A bolt inspection method and device based on data fusion

Abstract

The present invention provides a bolt inspection method and device based on data fusion, comprising the following steps: S1, constructing a torque preload curve and a preload sound time difference curve; S2, calculating the preload conversion coefficient after installation; S3, constructing a preload attenuation prediction curve, calculating the predicted re-tightening time, and re-tightening after the predicted re-tightening time is reached; S4, constructing a preload conversion coefficient attenuation prediction curve; S5, calculating the predicted scrap date according to the preload conversion coefficient attenuation prediction curve, and replacing the bolt after the predicted scrap date is reached. Beneficial effects of the present invention: by calculating the predicted re-tightening time and the predicted scrap time and scrap time, the inspection time interval is reasonably planned, the work efficiency of the inspection is improved, the torque preload curve and the preload sound time difference curve of the undone bolts are used to calculate the preload and preload conversion coefficient in the subsequent inspection and re-tightening process, the detection accuracy is ensured, and the equipment operation safety is improved.

Description

Technical Field

[0001] The invention belongs to the technical field of bolt inspection data processing, and in particular relates to a bolt inspection method and device based on data fusion. Background Art

[0002] There are several methods for inspecting bolts: First, the inspection method of bolts mainly relies on mechanical torque tools, such as torque wrenches to check the torque value of the bolts. However, the torque value can only be used as an indirect indicator of the bolt preload and cannot accurately reflect the actual connection status of the bolt. The relationship between torque and preload is not simple and is affected by many factors, such as friction coefficient, bolt surface condition, thread characteristics, etc. Therefore, relying solely on torque wrench inspection cannot accurately evaluate the actual preload of the bolt, and the error is large, which may cause potential risks in engineering fields with high safety requirements (such as bridges, heavy machinery, etc.); second, ultrasonic intelligent bolt monitoring technology is gradually applied to bolt preload monitoring. The existing ultrasonic intelligent bolt system adopts a multi-channel architecture. By configuring independent sensor nodes on each bolt, the status of multiple bolts is monitored in real time. This technology can reflect the change of bolt preload by measuring the acoustic time difference of ultrasonic propagation. However, although the existing multi-channel ultrasonic intelligent bolt monitoring system has made certain breakthroughs in accuracy, it is difficult to popularize this system in large-scale applications due to its high equipment cost, complex installation, and the need to equip each bolt with an independent sensor node. In addition, during the bolt inspection process, it is necessary to determine whether to re-tighten according to the bolt preload situation. The current inspection interval is obtained based on experience, and the bolt re-tightening time cannot be predicted, which often leads to frequent inspections and low work efficiency, and premature re-tightening shortens the service life of the bolt. Summary of the invention

[0003] In view of this, the present invention aims to propose a bolt inspection method and device based on data fusion, in order to solve at least one of the above-mentioned technical problems.

[0004] To achieve the above object, the technical solution of the present invention is achieved as follows:

[0005] A first aspect of the present invention provides a bolt inspection method based on data fusion, comprising the following steps:

[0006] S1. Collect torque data, preload data, and ultrasonic time difference data of uninstalled bolts to construct torque preload curve and preload acoustic time difference curve;

[0007] S2, collecting the torque and ultrasonic time difference data after installation to calculate the preload conversion coefficient after installation;

[0008] S3, collecting the time of multiple inspections and ultrasonic time difference data, constructing a preload attenuation prediction curve, calculating the predicted re-tightening time, and re-tightening after the predicted re-tightening time is reached;

[0009] S4. Constructing a preload conversion coefficient attenuation prediction curve based on the time and torque data of multiple re-tightening and the preload conversion coefficient after installation;

[0010] S5. The predicted scrapping date of the bolt is calculated based on the attenuation prediction curve of the preload conversion coefficient, and the bolt is replaced after the predicted scrapping time is reached.

[0011] Furthermore, in S3, when the pre-tightening force decays to the pre-tightening force threshold, the dispersion of the bolts in the same group is judged before re-tightening, and the specific steps are as follows:

[0012] S31, calculating the dispersion of the preload force of the same group of bolts through the coefficient of variation, and calculating the dispersion of the preload force attenuation prediction curve of the same group of bolts through the coefficient of variation;

[0013] S32. If the coefficient of variation of the preload force is greater than or equal to the threshold value, and the coefficient of variation of the preload force attenuation prediction curve is greater than or equal to the threshold value, replace the bolt with the largest slope of the preload force attenuation prediction curve;

[0014] S33. Otherwise, re-tighten according to the shortest predicted re-tightening time of the bolts in the same group.

