Bolt tightening quality defect detection method, device, equipment and storage medium

Abstract

This invention provides a method, apparatus, device, and storage medium for detecting bolt tightening quality defects, relating to the field of intelligent detection technology. The method includes: acquiring tightening sequence data generated during the tightening of a current bolt; inputting the tightening sequence data into a prediction model; obtaining tightening data prediction results generated and output by the prediction model; the tightening data prediction results include: tightening prediction detail data, abnormal inflection points in the tightening prediction detail data, and prediction results of tightening quality defects for the current bolt. Through the high-resolution tightening prediction detail data generated and output by the prediction model, instantaneous tightening data mutations and subtle inflection points can be captured. Furthermore, the abnormal inflection points and tightening quality defect prediction results output by the prediction model overcome the problem that existing threshold methods can only provide warnings of abnormalities but cannot accurately determine the location and type of abnormality. Therefore, this invention significantly improves the accuracy of bolt tightening quality defect detection.

Description

Technical Field

[0001] This invention relates to the field of intelligent detection technology, and more specifically, to a method, apparatus, equipment, and storage medium for detecting bolt tightening quality defects. Background Technology

[0002] Bolts are indispensable fasteners in modern assembly processes. The quality of bolt tightening directly affects the safety and reliability of the final product. For example, in the prototype manufacturing stage of automobile manufacturing (i.e., the stage of making prototypes during new car development), the quality control of bolt tightening at some key connection points of the vehicle is very important. If the tightening quality of these key connection points does not meet the requirements, it will directly lead to safety hazards or malfunctions in the vehicle. For example, improper torque at the chassis connection points of the vehicle will cause abnormal noises and vibrations when the vehicle is driving, and in severe cases, it may even cause parts to fall off. Inaccurate torque on the engine block bolts will lead to air and oil leaks in the engine, affecting engine performance and even causing mechanical failures.

[0003] In related technologies, a torque control scheme based on "torque-angle threshold judgment" is commonly used to detect the tightening quality of bolts. This scheme collects data every fixed time interval (e.g., 10 milliseconds) using hardware such as torque sensors and angle encoders, and stops tightening when the torque or angle reaches a preset threshold. This scheme is limited by the sensor's acquisition rate, resulting in low resolution of the acquired data. It is difficult to capture instantaneous changes in data within 3-5 milliseconds, and it only judges abnormal data by static threshold comparison. The analysis logic is relatively rigid, leading to a high false judgment rate and making it difficult to meet the requirements of high-precision control. Summary of the Invention

[0004] The problem addressed by this invention is how to improve the accuracy of detecting defects in bolt tightening quality.

[0005] To address the aforementioned problems, this invention provides a method, apparatus, equipment, and storage medium for detecting bolt tightening quality defects.

[0006] In a first aspect, the present invention provides a method for detecting bolt tightening quality defects, comprising: Acquire tightening sequence data generated during the tightening of the current bolt; the tightening sequence data is one or more sets of data sampled at a first time interval, and each data point in the tightening sequence data is used to characterize the tightening state quantity; The tightening sequence data is input into the prediction model to obtain the tightening data prediction result generated and output by the prediction model; the tightening data prediction result includes at least: tightening prediction detail data, abnormal inflection points in the tightening prediction detail data, and the tightening quality defect prediction result of the current bolt; the time interval between every two adjacent data points in the tightening prediction detail data is the second time interval, and the second time interval is less than the first time interval.

[0007] Optionally, the prediction model generates the tightening data prediction result in the following manner: Based on the tightening sequence data, and combined with the tightening data change patterns that match the current bolt connection scenario, tightening prediction detail data is generated. Based on the comparison results between the tightening prediction detail data and the standard data, abnormal inflection points in the tightening prediction detail data are generated. Based on the tightening prediction details and the abnormal inflection points in the tightening prediction details, and combined with the relationship between the preset tightening quality defect type and the tightening curve, the tightening quality defect prediction result of the current bolt is generated.

[0008] Optionally, the step of generating tightening prediction detail data based on the tightening sequence data and combining it with the tightening data change patterns that match the current bolt connection scenario includes: Search the database for a tightening data change pattern that matches the current bolt connection scenario; If a matching tightening data change pattern exists, the tightening data change pattern is used to predict the tightening detail data corresponding to the current bolt tightening process, so as to obtain the tightening prediction detail data. If no matching tightening data change pattern exists, the preset default change pattern is used as the matching tightening data change pattern. Then, the tightening data change pattern is used to predict the tightening detail data corresponding to the current bolt tightening process, so as to obtain the tightening prediction detail data.

