Bolt tightening quality real-time monitoring method and device
By generating torque-angle curves and extracting multidimensional features using the sliding window method, the problem of lack of monitoring in traditional bolt tightening operations is solved, enabling real-time monitoring of the reliability and safety of bolt connections and constructing a closed-loop management system for the entire process.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- JIANGSU XCMG CONSTRUCTION MACHINERY RESEARCH INSTITUTE LTD
- Filing Date
- 2025-11-27
- Publication Date
- 2026-04-21
AI Technical Summary
Traditional bolt tightening operations lack comprehensive monitoring of the tightening process, resulting in a lack of assurance regarding the quality and reliability of bolt connections.
By acquiring the torque and angle during the bolt tightening process from the tightening machine during the offline training phase, a torque-angle curve is generated. Data preprocessing and tightening quality classification are performed. Multidimensional features are extracted using the sliding window method, and feature threshold intervals are generated based on statistics. These are then configured into the monitoring system for real-time monitoring.
It has achieved systematic monitoring of the tightening process, enabling early and accurate identification of various abnormal modes, ensuring the reliability and safety of bolt connections, and forming a closed-loop management system covering the entire process from threshold training to real-time monitoring.
Smart Images
Figure CN121901916A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of manufacturing technology, and in particular to a method and device for real-time monitoring of bolt tightening quality. Background Technology
[0002] As the manufacturing industry moves towards digitalization and intelligence, the complexity and precision of production equipment are increasing, placing unprecedented demands on the reliability of equipment connection points. Bolt connections, as the most basic and widespread connection method in mechanical equipment, directly determine the overall structural safety, sealing performance, operational stability, and lifespan of the equipment through their tightening quality. Traditional bolt tightening operations mainly rely on operators using tightening guns, and quality control typically involves manually recording torque values on paper work orders. This lack of comprehensive monitoring of the tightening process makes it impossible to identify abnormal conditions such as sticking or slipping. Summary of the Invention
[0003] The purpose of this invention is to overcome the shortcomings of the prior art and provide a method and device for real-time monitoring of bolt tightening quality, which solves the problem that traditional bolt tightening operations lack comprehensive monitoring of the tightening process, and the quality and reliability of bolt connections cannot be guaranteed.
[0004] To achieve the above objectives, the present invention is implemented using the following technical solution: In a first aspect, the present invention provides a method for real-time monitoring of bolt tightening quality, comprising: During the offline training phase, the torque and angle during the bolt tightening process are obtained from the tightening machine to generate a torque-angle curve; The torque-angle curves are preprocessed and tightening quality is classified to generate sample curves and sample types. Multidimensional features are extracted from the sample curves using the sliding window method to generate multidimensional features. The multidimensional features of the sample curves for each sample type are summarized and analyzed, and the corresponding feature threshold intervals are generated using statistics. The feature threshold range for each sample type is configured to be monitored in real time by the monitoring system, and the real-time tightening quality is obtained during the online monitoring phase.
[0005] Optionally, the data preprocessing includes smoothing, denoising, normalizing, and interval filtering of the torque-angle curve.
[0006] Optionally, the tightening quality classification includes classifying the torque-angle curve into normal tightening and abnormal tightening based on process knowledge. The abnormal tightening includes sticking, slight sticking, jamming, slight jamming, repeated tightening, excessive tightening torque, and large slope fluctuation.
[0007] Optionally, the multidimensional feature extraction of the sample curve based on the sliding window method includes extracting multidimensional feature parameters of the sample curve within each sliding window. The multidimensional feature parameters include peak torque, final tightening angle, interval slope, root mean square error, number of extreme values, and abrupt change point data.
[0008] Optionally, the step of summarizing and analyzing the multidimensional features of the sample curves for each sample type and generating corresponding feature threshold intervals using statistics includes: For sample curves of the same sample type, calculate the mean and standard deviation of each feature within each sliding window; Based on the mean and the standard deviation, the feature threshold range of each dimension of the feature within each sliding window under the sample type is determined according to the 3σ principle.
[0009] Optionally, configuring the feature threshold range for each sample type to the monitoring system for real-time monitoring includes: During the online monitoring phase, the torque and angle during the bolt tightening process are obtained from the tightening machine to generate a torque-angle curve; The torque-angle curve is preprocessed, and multidimensional features are extracted from the preprocessed sample curve using the sliding window method to generate multidimensional features. The multidimensional features are matched with the feature threshold intervals of each sample type, and the sample type that matches the feature threshold interval the most times is taken as the real-time tightening quality.