[0015] Furthermore, in S5, after the predicted scrapping time is reached, before the bolts of the same group are replaced, the dispersion of the bolts of the same group is judged, and the specific steps are as follows:

[0016] S51. Calculate the dispersion of the preload conversion coefficient attenuation prediction curve of the same group of bolts through the coefficient of variation;

[0017] S52: If the coefficient of variation of the preload conversion coefficient attenuation prediction curve is greater than or equal to the threshold, replace the bolt with the largest slope of the preload conversion coefficient attenuation prediction curve;

[0018] S53: If the coefficient of variation of the preload conversion coefficient attenuation prediction curve is less than a threshold value, the bolts in the same group are replaced after the shortest predicted scrap time among the bolts in the same group is reached.

[0019] Furthermore, the steps of constructing the preload force attenuation prediction curve are as follows:

[0020] A1. Collect all preload force data of multiple bolts obtained after installation and arrange them according to the test time to form a two-dimensional array of preload force;

[0021] A2. Fitting the preload force two-dimensional array to obtain a preload force prediction function;

[0022] A3. After re-tightening, the slope of the pre-tightening force conversion coefficient attenuation prediction curve is fitted by the pre-tightening force prediction function of the previous inspection cycle to obtain the first pre-tightening force prediction function and calculate the predicted re-tightening time;

[0023] A4. After re-tightening, repeat steps A1 and A2 to obtain a second preload prediction function as the preload prediction function of this inspection cycle.

[0024] Furthermore, the process of constructing the preload acoustic time difference curve according to the preload force data and the ultrasonic time difference data in S1 is as follows:

[0025] The simulated tightening method is used to construct the preload acoustic time difference curve. The axial force loading is realized by the bolt tightening process to ensure that the stress state of the bolt is consistent with the actual state. By obtaining multiple sets of data points of preload data and ultrasonic time difference, the polynomial fitting function is constructed and calculated:

[0026] ;

[0027] In the formula, is the ultrasonic time difference, is the preload force, Represents a first-order or n-order polynomial function;

[0028] The torque preload curve and the preload sound time difference curve are constructed simultaneously. In the process of simulated tightening, multiple sets of torque and preload data points are obtained to construct and calculate the polynomial fitting function.

[0029] ;

[0030] In the formula, is the torque, is the preload force, Represents a first-order or n-th-order polynomial function.

[0031] A second aspect of the present invention provides a bolt spot inspection device based on data fusion, comprising:

[0032] A collection module configured to collect torque data, ultrasonic time difference data, preload data, and bolt information data of the bolt;

[0033] The information processing module is configured to perform the following steps:

[0034] B1. Calculate the torque preload curve and preload sound time difference curve of the unmounted bolt according to the torque data, ultrasonic time difference data and preload data of the unmounted bolt, and match and store them with the bolt information data;

[0035] B2. Calculate the preload conversion coefficient after installation based on the installation torque and ultrasonic time difference data collected by the acquisition module;

[0036] B3. According to the acquisition module, the time of multiple inspections and ultrasonic time difference data are collected to construct a preload attenuation prediction curve, calculate the predicted re-tightening time, and re-tighten after the predicted re-tightening time is reached;

[0037] B4. The acquisition module collects the time and torque data of multiple re-tightening, and constructs a preload conversion coefficient attenuation prediction curve in combination with the preload conversion coefficient after installation;

[0038] B5. Calculate the predicted scrap date of the bolt based on the attenuation prediction curve of the preload conversion coefficient, and replace the bolt when the predicted scrap date is reached.

[0039] Furthermore, an ultrasonic RFID integrated patch is provided in each bolt;

[0040] The acquisition module includes a torque wrench, an ultrasonic detection head, and an RFID reader;

[0041] The torque wrench is used to collect torque data for bolt installation and re-tightening;

[0042] The ultrasonic detection head is used to collect ultrasonic time difference data of the bolt;

[0043] The RFID reader is used to collect bolt information data;

[0044] The information processing module receives and collects bolt information data, matches torque data and ultrasonic time difference data of the uninstalled bolts and stores them in the storage;

[0045] The information processing module matches the calculated preload force attenuation prediction curve and preload force conversion coefficient attenuation prediction curve with the bolt information data and stores them in the storage.

[0046] A third aspect of the present invention provides an electronic device, comprising a processor and a memory that is communicatively connected to the processor and is used to store instructions executable by the processor, wherein the processor is used to execute the method described in the first aspect above.

[0047] The fourth aspect of the present invention provides a server, comprising at least one processor, and a memory communicatively connected to the processor, the memory storing instructions executable by the at least one processor, the instructions being executed by the processor so that the at least one processor executes the method described in the first aspect.