[0009] Optionally, after generating tightening prediction detail data based on the tightening sequence data and the tightening data change pattern matching the current bolt connection scenario, the method further includes: For each sampling time point of the tightening sequence data, the prediction data point of the time point closest to the sampling time point is extracted from the tightening prediction detail data, and the extracted prediction data point is compared with the data point in the tightening sequence data corresponding to the sampling time point to obtain the prediction deviation; The prediction deviation is obtained by comparing the data points corresponding to the tightening sequence data. If the prediction deviation meets the preset conditions, the tightening prediction detail data generated by the prediction model is output; otherwise, the parameters of the prediction model are adjusted.

[0010] Optionally, generating abnormal inflection points in the tightening prediction detail data based on the comparison results between the tightening prediction detail data and standard data includes: Determine the probability of deviation between each data point in the tightening prediction detail data and the standard data in a preset feature; the preset feature is a quantified value characterizing the data change pattern; If the deviation probability of the data point is greater than a preset threshold, the data point is identified as an abnormal inflection point.

[0011] Optionally, the tightening quality defect prediction result includes: the tightening quality defect prediction type and its corresponding confidence level; the confidence level is used to characterize the prediction accuracy of the tightening quality defect prediction type.

[0012] Optionally, the tightening sequence data includes: torque sequence data and / or angle sequence data.

[0013] Optionally, it also includes: displaying the tightening sequence data and / or the tightening data prediction results in a visualization interaction module; and / or, The tightening sequence data and the tightening data prediction results are stored in the data storage module.

[0014] Secondly, the present invention provides a bolt tightening quality defect detection device, comprising: The data acquisition module is used to acquire tightening sequence data generated during the tightening of the current bolt; the tightening sequence data is one or more sets of data sampled at a first time interval, and each data point in the tightening sequence data is used to characterize the tightening state quantity. The model prediction module is used to input the tightening sequence data into the prediction model to obtain the tightening data prediction results generated and output by the prediction model. The tightening data prediction results include: tightening prediction detail data, abnormal inflection points in the tightening prediction detail data, and the tightening quality defect prediction results of the current bolt. The time interval between every two data points in the tightening prediction detail data is the second time interval, which is less than the first time interval.

[0015] Thirdly, the present invention provides an electronic device, including a memory and a processor; The memory is used to store computer programs; The processor is configured to, when executing the computer program, implement the bolt tightening quality defect detection method as described in the first aspect.

[0016] Fourthly, the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the bolt tightening quality defect detection method as described in the first aspect.

[0017] The beneficial effects of the bolt tightening quality defect detection method, apparatus, equipment, and storage medium of the present invention are as follows: Tightening sequence data generated during the tightening of the current bolt is acquired. The tightening sequence data consists of one or more sets of data sampled at a first time interval. Each data point in the tightening sequence data is used to characterize the tightening state quantity, providing data support for subsequent tightening data analysis and prediction. The tightening sequence data is input into a prediction model to obtain the tightening data prediction results generated and output by the prediction model. The tightening data prediction results include: tightening prediction detail data, abnormal inflection points in the tightening prediction detail data, and the current bolt tightening quality defect prediction results. The time interval between every two data points in the tightening prediction detail data is a second time interval, which is shorter than the first time interval. The high-resolution tightening prediction detail data generated and output by the prediction model can capture instantaneous tightening data mutations and detect some subtle inflection points. Furthermore, the abnormal inflection points and tightening quality defect prediction results output by the prediction model overcome the problem that existing threshold methods can only warn of abnormalities but cannot accurately determine the location and type of abnormality. Therefore, the present invention significantly improves the accuracy of bolt tightening quality defect detection. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the structure of a bolt tightening quality defect detection system according to one embodiment; Figure 2 This is a flowchart of a bolt tightening quality defect detection method according to an embodiment of the present invention; Figure 3 A flowchart for generating tightening data prediction results for a prediction model in one embodiment; Figure 4 A flowchart for generating detailed tightening prediction data for a prediction model in one embodiment; Figure 5 A flowchart for generating anomalous inflection points in tightening prediction detail data for a prediction model of one embodiment; Figure 6 This is a schematic diagram of a bolt tightening quality defect detection device according to an embodiment of the present invention; Figure 7 This is a schematic diagram of the structure of an electronic device according to an embodiment of the present invention. Detailed Implementation

[0019] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Although some embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of the present invention. It should be understood that the accompanying drawings and embodiments of the present invention are for illustrative purposes only and are not intended to limit the scope of protection of the present invention.

[0020] It should be understood that the various steps described in the method embodiments of the present invention may be performed in different orders and / or in parallel. Furthermore, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present invention is not limited in this respect.