[0010] Secondly, the present invention provides a real-time monitoring device for bolt tightening quality, comprising: The data acquisition module is configured to acquire the torque and angle during the bolt tightening process from the tightening machine during the offline training phase, and generate a torque-angle curve. The data processing module is configured to perform data preprocessing and tightening quality classification on the torque-angle curve, and generate sample curves and sample types; The feature extraction module is configured to perform multidimensional feature extraction on the sample curve based on the sliding window method to generate multidimensional features; The analysis and statistics module is configured to perform a summary analysis of the multidimensional features of the sample curves for each sample type and generate corresponding feature threshold intervals using statistics. The configuration application module is configured to configure the feature threshold range of each sample type to the monitoring system for real-time monitoring, and to obtain the real-time tightening quality during the online monitoring phase.
[0011] Thirdly, the present invention provides an electronic device, including a processor and a storage medium; The storage medium is used to store instructions; The processor is configured to operate according to the instructions to perform the steps according to the method described above.
[0012] Fourthly, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method.
[0013] Fifthly, the present invention provides a computer program product, including a computer program / instructions that, when executed by a processor, implement the steps of the above-described method.
[0014] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: This invention provides a method and device for real-time monitoring of bolt tightening quality. 1) It constructs a systematic multi-state historical dataset and statistical summary analysis of bolt tightening. By systematically collecting tightening process data of normal and various typical abnormalities (such as sticking, slipping, jamming, etc.), a standardized training sample set is formed. Through statistical summary analysis, the characteristic threshold ranges of each dimension of each sample type are obtained, effectively solving the problem of poor adaptability caused by the reliance on a single threshold in traditional methods, and laying a data and algorithm foundation for unified and accurate tightening quality judgment.
[0015] 2) A method for dynamic monitoring and anomaly diagnosis of the tightening process based on multidimensional features. In real-time monitoring, a sliding window technique is used to dynamically calculate the multidimensional features of the real-time data stream, and these features are compared with corresponding precise thresholds, enabling real-time and highly sensitive detection of anomalies in the tightening process. Compared to monitoring a single torque or angle, this method can identify multiple anomaly patterns earlier and more accurately.
[0016] 3) It achieves closed-loop management and integrated application of the entire process from threshold training and configuration to real-time monitoring. Standardized thresholds are generated through offline training, bound to specific equipment and processes by the configuration module, and then distributed to the monitoring system. Finally, they are applied and recorded in real-time judgment on the production line, forming an automated closed loop of "training-configuration-monitoring". This architecture ensures that the algorithm thresholds can be continuously optimized, realizing the standardization and digitalization of tightening process quality management. Attached Figure Description
[0017] Figure 1 This is a flowchart illustrating the real-time monitoring method for bolt tightening quality provided in an embodiment of the present invention; Figure 2 This is a flowchart illustrating the offline training phase provided in an embodiment of the present invention; Figure 3 This is a flowchart illustrating the online monitoring phase provided in an embodiment of the present invention. Detailed Implementation
[0018] The present invention will be further described below with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and should not be used to limit the scope of protection of the present invention.
[0019] Example 1
[0020] like Figure 1 As shown, the present invention provides a method for real-time monitoring of bolt tightening quality, comprising the following steps: Step S1: Obtain the torque and angle during the bolt tightening process from the tightening machine, and generate a torque-angle curve. During the offline training phase, the relevant data is historical data.
[0021] Step S2: Perform data preprocessing and tightening quality classification on the torque-angle curve to generate sample curves and sample types.
[0022] Specifically, in this embodiment, data preprocessing includes smoothing, denoising, normalizing, and interval filtering of the torque-angle curve to remove invalid data segments and form standardized historical data input (sample curve), providing a reliable foundation for subsequent feature extraction.
[0023] Tightening quality classification includes classifying torque-angle curves into normal tightening and abnormal tightening based on process knowledge. Abnormal tightening includes sticking and slipping, slight sticking and slipping, jamming, slight jamming, repeated tightening, exceeding the tightening torque limit, and large slope fluctuations.
[0024] Step S3: Extract multidimensional features from the sample curves using the sliding window method to generate multidimensional features.
[0025] Multidimensional feature parameters are extracted from the sample curves within each sliding window. These parameters include peak torque, final tightening angle, interval slope, root mean square error, number of extrema, and abrupt change data. The multidimensional features can be represented by a feature matrix, where each row of the feature matrix contains the multidimensional feature parameters for each sliding window.