[0048] A fifth aspect of the present invention provides a computer-readable storage medium storing a computer program, which implements the method described in the first aspect when executed by a processor.

[0049] Compared with the prior art, the bolt inspection method and device based on data fusion described in the present invention has the following beneficial effects:

[0050] The bolt inspection method based on data fusion described in the present invention calculates the predicted re-tightening time and the predicted scrapping time, reasonably plans the inspection time interval, improves the inspection work efficiency, calculates the pre-tightening force and pre-tightening force conversion coefficient in the subsequent inspection and re-tightening process through the torque pre-tightening force curve and the pre-tightening force sound time difference curve of the uninstalled bolt, ensures the detection accuracy, and improves the equipment operation safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] The accompanying drawings constituting a part of the present invention are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the accompanying drawings:

[0052] Figure 1 The figure is a flow chart of the spot inspection method described in an embodiment of the present invention. DETAILED DESCRIPTION

[0053] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments may be combined with each other.

[0054] In the description of the present invention, it is to be understood that the terms "first", "second", etc. are used for descriptive purposes only and are not to be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first", "second", etc. may explicitly or implicitly include one or more of the feature. In the description of the present invention, unless otherwise specified, "plurality" means two or more.

[0055] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood by specific circumstances.

[0056] The present invention will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments.

[0057] Embodiment 1:

[0058] like Figure 1 As shown, a bolt inspection method based on data fusion includes the following steps:

[0059] S1. Collect torque data, preload data, and ultrasonic time difference data of uninstalled bolts to construct torque preload curve and preload acoustic time difference curve;

[0060] The process of constructing the preload acoustic time difference curve according to the preload data and ultrasonic time difference data in S1 is as follows:

[0061] The principle of ultrasonic bolt axial force test is: ultrasonic It is emitted through one end of the bolt at the moment and reflected at the other end. The time difference is received back at the moment. With bolt length There is a theoretical linear relationship between

[0062] ;

[0063] In the formula, is the speed of ultrasound in the material.

[0064] From elastic mechanics, we know that within the elastic range of the material, there is also a linear relationship between the bolt length and the axial force of the bolt, that is,

[0065] ;

[0066] In the formula, is Young's elastic modulus of the bolt material, is the length change.

[0067] Therefore, within the elastic range of the material, the axial force of the bolt and the acoustic time difference between ultrasonic emission and reception also satisfy a linear relationship, that is,

[0068] ;

[0069] In the formula, Theoretically, it is only necessary to calibrate the bolt by stretching it before it is put into service. This coefficient can be used to detect the preload of the bolt in subsequent service, i.e., "preload-sound time difference". Based on this principle, the calibration of smart bolts on the market uses a tensile testing machine to apply axial quasi-static tensile loads of different amplitudes to the smart bolts. , and establish the relationship between the tensile load and the acoustic time difference through linear fitting or polynomial fitting, that is, The curve is used as the judgment data for calculating the preload force of the subsequent bolts during service.

[0070] But in fact, the propagation speed of sound is not fixed, and the stress state inside the bolt will affect the propagation speed of ultrasonic waves. The tensile method is used for calibration, and the uniaxial equivalent stress on the thread interface of the bolt is for:

[0071] ;

[0072] In the formula, is the equivalent stress diameter of the bolt.

[0073] In actual service, the bolts are tightened and loaded, which results in the axial force Loading value with tensile testing machine The equivalent stress inside the bolt is different. When tightened, the stress on the bolt is tensile stress. , tangential stress Biaxial stress, according to the fourth strength theory, the equivalent stress of the bolt for ;

[0074] In the formula, Obviously,

[0075] ;

[0076] The difference in internal stress of the material leads to the use of tensile method During the curve calibration and the actual bolt tightening process, the sound velocity of the ultrasonic wave in the material The most direct consequence is that the preload force calculated later using the calibration curve will have a large error.

[0077] The simulated tightening method is used to construct the preload acoustic time difference curve. The axial force loading is realized by the bolt tightening process to ensure that the stress state of the bolt is consistent with the actual state. By obtaining multiple sets of data points of preload data and ultrasonic time difference, the polynomial fitting function is constructed and calculated:

[0078] ;

[0079] In the formula, is the ultrasonic time difference, is the preload force, Represents a first-order or n-order polynomial function. The preload acoustic time difference curve can be used to calculate the current preload through ultrasonic time difference during the inspection process, and the test result is more accurate.