[0021] The term "comprising" and its variations as used herein are open-ended, meaning "including but not limited to"; the term "based on" means "at least partially based on"; the term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments"; and the term "optionally" means "optional embodiments". Definitions of other terms will be given in the following description. It should be noted that the concepts of "first," "second," etc., mentioned in this invention are used only to distinguish different devices, modules, or units, and are not intended to limit the order of functions performed by these devices, modules, or units or their interdependencies.

[0022] It should be noted that the terms "a" and "a plurality of" used in this invention are illustrative rather than restrictive. Those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0023] The names of the messages or information exchanged between the multiple devices in the embodiments of the present invention are for illustrative purposes only and are not intended to limit the scope of these messages or information.

[0024] like Figure 1 As shown, a bolt tightening quality defect detection system includes a data acquisition device 110 and a processor 120. The data acquisition device 110 includes a torque sensor 111, an angle sensor 112, and a data acquisition card 113. The torque sensor 111 is used to measure and acquire the electrical signal corresponding to the torque applied during bolt tightening. The angle sensor 112 is used to measure and acquire the electrical signal corresponding to the rotation angle during bolt tightening. The data acquisition card 113 is used to convert the electrical signals acquired by the torque sensor 111 and the angle sensor 112 into digital signals to obtain tightening sequence data, and output it to the processor 120.

[0025] The processor 120 can be an industrial computer or other processing device with data processing and analysis capabilities. The processor 120 is used to process and analyze the tightening sequence data generated during the bolt tightening process acquired by the data acquisition device 110.

[0026] In some embodiments, the bolt tightening quality defect detection system further includes a visualization interaction module 130 and a data storage module 140. The visualization interaction module 130 is used to display tightening sequence data and tightening quality defect prediction results obtained based on the tightening sequence data processing. The data storage module 140 is used to store the tightening sequence data and the tightening quality defect prediction results obtained based on the tightening sequence data processing.

[0027] like Figure 2 As shown, this embodiment of the invention provides a method for detecting bolt tightening quality defects. This method is applied in a processor 120 and includes the following steps: Step S210: Obtain the tightening sequence data generated during the tightening of the current bolt.

[0028] Specifically, taking automobile manufacturing as an example, bolts can currently be bolts at critical connection points. Critical connection points refer to the connection positions of bolted components in automobile manufacturing that play a decisive role in safety and reliability, including engine block connection bolts, chassis suspension system connection points, steering system connection components, etc.

[0029] Specifically, the tightening sequence data consists of one or more sets of data sampled at a first time interval. Each data point in the tightening sequence data represents a tightening state quantity. The first time interval between every two data points in the tightening sequence data is determined by the sensor sampling rate in the data acquisition device 110. For example, for the torque sensor 111 and the angle sensor 112, data is sampled at a rate of once every 10 milliseconds, corresponding to a first time interval of 10 milliseconds. The torque sensor 111 and the angle sensor 112 sample synchronously, transmitting the sampled analog electrical signals (such as current signals) to the data acquisition card 113. The data acquisition card 113 converts the two analog electrical signals into digital signals and sends them to the processor 120, obtaining two synchronous torque sequence data and angle sequence data, thus obtaining the tightening sequence data. In this embodiment, each data point in the tightening sequence data can represent the torque and angle during the tightening process.

[0030] In some embodiments, after acquiring the tightening sequence data, it is necessary to preprocess the tightening sequence data to remove noise and interference, thereby avoiding affecting subsequent data processing. For example, preprocessing the tightening sequence data may include the following steps: Step S211: Cleaning outliers: The tightening sequence data may contain outliers (such as occasional sensor vibrations that suddenly measure a torque far exceeding the normal range). This step will delete these "outliers," leaving only the data within the normal range. The criterion for judging outliers is: if a data point exceeds the normal range (set based on historical data) by 20%, it is judged as an outlier.

[0031] Step S212: Dynamic Denoising: There may be noise in the tightening sequence data (such as small errors in the sensor itself). In this embodiment, existing denoising algorithms (such as wavelet transform) are used to remove noise, but the denoising intensity is adjusted according to the bolt connection scenario. For example, in the hard connection scenario, the torque itself fluctuates greatly, so less noise is removed to avoid removing useful details; in the soft connection scenario, the torque fluctuates less, so more noise is removed to make the data more stable.

[0032] Step S213: Normalization: Since the unit of torque is Newton-meter and the unit of angle is degree, the two units are different. Therefore, it is necessary to convert both types of data into values ​​between 0 and 1. For example, the maximum torque corresponds to 1, the minimum torque corresponds to 0, the maximum angle corresponds to 1, and the minimum angle corresponds to 0, in order to eliminate the unit difference and facilitate subsequent calculations.