[0026] Step S4: Summarize and analyze the multidimensional features of the sample curves for each sample type, and use statistics to generate the corresponding feature threshold intervals.
[0027] The specific process includes: For sample curves of the same sample type, calculate the mean and standard deviation of each feature within each sliding window; Based on the mean and standard deviation, the feature threshold range of each dimension of the feature within each sliding window under the sample type is determined according to the 3σ principle.
[0028] Taking peak torque as an example, its mean and standard deviation within the first window are μ and σ, respectively, and its characteristic threshold interval within the first window is: 1σ interval: [μ-σ, μ+σ] 2σ interval: [μ-2σ, μ+2σ] 3σ interval: [μ-3σ, μ+3σ] Step S5: Configure the feature threshold range for each sample type to the monitoring system for real-time monitoring.
[0029] Threshold parameter configuration and management: The feature threshold ranges for each sample type are imported into the system's backend database, and threshold parameters can be configured, modified, and version managed through a unified interface. The system supports configuring different threshold templates by workstation, bolt specification, equipment number, and other dimensions to ensure that the threshold settings match the actual assembly conditions.
[0030] Threshold Template Import, Export, and Sharing: To facilitate system maintenance and multi-workstation collaborative management, the module provides functions for importing, exporting, and sharing threshold templates. Users can export threshold templates trained under specific working conditions as standardized files for rapid deployment on other production lines; they can also synchronize templates to the server via the network to achieve model sharing and unified updates across factories and equipment.
[0031] Step S6: Obtain the torque and angle during the bolt tightening process from the tightening machine, and generate a torque-angle curve. During the online monitoring phase, the relevant data is real-time.
[0032] Step S7: Perform data preprocessing on the torque-angle curve, and extract multidimensional features from the preprocessed sample curves using the sliding window method to generate multidimensional features.
[0033] Step S8: Match the multidimensional features with the feature threshold range of each sample type, and take the sample type that matches the feature threshold range the most times as the real-time tightening quality.
[0034] When the system identifies abnormal curves, it automatically generates alarm information and provides corresponding handling suggestions based on the fault mode library. These suggestions include methods for handling abnormal situations such as sticking / slipping and repeated tightening, assisting operators in responding quickly and ensuring the safety and reliability of the bolt assembly process. The module supports pushing real-time analysis results to the operating terminal or error prevention screen, displaying curve graphs, pass / abnormal status, and abnormality type prompts, enabling real-time on-site monitoring and visual management.
[0035] Furthermore, data storage and report management: Real-time monitored tightening curve data, characteristic values, judgment results, and anomaly information are all stored in the database, supporting multi-level aggregation and querying by group—factory—production line—workstation. The system can generate visual reports for production process analysis, historical data traceability, and quality statistics, providing data support for process optimization and management decisions.
[0036] In summary, the real-time bolt tightening quality monitoring method provided in this embodiment of the invention, during the offline training phase, such as... Figure 2 As shown, in terms of tightening data acquisition, the tightening machine controller collects key parameters such as torque and angle during the bolt tightening process in real time, and filters the effective range to form standardized data input. Regarding bolt tightening data threshold training, the system is classified according to the type of threshold training scenario, introducing a classification data import module and a feature algorithm module to generate multi-dimensional feature value information such as RMSE, slope, and extreme point count based on a sliding window. The feature value threshold is calculated using the collected historical data, providing threshold configuration and application after the tightening equipment is trained. In the online monitoring phase, such as... Figure 3 As shown, in terms of real-time bolt tightening analysis, data is received and tightening curves are generated in real time, tightening results are analyzed in real time, and correction schemes for abnormal tightening data are provided. Bolt tightening reports are established at different levels, such as group, factory, production line, and section, to achieve multi-dimensional analysis of bolt tightening detection data.
[0037] Example 2
[0038] This invention provides a real-time monitoring device for bolt tightening quality, comprising: The data acquisition module is configured to acquire the torque and angle during the bolt tightening process from the tightening machine during the offline training phase, and generate a torque-angle curve. The data processing module is configured to preprocess the torque-angle curve and classify the tightening quality, generating sample curves and sample types; The feature extraction module is configured to extract multidimensional features from the sample curves based on the sliding window method, and generate multidimensional features. The analysis and statistics module is configured to summarize and analyze the multidimensional features of the sample curves for each sample type and generate corresponding feature threshold intervals using statistics. The configuration application module is configured to assign the feature threshold range for each sample type to the monitoring system for real-time monitoring, and to obtain the real-time tightening quality during the online monitoring phase.