[0080] The torque preload curve can be used as a basis for determining the degradation of bolt size, performance and thread quality after service. Theoretically, the torque-preload conversion of the bolt satisfies ;

[0081] That is, the bolt preload conversion factor (i.e. the slope of the "torque-preload" curve) is

[0082] ;

[0083] In the formula, is the torque, is the preload force, is the thread pitch diameter, is the thread angle (30° for metric threads), is the thread friction coefficient, is the thread pitch, is the equivalent friction diameter of the bolt head, is the friction coefficient of the bolt head. Since the parameters in the formula are all constants, for a bolt, its preload conversion coefficient is In theory, it is a constant value. But in reality, as the service life of the bolt increases, the various interfaces of the bolt will wear out, and the friction coefficient of the contact surface will gradually increase, resulting in The value increases, and the torque-preload conversion coefficient increases. The most direct consequence is that the old bolt needs a larger torque to be tightened to the target preload, or the shear stress inside the bolt is too large before the target preload is reached, causing the bolt to break.

[0084] The "torque-preload" curve is calibrated during the factory inspection, and the preload conversion coefficient is adjusted during the subsequent tightening and re-tightening process. By conducting statistical analysis on the attenuation of the bolts, the performance attenuation of the bolts can be obtained, and the time point for actively scrapping and replacing the bolts can be predicted, which greatly improves the reliability of the connection of high-end equipment.

[0085] The torque preload curve and the preload sound time difference curve are constructed simultaneously. In the process of simulated tightening, multiple sets of torque and preload data points are obtained to construct and calculate the polynomial fitting function.

[0086] ;

[0087] In the formula, is the torque, is the preload force, Represents a first-order or n-th-order polynomial function.

[0088] S2, collecting the torque and ultrasonic time difference data after installation to calculate the preload conversion coefficient after installation;

[0089] The preload force at the corresponding moment needs to be obtained in the three stages of installation, inspection and re-tightening. The acquisition method is to first communicate with the RFID chip implanted in the bolt head, obtain the bolt ID, and obtain the corresponding bolt ID in the database. Then, the piezoelectric ceramic piece at the end of the bolt is energized to obtain the acoustic time difference of the bolt during installation, inspection, and re-tightening. and ;

[0090] Finally, the bolt preload force of this link can be obtained by substituting it into the calibration function, that is,

[0091] ;

[0092] ;

[0093] ;

[0094] The torque at the corresponding moment needs to be obtained during installation and re-tightening. The torque can be directly obtained through the torque sensor inside the torque wrench. Torque value.

[0095] In the whole life cycle of a bolt, there are usually multiple installation, inspection, and re-tightening links. The subscript i is used to represent a link that occurs i times repeatedly, for example , .

[0096] In the two links of installation and re-tightening, it is necessary to obtain the preload conversion coefficient at the corresponding moment. According to the formula, the calculation method of the preload conversion coefficient is:

[0097] ;

[0098] ;

[0099] S3, collecting the time and ultrasonic time difference data of multiple inspections, constructing a preload attenuation prediction curve, calculating the predicted re-tightening time, and re-tightening after the predicted re-tightening time is reached;

[0100] By calculating and predicting the re-tightening time, we can reasonably plan the inspection interval time. While ensuring safety, we can increase the inspection interval time to reduce workload and improve work efficiency.

[0101] The preload force is calculated by ultrasonic time difference according to the preload force acoustic time difference curve.

[0102] After the bolt is installed and has undergone multiple inspections, sufficient preload data can be obtained for a single bolt to support preload attenuation analysis. Arrange all preload data obtained since the bolt was installed by test time to form the following two-dimensional array:

[0103] ;

[0104] In the formula, Indicates the interval between the time when the operation is performed and the time when the bolt is installed. .

[0105] By performing linear fitting or polynomial fitting on the array in the above formula, the corresponding bolt preload attenuation prediction curve can be obtained, that is,

[0106] ;

[0107] In the formula, represents a first-order or n-order polynomial function, represents any time after the installation time, Indicated in The residual preload of the bolt at time, Indicates the date corresponding to the installation phase. By entering different time , the residual preload of the bolt at any time can be predicted .

[0108] The preload loosening curve will experience a "fast-slow-fast" process. Within 24 hours of installation, the preload will usually decay rapidly by 10% to 20%, and then turn into a continuous slow decay. When the preload is less than 50%, it gradually enters the rapid loosening stage, and the decay speed gradually accelerates, and finally the bolt loosens. Therefore, the re-tightening time is set to the time when the preload decays to In the present invention, the re-tightening time is set when the pre-tightening force decays to hour, is the target preload, Substitution In the result:

[0109] ;

[0110] The bolt preload can be reduced to date , which is the date when the next tightening is required. express The inverse function of .