[0033] Step S220: Input the tightening sequence data into the prediction model to obtain the tightening data prediction results generated and output by the prediction model.

[0034] Specifically, the tightening data prediction results include at least: tightening prediction detail data, abnormal inflection points in the tightening prediction detail data, and the prediction results of the current bolt tightening quality defects.

[0035] Specifically, the time interval between every two adjacent data points in the tightening prediction detail data is the second time interval, which is shorter than the first time interval. For example, if the first time interval is 10 milliseconds, the second time interval can be 1 millisecond. That is, if the tightening sequence data includes a torque of 0.5 at millisecond 0 and a torque of 0.8 at millisecond 10, then the tightening prediction detail data includes a torque of 0.5 at millisecond 0, a torque of 0.53 at millisecond 1, a torque of 0.56 at millisecond 2, ..., a torque of 0.77 at millisecond 9 and a torque of 0.8 at millisecond 10. In other words, the prediction model can output tightening prediction detail data with higher time resolution, showing more subtle data changes and not missing transient issues in the bolt tightening process.

[0036] Specifically, by identifying abnormal inflection points in the tightening prediction details output by the prediction model, early problems such as bolt thread damage can be detected in advance, preventing potentially defective parts from flowing into the next process.

[0037] Specifically, the current prediction results for bolt tightening quality defects include: the predicted type of tightening quality defect and its corresponding confidence level, where the confidence level is used to characterize the prediction accuracy of the predicted type of tightening quality defect. In this way, compared with the existing single threshold judgment method, which can only obtain the result of whether it is abnormal, the prediction model can output specific defect prediction types, and can distinguish between seemingly similar defect types such as bolt stripping, repeated tightening, and tool error.

[0038] In some embodiments, the prediction model is a pre-trained deep learning-based artificial intelligence model that can learn from a large amount of historical data, grasp patterns, and reason and predict new data without the need to design separate algorithms for each task.

[0039] In this embodiment, tightening sequence data generated during the tightening of the current bolt is acquired. The tightening sequence data consists of one or more sets of data sampled at a first time interval. Each data point in the tightening sequence data is used to characterize the tightening state quantity, providing data support for subsequent tightening data analysis and prediction. The tightening sequence data is input into the prediction model to obtain the tightening data prediction results generated and output by the prediction model. The tightening data prediction results include: tightening prediction detail data, abnormal inflection points in the tightening prediction detail data, and the prediction results of tightening quality defects of the current bolt. The time interval between every two data points in the tightening prediction detail data is the second time interval, which is shorter than the first time interval. The high-resolution tightening prediction detail data generated and output by the prediction model can capture instantaneous tightening data mutations and detect some subtle inflection points. Furthermore, the abnormal inflection points and tightening quality defect prediction results output by the prediction model overcome the problem that the existing threshold method can only warn of whether there is an anomaly but cannot accurately determine the location of the anomaly and the type of defect. Thus, the accuracy of bolt tightening quality defect detection is significantly improved.

[0040] Furthermore, this embodiment uses the existing data acquisition device 110 (torque sensor 111, angle sensor 112, and data acquisition card 113) in the production line, without the need to add or replace new hardware devices.

[0041] Optionally, such as Figure 3 As shown, the prediction model generates tightening data prediction results in the following ways: Step S310: The prediction model generates detailed tightening prediction data based on the tightening sequence data and the tightening data change pattern that matches the current bolt connection scenario.

[0042] Specifically, bolt connection scenarios include flexible connection scenarios and rigid connection scenarios. Flexible connection scenarios refer to the presence of elastic materials (such as gaskets or rubber) in the connected parts, which allow for a certain amount of displacement and vibration absorption. Rigid connection scenarios refer to the fact that the connected parts are all rigid bodies (such as steel and cast iron), which rely on high preload to ensure a tight fit.

[0043] Specifically, during the training of the prediction model, a large amount of historical tightening sequence data is input into the model so that it can remember the tightening data variation patterns under different connection scenarios. For example, in the hard connection scenario, the tightening data variation pattern is fast at first and then slows down; in the soft connection scenario, the tightening data variation pattern is uniform and linear. These tightening data variation patterns corresponding to different connection scenarios are pre-stored in the database.

[0044] Specifically, the prediction model derives detailed tightening prediction data based on the input tightening sequence data and the matching tightening data variation patterns. For example, if the input tightening sequence data to the prediction model is torque A at millisecond 0 and torque B at millisecond 10, the prediction model derives the torque value for each moment from millisecond 1 to millisecond 9 based on the matching tightening data variation patterns, completing the detailed data within 10 milliseconds to achieve a 1-millisecond resolution.

