disk storage device, flash memory, or other non-volatile solid-state storage device. In some instances, the memory 402 may further include memory remotely located relative to the processor 401, which can be connected to the device / terminal / server via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0122] Input device 403 can be used to receive input digital or character information, and to generate key signal inputs related to user settings and function control of the device. Output device 404 may include display devices such as a display screen.

[0123] Example 5

[0124] Embodiment 5 of the present invention also provides a storage medium containing computer-executable instructions, which, when executed by a computer processor, are used to perform a device detection method, the method comprising:

[0125] The raw noise of the target device is collected, and the processing cycle corresponding to the raw noise is determined based on the trigger signal.

[0126] The noise characteristic signal of the original noise is determined within each of the aforementioned processing cycles.

[0127] The equipment status is determined based on the comparison results between each noise characteristic signal and the preset processing object template signal.

[0128] Of course, the computer-executable instructions provided in the embodiments of the present invention are not limited to the method operations described above, but can also perform related operations in the device detection method provided in any embodiment of the present invention.

[0129] Based on the above description of the implementation methods, those skilled in the art can clearly understand that the present invention can be implemented using software and necessary general-purpose hardware, and of course, it can also be implemented using hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk, or optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0130] It is worth noting that in the embodiments of the above-mentioned equipment testing device, the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of each functional unit are only for easy differentiation and are not used to limit the scope of protection of the present invention.

[0131] Note that the above description is merely a preferred embodiment of the present invention and the technical principles employed. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and various obvious changes, readjustments, and substitutions can be made without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in detail through the above embodiments, the present invention is not limited to the above embodiments, and may include many other equivalent embodiments without departing from the concept of the present invention, the scope of which is determined by the scope of the appended claims.

Claims

1. A device state detection method characterized by, The method comprises: Collecting original noise of a target device, and determining a processing period corresponding to the original noise according to a trigger signal; Determining a noise feature signal of the original noise in each processing period; Obtaining noise data of at least one processing object; Extracting a noise feature signal of each noise data as a preset processing object template signal corresponding to the processing object; Dividing each preset processing object template signal into at least one object coarse classification according to the similarity of each preset processing object template signal; Obtaining a processing object of the target device, and determining an object coarse classification corresponding to the processing object; Obtaining the preset processing object template signal in the object coarse classification; Comparing the noise feature signal of the original noise with the preset processing object template signal to determine the processing progress of the target device, and predicting the range of the noise feature signal of the next stage according to the preset processing object model signal to obtain a prediction result, which is used for comparison with the noise feature signal of the original noise generated subsequently by the target device to determine whether an abnormality occurs in the production process of the target device.

2. The method of claim 1, wherein, The collecting original noise of a target device, and determining a processing period corresponding to the original noise according to a trigger signal comprises: Collecting original noise at a spindle of a machine tool of the target device; Dividing the original noise into at least one processing period according to a trigger signal, wherein the trigger signal comprises a device flushing pulse signal.

3. The method of claim 1, wherein, The determining a noise feature signal of the original noise in each processing period comprises: Processing the original noise in each processing period using a preset processing operation to obtain a noise feature signal; The preset processing operation comprises at least one of a high-frequency component mean, a high-frequency component standard deviation, a low-frequency component mean, and a high-low frequency component standard deviation.

4. The method of claim 1, wherein, The method further comprises: Labeling a feature interval of each noise feature signal in the object coarse classification.

5. The method according to claim 1 or 4, characterized in that, The method further comprises: Determining an intersection-over-union of the noise feature signal and the preset processing object template signal; Comparing the intersection-over-union with a threshold value, if the intersection-over-union is less than the threshold value, determining that the target device is in a fault state, and if the intersection-over-union is greater than or equal to the threshold value, determining that the target device is in a normal state.

6. The method of claim 1, wherein, Further comprising: Determining a matching length of the noise feature signal in the processing period and the preset processing object template signal, and taking the matching length as the processing progress of the target device.

7. An apparatus state detection device, characterized by comprising: The device comprises: A period determination module for collecting original noise of a target device, and determining a processing period corresponding to the original noise according to a trigger signal; A feature signal module for determining a noise feature signal of the original noise in each processing period; The state detection module is configured to acquire noise data of at least one machining object, extract a noise feature signal of each of the noise data as a preset machining object template signal corresponding to the machining object, divide each of the preset machining object template signals into at least one object coarse classification according to a similarity of each of the preset machining object template signals, acquire a machining object of the target equipment, determine an object coarse classification corresponding to the machining object, acquire the preset machining object template signal in the object coarse classification, compare the noise feature signal of the original noise with the preset machining object template signal, determine a machining progress of the target equipment, and predict a range of a noise feature signal of a next stage according to the preset machining object template signal to obtain a prediction result, wherein the prediction result is used for comparison with a noise feature signal of original noise generated subsequently by the target equipment to determine whether an abnormality occurs in a production process of the target equipment.

8. A computer device, comprising: The computer device includes: one or more processors; a memory for storing one or more programs; when the one or more programs are executed by the one or more processors, the one or more processors implement the device state detection method as claimed in any one of claims 1-6.

9. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the device state detection method as claimed in any one of claims 1-6.

Images