[0111] The steps to construct the preload attenuation prediction curve are as follows:

[0112] A1. Collect all preload force data of multiple bolts obtained after installation and arrange them according to the test time to form a two-dimensional array of preload force;

[0113] A2. Fitting the preload force two-dimensional array to obtain a preload force prediction function;

[0114] A3. After re-tightening, the pre-tightening force prediction function of the previous inspection cycle is fitted with the slope of the pre-tightening force conversion coefficient attenuation prediction curve to obtain the first pre-tightening force prediction function and calculate the predicted re-tightening time; the slope of the pre-tightening force conversion coefficient attenuation prediction curve can be obtained by fitting the slope of the pre-tightening force conversion coefficient attenuation prediction curve.

[0115] A4. After re-tightening, repeat steps A1 and A2 to obtain a second preload prediction function as the preload prediction function of this inspection cycle.

[0116] If not enough inspection data is obtained at the end of re-tightening to form the preload prediction function of this re-tightening cycle, the inspection frequency can be planned by fitting the preload prediction function of the previous inspection cycle through the slope of the preload conversion coefficient attenuation prediction curve to prevent low work efficiency caused by blind inspection.

[0117] In S3, when the preload force decays to the preload force threshold, the dispersion of the bolts in the same group is judged before re-tightening. The specific steps are as follows:

[0118] S31, calculating the dispersion of the preload force of the same group of bolts through the coefficient of variation, and calculating the dispersion of the preload force attenuation prediction curve of the same group of bolts through the coefficient of variation;

[0119] Preload dispersion and torque dispersion analysis are required in different links. Here, the bolt preload dispersion analysis of the same bolt group in the installation link is introduced as an example. The calculation process of preload dispersion and torque dispersion is the same. Here, only preload dispersion is taken as an example.

[0120] After multiple bolts in the same bolt group are installed, sufficient preload data can be obtained to support preload dispersion analysis. The installation preload data of n bolts are organized into the following one-dimensional array;

[0121] ;

[0122] Considering that enterprises need to conduct comparative analysis on the discreteness of bolt groups in different parts, different environments, and different service conditions, it is necessary to select a normalized discreteness measurement index as the judgment standard. Therefore, the coefficient of variation is selected. It is more suitable as the analysis basis of preload dispersion. For the above array, the formula is as follows:

[0123] ;

[0124] In the formula, is the population standard deviation of the array, is the mean of the array, calculated as follows:

[0125] ;

[0126] ;

[0127] In addition, the conventional bolt group preload dispersion analysis method also includes the preload upper deviation Deviation from preload , the calculation method is:

[0128] ;

[0129] ;

[0130] The analysis principle of preload force in other links is the same.

[0131] S32: If the coefficient of variation of the preload force is greater than or equal to a threshold value (the threshold value range is 0.05-0.2, preferably 0.1), and the coefficient of variation of the preload force attenuation prediction curve is greater than or equal to a threshold value (the threshold value range is 0.05-0.2, preferably 0.1), then replace the bolt with the largest slope of the preload force attenuation prediction curve;

[0132] S33. Otherwise, re-tighten according to the shortest predicted re-tightening time of the bolts in the same group.

[0133] The coefficient of variation is used to determine the consistency of the preload attenuation process of the same group of bolts. Replacing individual bolts that have attenuated too quickly can reduce the re-tightening cycle of the entire group of bolts. By predicting the re-tightening time of the same group of bolts, the number of inspections can be appropriately reduced, which facilitates batch management of bolts and improves inspection efficiency.

[0134] S4. Constructing a preload conversion coefficient attenuation prediction curve based on the time and torque data of multiple re-tightening and the preload conversion coefficient after installation;

[0135] The torque data is collected during re-tightening, and the preload force is calculated from the torque data through the torque preload force curve. Then, the preload force conversion coefficient is calculated based on the preload force and torque.

[0136] After the bolt is installed and has been tightened several times, sufficient preload conversion coefficient data can be obtained for a single bolt to support performance degradation analysis. Arrange all preload conversion coefficient data obtained since the bolt was installed by test time to form the following two-dimensional array

[0137] ;

[0138] In the formula, Indicates the interval between the time when the operation is performed and the time when the bolt is installed. .

[0139] By performing linear fitting or polynomial fitting on the array in the above formula, the preload attenuation prediction curve of the corresponding bolt can be obtained;

[0140] ;

[0141] In the formula, represents a first-order or n-order polynomial function, represents any time after the installation time, Indicated in The bolt preload conversion coefficient value at time, Indicates the date corresponding to the installation phase. By entering different time , the bolt preload conversion coefficient at any time can be predicted The increase of bolt preload conversion coefficient represents the attenuation of bolt performance, and the two are negatively correlated.