[0045] In some embodiments, such as Figure 4 As shown, the prediction model generates detailed tightening prediction data based on tightening sequence data and combined with the tightening data variation patterns that match the current bolt connection scenario. This includes the following steps: Step S410: Call the database that stores the data change patterns.

[0046] Step S420: Search the database for a tightening data change pattern that matches the current bolt connection scenario.

[0047] Step S430: If a matching tightening data change pattern is found, the tightening data change pattern is used to predict the tightening detail data corresponding to the current bolt tightening process, so as to obtain the tightening prediction detail data.

[0048] Step S440: If no matching tightening data change pattern is found, the preset default change pattern is used as the matching tightening data change pattern. Then, the tightening data change pattern is used to predict the tightening detail data corresponding to the current bolt tightening process, so as to obtain the tightening prediction detail data.

[0049] Step S450: For each sampling time point of the tightening sequence data, extract the prediction data point of the time point closest to the sampling time point from the tightening prediction detail data, and compare the extracted prediction data point with the data point in the tightening sequence data corresponding to the sampling time point to obtain the prediction deviation.

[0050] For example, by comparing the derived torque prediction value at the 9th second with the actual sampled torque value at the 10th second, the deviation between the two can be obtained, i.e., the prediction deviation.

[0051] Step S460: If the prediction deviation meets the preset conditions, the prediction model outputs the prediction detail data; otherwise, the parameters of the prediction model are adjusted.

[0052] Specifically, the preset condition is that the percentage of prediction deviation is less than a preset threshold, such as less than 5%. The percentage of prediction deviation is the percentage of the absolute difference between the predicted value and the actual sampled value to the actual measured value.

[0053] Specifically, the accuracy of the derived tightening prediction details is ensured by performing deviation detection on the tightening prediction details data.

[0054] Step S320: Based on the comparison results between the tightening prediction detail data and the standard data, the prediction model generates abnormal inflection points in the tightening prediction detail data.

[0055] Specifically, the standard data is the curve corresponding to the standard tightening data, such as the standard curve that matches the current bolt.

[0056] In this embodiment, the prediction model learns the curve corresponding to the standard tightening data, and then compares it with the tightening prediction detail data to find the data points of the curve corresponding to the tightening data that deviate from the standard. These points are the abnormal inflection points.

[0057] In some embodiments, such as Figure 5 As shown, based on the comparison results between the tightening prediction detail data and the standard data, abnormal inflection points in the tightening prediction detail data are generated, including: Step S510: Determine the probability of deviation between each data point in the tightening prediction detail data and the standard data in the preset features.

[0058] Specifically, the preset feature is a quantitative value that characterizes the pattern of data change, such as the change in torque per millisecond.

[0059] Specifically, the deviation probability characterizes the degree of difference between a data point in the tightening prediction detail data and the standard data; that is, the similarity between the data point in the tightening prediction detail data and the standard data. The lower the similarity, the higher the deviation probability. For example, in the standard data, the torque change per millisecond does not exceed 10%. If the torque change per millisecond of a data point in the tightening prediction detail data is 30%, then the deviation probability of that data point is high. The deviation probability can be calculated using the existing cosine similarity method.

[0060] Step S520: If the deviation probability of a data point is greater than a preset threshold, then the data point is identified as an abnormal inflection point.

[0061] This embodiment can avoid missing minor anomalies in some data points by judging the abnormal inflection points of each data point in the tightening prediction detail data.

[0062] Step S330: The prediction model generates the tightening quality defect prediction result of the current bolt based on the tightening prediction detail data and the abnormal inflection points in the tightening prediction detail data, combined with the relationship between the preset tightening quality defect type and the tightening curve.

[0063] Specifically, a large number of tightening curves corresponding to different defect types are pre-inputted into the prediction model, allowing the model to memorize the correspondence between defect types and curve characteristics. For example, the curve corresponding to a stripped bolt defect is a sudden drop in torque followed by stabilization; the curve corresponding to a repetitive tightening defect is a torque that first rises, then falls, and then rises again; and the curve corresponding to a thread damage defect is an abnormal inflection point that appears before the torque exceeds the threshold.

[0064] The prediction model is based on the tightening prediction details and the abnormal inflection points in the tightening prediction details. It combines the pre-set relationship between the tightening quality defect type and the tightening curve to output the tightening quality defect prediction result, that is, the output tightening quality defect prediction type and its corresponding confidence level. The confidence level is used to characterize the prediction accuracy of the tightening quality defect prediction type. For example, the output is thread damage with a confidence level of 96%.