[0039] Example 3
[0040] Based on the real-time monitoring method for bolt tightening quality provided in Embodiment 1, this embodiment of the invention provides an electronic device, including a processor and a storage medium; Storage media are used to store instructions; The processor is used to perform operations according to instructions to execute the steps according to the method described above.
[0041] Example 4
[0042] Based on the real-time monitoring method for bolt tightening quality provided in Embodiment 1, this embodiment of the invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the above method.
[0043] Example 5
[0044] Based on the real-time monitoring method for bolt tightening quality provided in Embodiment 1, this embodiment of the invention provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the steps of the above method.
[0045] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0046] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.
[0047] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.
[0048] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.
[0049] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.
Claims
1. A method for real-time monitoring of bolt tightening quality, characterized in that, include: During the offline training phase, the torque and angle during the bolt tightening process are obtained from the tightening machine to generate a torque-angle curve; The torque-angle curves are preprocessed and tightening quality is classified to generate sample curves and sample types. Multidimensional features are extracted from the sample curves using the sliding window method to generate multidimensional features. The multidimensional features of the sample curves for each sample type are summarized and analyzed, and the corresponding feature threshold intervals are generated using statistics. The feature threshold range for each sample type is configured to be monitored in real time by the monitoring system, and the real-time tightening quality is obtained during the online monitoring phase.
2. The method for real-time monitoring of bolt tightening quality according to claim 1, characterized in that, The data preprocessing includes smoothing, denoising, normalizing, and interval filtering of the torque-angle curve.
3. The method for real-time monitoring of bolt tightening quality according to claim 1, characterized in that, The tightening quality classification includes classifying the torque-angle curve into normal tightening and abnormal tightening based on process knowledge. Abnormal tightening includes sticking and slipping, slight sticking and slipping, jamming, slight jamming, repeated tightening, excessive tightening torque, and large slope fluctuation.
4. The method for real-time monitoring of bolt tightening quality according to claim 1, characterized in that, The multidimensional feature extraction of the sample curve based on the sliding window method includes extracting multidimensional feature parameters of the sample curve within each sliding window. The multidimensional feature parameters include peak torque, final tightening angle, interval slope, root mean square error, number of extreme values, and abrupt change point data.
5. The method for real-time monitoring of bolt tightening quality according to claim 1, characterized in that, The process of summarizing and analyzing the multidimensional features of the sample curves for each sample type, and generating corresponding feature threshold ranges using statistics, includes: For sample curves of the same sample type, calculate the mean and standard deviation of each feature within each sliding window; Based on the mean and the standard deviation, the feature threshold range of each dimension of the feature within each sliding window under the sample type is determined according to the 3σ principle.
6. The method for real-time monitoring of bolt tightening quality according to claim 1, characterized in that, The step of configuring the feature threshold range for each sample type to the monitoring system for real-time monitoring includes: During the online monitoring phase, the torque and angle during the bolt tightening process are obtained from the tightening machine to generate a torque-angle curve; The torque-angle curve is preprocessed, and multidimensional features are extracted from the preprocessed sample curve using the sliding window method to generate multidimensional features. The multidimensional features are matched with the feature threshold intervals of each sample type, and the sample type that matches the feature threshold interval the most times is taken as the real-time tightening quality.
7. A real-time monitoring device for bolt tightening quality, characterized in that, include: The data acquisition module is configured to acquire the torque and angle during the bolt tightening process from the tightening machine during the offline training phase, and generate a torque-angle curve. The data processing module is configured to perform data preprocessing and tightening quality classification on the torque-angle curve, and generate sample curves and sample types; The feature extraction module is configured to perform multidimensional feature extraction on the sample curve based on the sliding window method to generate multidimensional features; The analysis and statistics module is configured to perform a summary analysis of the multidimensional features of the sample curves for each sample type and generate corresponding feature threshold intervals using statistics. The configuration application module is configured to configure the feature threshold range of each sample type to the monitoring system for real-time monitoring, and to obtain the real-time tightening quality during the online monitoring phase.
8. An electronic device, characterized in that, Including processor and storage media; The storage medium is used to store instructions; The processor is configured to operate according to the instructions to perform the steps of the method according to any one of claims 1-6.
9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the program implements the steps of the method according to any one of claims 1-6.
10. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the method described in any one of claims 1-6.