[0142] The increase in the bolt preload conversion coefficient indicates that the bolt needs to be scrapped. According to the tightening connection theory, when the bolt cannot be tightened to the factory preset target preload, it should be regarded as scrapped.

[0143] The bolt preload conversion coefficient is mainly related to the thread friction coefficient. As the bolt is in service, the thread wear gradually increases, the thread friction coefficient increases, and the bolt preload conversion coefficient increases. At this time, the torque required to re-tighten the bolt to the target preload will increase. When the old bolt reaches the same target preload, the increase in friction coefficient causes the shear stress to increase. When the equivalent stress exceeds the material yield limit, the bolt is twisted off.

[0144] The target preload of the new bolt is set to the yield limit load. (Usually 80%).

[0145] ;

[0146] in is the material yield limit. Assuming that the bolt preload conversion factor rises to , under the same target preload, the torque required by the bolt also rises to According to the theory of elasticity, the shear stress of the old bolt is Also rose to .

[0147] ;

[0148] The critical condition for the old bolt to be just twisted off is But in fact, a certain safety margin is needed to ensure the safety of the connection structure.

[0149] In the formula, is the utilization factor of the old bolts (limit is 95%).

[0150] Combining, we can know that;

[0151] ;

[0152] Substitute it into:

[0153] ;

[0154] Since the above formula lacks necessary solution parameters, the exact solution cannot be calculated, but the range of the solution can be determined as:

[0155] ;

[0156] In order to ensure the safety of the connection structure, , that is, the increase ratio of the bolt preload conversion coefficient shall not exceed .Will Substitute into the formula get;

[0157] ;

[0158] This is the estimated scrap date of the current bolt. express The inverse function of .

[0159] S5. The predicted scrapping date of the bolt is calculated based on the attenuation prediction curve of the preload conversion coefficient, and the bolt is replaced after the predicted scrapping time is reached.

[0160] In S5, after reaching the predicted scrapping time, before replacing the same group of bolts, the dispersion of the same group of bolts is judged. The specific steps are as follows:

[0161] S51. Calculate the dispersion of the preload conversion coefficient attenuation prediction curve of the same group of bolts through the coefficient of variation;

[0162] S52, if the coefficient of variation of the preload conversion coefficient attenuation prediction curve is greater than or equal to a threshold value (the threshold value range is 0.05-0.2, preferably 0.1), replace the bolt with the largest slope of the preload conversion coefficient attenuation prediction curve;

[0163] S53. If the coefficient of variation of the preload conversion coefficient attenuation prediction curve is less than a threshold value (the threshold value range is 0.05-0.2, preferably 0.1), the bolts in the same group are replaced after the shortest predicted scrap time among the bolts in the same group is reached.

[0164] The consistency of the attenuation process of the preload conversion coefficient of the same group of bolts is determined by the coefficient of variation. Replacing individual aged bolts can increase the service life of the entire group of bolts. By predicting the service life of the same group, batch management of bolts is facilitated to improve inspection efficiency.

[0165] By calculating the predicted re-tightening time and predicted scrap time, the inspection time interval can be reasonably planned to improve the inspection work efficiency.

[0166] By calculating the torque preload curve of the undocked bolts and the preload sound time difference curve, the preload and preload conversion coefficient in the subsequent spot inspection and re-tightening process are ensured to ensure the detection accuracy and improve the safety of equipment operation.

[0167] A bolt inspection device based on data fusion, comprising:

[0168] A collection module configured to collect torque data, ultrasonic time difference data, preload data, and bolt information data of the bolt;

[0169] The information processing module is configured to perform the following steps:

[0170] B1. Calculate the torque preload curve and preload sound time difference curve of the unmounted bolt according to the torque data, ultrasonic time difference data and preload data of the unmounted bolt, and match and store them with the bolt information data;

[0171] B2. Calculate the preload conversion coefficient after installation based on the installation torque and ultrasonic time difference data collected by the acquisition module;

[0172] B3. According to the acquisition module, the time of multiple inspections and ultrasonic time difference data are collected to construct a preload attenuation prediction curve, calculate the predicted re-tightening time, and re-tighten after the predicted re-tightening time is reached;

[0173] B4. By collecting the time and torque data of multiple re-tightening according to the acquisition module, the preload conversion coefficient attenuation prediction curve is constructed in combination with the preload conversion coefficient after installation;

[0174] B5. Calculate the predicted scrap date of the bolt based on the attenuation prediction curve of the preload conversion coefficient, and replace the bolt when the predicted scrap date is reached.