[0065] If the prediction model is based solely on the tightening prediction detail data, combined with the pre-defined relationship between the tightening quality defect type and the tightening curve, and outputs the tightening quality defect prediction result, then, if the torque exceeds the threshold, the prediction model will judge it as tool error. If abnormal inflection points are added to the tightening prediction detail data, it will be judged as thread damage defect, thus improving the accuracy of bolt tightening quality defect detection.

[0066] In this optional embodiment, the prediction model generates detailed tightening prediction data based on tightening sequence data and the tightening data variation patterns matching the current bolt connection scenario. This allows for the acquisition of more subtle instantaneous changes during the tightening process. Based on the comparison results between the detailed tightening prediction data and standard data, the prediction model generates abnormal inflection points in the detailed tightening prediction data, enabling early detection of problems such as thread damage and preventing potentially defective components from flowing into the next process. Furthermore, based on the detailed tightening prediction data and the abnormal inflection points within it, and combined with the pre-defined relationship between tightening quality defect types and tightening curves, the prediction model generates a tightening quality defect prediction result for the current bolt. This result can identify specific defect types with high accuracy.

[0067] Optionally, the tightening sequence data, tightening prediction detail data, abnormal inflection points in the tightening prediction detail data, and tightening quality defect prediction results can be displayed in the visualization interaction module 130.

[0068] Specifically, the interface of the visualization interaction module 130 displays the curve corresponding to the tightening sequence data and the curve corresponding to the tightening prediction detail data. The two curves are displayed in different colors; for example, the black curve is the curve corresponding to the tightening sequence data, and the red curve is the curve corresponding to the tightening prediction detail data. Abnormal inflection points in the tightening prediction detail data can be marked on the curve corresponding to the tightening prediction detail data, for example, with a yellow flashing dot. When the operator hovers the cursor over the yellow flashing dot, a prompt box will pop up, displaying the time corresponding to the abnormal inflection point (e.g., the 5th millisecond), the torque change magnitude (e.g., a decrease of 15%), and the predicted defect type (e.g., thread damage). Furthermore, the visualization interaction module 130 supports zooming and panning operations on the curves to facilitate the operator's observation of curve details.

[0069] In this optional embodiment, the data is displayed to the operator through the visualization interaction module 130 so that the operator can intuitively observe the changes in the data.

[0070] Optionally, the tightening sequence data and the tightening quality defect prediction results obtained by processing the tightening sequence data are stored in the data storage module 140.

[0071] Specifically, the data storage module 140 stores all data from the tightening sequence data input to the prediction model to the tightening quality defect prediction result output by the prediction model, including: tightening sequence data, tightening prediction detail data, abnormal inflection points in the tightening prediction detail data, and tightening quality defect prediction results. In some embodiments, a time-series database (Infx database) can be used for storage. Since the time-series database specifically stores data generated in chronological order, it is convenient to query data according to tightening task number, tightening time, etc. In addition, the stored data can be used to optimize the prediction model. For example, new defect data collected later can be added to the database of tightening quality defect types and tightening curves, making the prediction accuracy of the prediction model increasingly higher.

[0072] like Figure 6 As shown, an embodiment of the present invention provides a bolt tightening quality defect detection device 600, comprising: The data acquisition module 610 is used to acquire tightening sequence data generated during the tightening of the current bolt; the tightening sequence data is one or more sets of data sampled at a first time interval, and each data point in the tightening sequence data is used to characterize the tightening state quantity. The model prediction module 620 is used to input the tightening sequence data into the prediction model to obtain the tightening data prediction result generated and output by the prediction model. The tightening data prediction result includes: tightening prediction detail data, abnormal inflection points in the tightening prediction detail data, and the tightening quality defect prediction result of the current bolt. The time interval between every two data points in the tightening prediction detail data is the second time interval, which is less than the first time interval.

[0073] Optionally, the model prediction module 620 is further configured to: The prediction module generates detailed tightening prediction data based on the tightening sequence data and the tightening data change pattern that matches the current bolt connection scenario. The prediction module generates abnormal inflection points in the tightening prediction detail data based on the comparison results between the tightening prediction detail data and the standard data. The prediction module generates a prediction result for the tightening quality defect of the current bolt based on the tightening prediction detail data and the abnormal inflection points in the tightening prediction detail data, combined with the relationship between the preset tightening quality defect type and the tightening curve.