[0175] Each bolt has an ultrasonic RFID integrated patch;

[0176] The acquisition module includes a torque wrench, an ultrasonic detection head, and an RFID reader;

[0177] The torque wrench is used to collect torque data for bolt installation and re-tightening;

[0178] The ultrasonic detection head is used to collect ultrasonic time difference data of the bolt;

[0179] The RFID reader is used to collect bolt information data;

[0180] The information processing module receives and collects bolt information data, matches the torque data of the uninstalled bolts, and stores the ultrasonic time difference data in the storage;

[0181] The information processing module matches the calculated preload force attenuation prediction curve and preload force conversion coefficient attenuation prediction curve with the bolt information data and stores them in the storage.

[0182] Beneficial effects:

[0183] The bolt identity recognition function is added to improve the detection accuracy. The existing ultrasonic intelligent bolt monitoring system must start its monitoring from the assembly stage. This is because the monitoring system needs to determine the initial length and real-time length of the bolt in order to calculate the preload force. If the existing ultrasonic intelligent bolt monitoring system is directly used for bolt inspection, the initial length of each bolt is different, which will lead to a huge test error, or even the inability to obtain the preload force of the bolt. This inspection system has added a bolt identity recognition function. By identifying the identity of the bolt, the inspection data is compared with the initial length data, and the "acoustic time difference-axial force" data is calibrated based on batch products, so that the ultrasonic testing system can be applied to the bolt inspection stage and greatly improve the detection accuracy.

[0184] Reduce detection costs. Existing multi-channel ultrasonic intelligent bolt monitoring systems usually require an independent sensor module for each bolt and real-time monitoring through a data acquisition network, which makes the system cost high and the equipment installation complicated. In contrast, the present invention adopts a single-channel ultrasonic intelligent detection system, which can reduce equipment costs while simplifying the installation and maintenance process. Combined with the RFID chip to store the initial data and acoustic time difference calibration information of the bolt, the present invention no longer requires a multi-channel system and multiple probes, greatly reducing the overall cost of the system. The manufacturing cost of a single bolt only increases the cost of the piezoelectric ceramic sheet + RFID chip.

[0185] The present invention integrates ultrasonic detection and RFID data reading into a simple inspection process, so that users can complete the bolt preload measurement with a single device. By reading the data in the RFID chip, the system automatically identifies the basic information of the bolt and calculates the current preload of the bolt through ultrasound, which is easy to operate.

[0186] Since the present invention can significantly reduce system costs and simplify the operation process, the technology has broad promotion prospects in small and medium-sized industrial equipment, regular inspection projects and various engineering applications. Compared with the existing high-cost, multi-channel real-time monitoring system, the present invention provides a cost-effective solution that can be applied in more scenarios, such as bolt inspection of bridges, tunnels, wind power, heavy machinery and equipment, etc.

[0187] Embodiment 2:

[0188] An electronic device comprises a processor and a memory which is communicatively connected to the processor and is used to store instructions executable by the processor, wherein the processor is used to execute the method described in the first embodiment above.

[0189] Embodiment three:

[0190] A server includes at least one processor and a memory communicatively connected to the processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the processor so that the at least one processor executes the method described in Example 1.

[0191] Embodiment 4:

[0192] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method described in embodiment 1 is implemented.

[0193] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. These modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope of the claims and specification of the present invention.

[0194] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A bolt inspection method based on data fusion, characterized in that: The following steps are involved: S1. Collect torque data, preload data, and ultrasonic time difference data of uninstalled bolts to construct torque preload curve and preload acoustic time difference curve; S2, collecting the torque and ultrasonic time difference data after installation to calculate the preload conversion coefficient after installation; S3, collecting the time of multiple inspections and ultrasonic time difference data, constructing a preload attenuation prediction curve, calculating the predicted re-tightening time, and re-tightening after the predicted re-tightening time is reached; S4. Constructing a preload conversion coefficient attenuation prediction curve based on the time and torque data of multiple re-tightening and the preload conversion coefficient after installation; S5. The predicted scrapping date of the bolt is calculated based on the attenuation prediction curve of the preload conversion coefficient, and the bolt is replaced after the predicted scrapping time is reached.

2. A bolt inspection method based on data fusion according to claim 1, characterized in that: In S3, when the preload force decays to the preload force threshold, the dispersion of the bolts in the same group is judged before re-tightening. The specific steps are as follows: S31, calculating the dispersion of the preload force of the same group of bolts through the coefficient of variation, and calculating the dispersion of the preload force attenuation prediction curve of the same group of bolts through the coefficient of variation; S32. If the coefficient of variation of the preload force is greater than or equal to the threshold value, and the coefficient of variation of the preload force attenuation prediction curve is greater than or equal to the threshold value, replace the bolt with the largest slope of the preload force attenuation prediction curve; S33. Otherwise, re-tighten according to the shortest predicted re-tightening time of the bolts in the same group.