[0074] Optionally, the step of generating tightening prediction detail data based on the tightening sequence data and combining it with the tightening data change patterns that match the current bolt connection scenario includes: Search the database for a tightening data change pattern that matches the current bolt connection scenario; If a matching tightening data change pattern is found, the tightening data change pattern is used to predict the tightening detail data corresponding to the current bolt tightening process, so as to obtain the tightening prediction detail data. If no matching tightening data change pattern is found, the preset default change pattern is used as the matching tightening data change pattern. Then, the tightening data change pattern is used to predict the tightening details data corresponding to the current bolt tightening process, so as to obtain the tightening prediction details data.

[0075] Optionally, after generating tightening prediction detail data based on the tightening sequence data and the tightening data change pattern matching the current bolt connection scenario, the method further includes: For each sampling time point of the tightening sequence data, the prediction data point of the time point closest to the sampling time point is extracted from the tightening prediction detail data, and the extracted prediction data point is compared with the data point in the tightening sequence data corresponding to the sampling time point to obtain the prediction deviation; The prediction deviation is obtained by comparing the data points corresponding to the tightening sequence data. If the prediction deviation meets the preset conditions, the prediction model outputs the tightening prediction details; otherwise, the prediction model is adjusted.

[0076] Optionally, generating abnormal inflection points in the tightening prediction detail data based on the comparison results between the tightening prediction detail data and standard data includes: Determine the probability of deviation between each data point in the tightening prediction detail data and the standard data in a preset feature; the preset feature is a quantified value characterizing the data change pattern; If the deviation probability of the data point is greater than a preset threshold, the data point is identified as an abnormal inflection point.

[0077] Optionally, the tightening quality defect prediction result includes: the tightening quality defect prediction type and its corresponding confidence level; the confidence level is used to characterize the prediction accuracy of the tightening quality defect prediction type.

[0078] Optionally, the tightening sequence data includes: torque sequence data and / or angle sequence data.

[0079] Optionally, it may also include: a visual interaction module and / or a data storage module; Display the tightening sequence data and / or the tightening data prediction results in the visualization and interaction module; The tightening sequence data and the tightening data prediction results are stored in the data storage module.

[0080] like Figure 7 As shown, an electronic device 700 provided in this embodiment of the invention includes a memory 710 and a processor 720; the memory 710 is used to store a computer program; the processor 720 is used to implement the bolt tightening quality defect detection method as described above when the computer program is executed.

[0081] Alternatively, an electronic device 700 includes a memory 710 and a processor 720 coupled to the memory 710; the memory 710 is configured to store a computer program; and the processor 720 is configured to perform the following operations when the computer program is executed: Acquire tightening sequence data generated during the tightening of the current bolt; the tightening sequence data is one or more sets of data sampled at a first time interval, and each data point in the tightening sequence data is used to characterize the tightening state quantity; The tightening sequence data is input into the prediction model to obtain the tightening data prediction result generated and output by the prediction model; the tightening data prediction result includes at least: tightening prediction detail data, abnormal inflection points in the tightening prediction detail data, and the tightening quality defect prediction result of the current bolt; the time interval between every two data points in the tightening prediction detail data is the second time interval, which is less than the first time interval.

[0082] This invention provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the bolt tightening quality defect detection method described above.

[0083] Alternatively, a non-volatile computer-readable storage medium storing a computer program that, when executed by a processor, causes the processor to perform the following operations: Acquire tightening sequence data generated during the tightening of the current bolt; the tightening sequence data is one or more sets of data sampled at a first time interval, and each data point in the tightening sequence data is used to characterize the tightening state quantity; The tightening sequence data is input into the prediction model to obtain the tightening data prediction result generated and output by the prediction model; the tightening data prediction result includes at least: tightening prediction detail data, abnormal inflection points in the tightening prediction detail data, and the tightening quality defect prediction result of the current bolt; the time interval between every two data points in the tightening prediction detail data is the second time interval, which is less than the first time interval.

[0084] The present invention will now be described an electronic device 700 that can serve as a server or client of the present invention, which is an example of a hardware device that can be applied to various aspects of the present invention. Electronic device 700 is intended to represent various forms of digital electronic computer devices, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. Electronic device 700 can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0085] Electronic device 700 includes a computing unit that can perform various appropriate actions and processes based on a computer program stored in read-only memory (ROM) or a computer program loaded from a storage unit into random access memory (RAM). The RAM may also store various programs and data required for device operation. The computing unit, ROM, and RAM are interconnected via a bus. Input / output (I / O) interfaces are also connected to the bus.

[0086] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc. In this application, the units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of the embodiments of the present invention according to actual needs. Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated units can be implemented in hardware or as software functional units.