3. The bolt inspection method based on data fusion according to claim 1 is characterized in that: In S5, after the predicted scrapping time is reached, before the bolts of the same group are replaced, the dispersion of the bolts of the same group is judged, and the specific steps are as follows: S51. Calculate the dispersion of the preload conversion coefficient attenuation prediction curve of the same group of bolts through the coefficient of variation; S52, if the coefficient of variation of the preload conversion coefficient attenuation prediction curve is greater than or equal to the threshold, replace the bolt with the largest slope of the preload conversion coefficient attenuation prediction curve; S53: If the coefficient of variation of the preload conversion coefficient attenuation prediction curve is less than a threshold value, the bolts in the same group are replaced after the shortest predicted scrap time among the bolts in the same group is reached.

4. The bolt inspection method based on data fusion according to claim 1 is characterized in that: The steps to construct the preload attenuation prediction curve are as follows: A1. Collect all preload force data of multiple bolts obtained after installation and arrange them according to the test time to form a two-dimensional array of preload force; A2. Fitting the preload force two-dimensional array to obtain a preload force prediction function; A3. After re-tightening, the slope of the pre-tightening force conversion coefficient attenuation prediction curve is fitted by the pre-tightening force prediction function of the previous inspection cycle to obtain the first pre-tightening force prediction function and calculate the predicted re-tightening time; A4. After re-tightening, repeat steps A1 and A2 to obtain a second preload prediction function as the preload prediction function of this inspection cycle.

5. The bolt inspection method based on data fusion according to claim 1 is characterized in that: The process of constructing the preload acoustic time difference curve according to the preload data and the ultrasonic time difference data in S1 is as follows: The simulated tightening method is used to construct the preload acoustic time difference curve. The axial force loading is realized by the bolt tightening process to ensure that the stress state of the bolt is consistent with the actual state. By obtaining multiple sets of data points of preload data and ultrasonic time difference, the polynomial fitting function is constructed and calculated: ； In the formula, is the ultrasonic time difference, is the preload force, Represents a first-order or n-order polynomial function; The torque preload curve and the preload sound time difference curve are constructed simultaneously. In the process of simulated tightening, multiple sets of torque and preload data points are obtained to construct and calculate the polynomial fitting function. ； In the formula, is the torque, is the preload force, Represents a first-order or n-th-order polynomial function.

6. A bolt inspection device based on data fusion, characterized in that: include: A collection module configured to collect torque data, ultrasonic time difference data, preload data, and bolt information data of the bolt; The information processing module is configured to perform the following steps: B1. Calculate the torque preload curve and preload sound time difference curve of the unmounted bolt according to the torque data, ultrasonic time difference data and preload data of the unmounted bolt, and match and store them with the bolt information data; B2. Calculate the preload conversion coefficient after installation based on the installation torque and ultrasonic time difference data collected by the acquisition module; B3. According to the acquisition module, the time of multiple inspections and ultrasonic time difference data are collected to construct a preload attenuation prediction curve, calculate the predicted re-tightening time, and re-tighten after the predicted re-tightening time is reached; B4. The acquisition module collects the time and torque data of multiple re-tightening, and constructs a preload conversion coefficient attenuation prediction curve in combination with the preload conversion coefficient after installation; B5. Calculate the predicted scrap date of the bolt based on the attenuation prediction curve of the preload conversion coefficient, and replace the bolt when the predicted scrap date is reached.

7. The bolt inspection device based on data fusion according to claim 6 is characterized in that: Each bolt has an ultrasonic RFID integrated patch; The acquisition module includes a torque wrench, an ultrasonic detection head, and an RFID reader; The torque wrench is used to collect torque data for bolt installation and re-tightening; The ultrasonic detection head is used to collect ultrasonic time difference data of the bolt; The RFID reader is used to collect bolt information data; The information processing module receives and collects bolt information data, matches torque data and ultrasonic time difference data of the uninstalled bolts and stores them in the storage; The information processing module matches the calculated preload force attenuation prediction curve and preload force conversion coefficient attenuation prediction curve with the bolt information data and stores them in the storage.

8. An electronic device, comprising a processor and a memory connected to the processor for storing instructions executable by the processor, characterized in that: The processor is used to execute the method described in any one of claims 1 to 4.

9. A server, characterized in that: The method comprises at least one processor and a memory in communication with the processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the processor so that the at least one processor executes the method according to any one of claims 1 to 4.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 4 is implemented.

Images