[0087] While the present invention has been disclosed above, its scope of protection is not limited thereto. Those skilled in the art can make various changes and modifications without departing from the spirit and scope of the present invention, and all such changes and modifications will fall within the scope of protection of the present invention.

Claims

1. A method of detecting a tightening quality defect of a bolt, characterized by, include: Obtain the tightening sequence data generated during the tightening of the current bolt; The tightening sequence data is one or more sets of data sampled at a first time interval, and each data point in the tightening sequence data is used to characterize the tightening state quantity. The tightening sequence data is input into the prediction model to obtain the tightening data prediction result generated and output by the prediction model; the tightening data prediction result includes at least: tightening prediction detail data, abnormal inflection points in the tightening prediction detail data, and the tightening quality defect prediction result of the current bolt; the time interval between every two adjacent data points in the tightening prediction detail data is the second time interval, and the second time interval is less than the first time interval.

2. The method of claim 1, wherein The prediction model generates the tightening data prediction results in the following ways: Based on the tightening sequence data, and combined with the tightening data change patterns that match the current bolt connection scenario, tightening prediction detail data is generated. Based on the comparison results between the tightening prediction detail data and the standard data, abnormal inflection points in the tightening prediction detail data are generated. Based on the tightening prediction details and the abnormal inflection points in the tightening prediction details, and combined with the relationship between the preset tightening quality defect type and the tightening curve, the tightening quality defect prediction result of the current bolt is generated.

3. The method of claim 2, wherein Based on the tightening sequence data, and combined with the tightening data change patterns matching the current bolt connection scenario, tightening prediction detail data is generated, including: Search the database for a tightening data change pattern that matches the current bolt connection scenario; If a matching tightening data change pattern exists, the tightening data change pattern is used to predict the tightening detail data corresponding to the current bolt tightening process, so as to obtain the tightening prediction detail data. If no matching tightening data change pattern exists, the preset default change pattern is used as the matching tightening data change pattern. Then, the tightening data change pattern is used to predict the tightening detail data corresponding to the current bolt tightening process, so as to obtain the tightening prediction detail data.

4. The method of claim 3, wherein After generating tightening prediction detail data based on the tightening sequence data and the tightening data change pattern matching the current bolt connection scenario, the process further includes: For each sampling time point of the tightening sequence data, the prediction data point of the time point closest to the sampling time point is extracted from the tightening prediction detail data, and the extracted prediction data point is compared with the data point in the tightening sequence data corresponding to the sampling time point to obtain the prediction deviation; If the prediction deviation meets the preset conditions, the tightening prediction detail data generated by the prediction model is output; otherwise, the parameters of the prediction model are adjusted.

5. The method of claim 2, wherein The process of generating abnormal inflection points in the tightening prediction detail data based on the comparison results between the tightening prediction detail data and the standard data includes: Determine the probability of deviation between each data point in the tightening prediction detail data and the standard data in a preset feature; the preset feature is a quantified value characterizing the data change pattern; If the deviation probability of the data point is greater than a preset threshold, the data point is identified as an abnormal inflection point.

6. The method of claim 2, wherein The tightening quality defect prediction results include: tightening quality defect prediction types and their corresponding confidence levels; the confidence levels are used to characterize the prediction accuracy of the tightening quality defect prediction types.

7. The method of detecting a tightening quality defect of a bolt according to any one of claims 1 to 6, characterized in that, The tightening sequence data includes: torque sequence data and / or angle sequence data.

8. The method of detecting a tightening quality defect of a bolt according to any one of claims 1 to 7, characterized in that, Also includes: Display the tightening sequence data and / or the tightening data prediction results in the visualization and interaction module; And / or, The tightening sequence data and the tightening data prediction results are stored in the data storage module.

9. A device for detecting defects in bolt tightening quality, characterized in that, include: The data acquisition module is used to acquire the tightening sequence data generated during the tightening of the current bolt; The tightening sequence data is one or more sets of data sampled at a first time interval, and each data point in the tightening sequence data is used to characterize the tightening state quantity. The model prediction module is used to input the tightening sequence data into the prediction model to obtain the tightening data prediction results generated and output by the prediction model. The tightening data prediction results include: tightening prediction detail data, abnormal inflection points in the tightening prediction detail data, and the tightening quality defect prediction results of the current bolt. The time interval between every two data points in the tightening prediction detail data is the second time interval, which is shorter than the first time interval.

10. An electronic device, characterized in that, Including memory and processor; The memory is used to store computer programs; The processor is configured to, when executing the computer program, implement the bolt tightening quality defect detection method as described in any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that, The storage medium stores a computer program, which, when executed by a processor, implements the bolt tightening quality defect detection method as described in any one of claims 1 to 8.

Images