Steel structure bolt connection pre-tightening force monitoring and compensation system

Abstract

The application discloses a steel structure bolt connection pre-tightening force monitoring and compensation system and relates to the technical field of building steel structures, which comprises the following steps: a pre-tightening force monitoring module collects acoustic-elastic signals in real time through an ultrasonic sensor and calculates a pre-tightening force value; an attenuation characteristic identification module extracts attenuation characteristic parameters and determines an attenuation category; a hierarchical compensation decision module receives the attenuation category and a target pre-tightening force, generates two-stage control instructions, and controls a pre-tightening force loading unit to exert a tensile load on a bolt; a fine adjustment module outputs fine adjustment instructions based on the deviation between real-time pre-tightening force and the target value within a second holding duration; and an execution module executes corresponding instructions. The application solves the technical problems of difficult prediction of pre-tightening force attenuation, insufficient compensation accuracy and inability to monitor in real time by identifying the attenuation characteristics of individual bolts and implementing hierarchical dynamic compensation, and high-precision control of pre-tightening force is realized.

Description

Technical Field

[0001] This invention relates to the field of building steel structure technology, specifically to a system for monitoring and compensating the preload of bolted connections in steel structures. Background Technology

[0002] In the construction of high-strength bolted connections in steel structures, preload control is the core link to ensure node stiffness and structural safety. However, after the bolts are tightened, due to the embedment loss in the thread meshing area, material creep and temperature effects, the preload will rapidly decrease by 1%-5% in a short period of time (usually within 25 minutes), and the degree of decrease is significantly discrete due to the roughness of the contact surface, ambient temperature and bolt specifications. Currently, mainstream technologies mainly rely on torque or rotation methods for one-time tensioning, followed by ultrasonic testing for post-tensioning sampling. This approach has three inherent drawbacks: First, the natural decay of preload is difficult to predict, and existing technologies lack dynamic identification methods for individual bolt decay characteristics, leading to blind compensation strategies. Second, the compensation mechanism is imperfect; traditional fixed-ratio overtensioning (e.g., a uniform 12%) cannot adapt to different working conditions, and ultrasonic sensors are difficult to place in scenarios with short bolts (length-to-diameter ratio <4) and blind holes, resulting in real-time monitoring failure. Third, the temperature-stress coupling effect is not decoupled; a 10°C change in ambient temperature can cause a 5%-10% drift in preload, and existing systems lack temperature adaptive correction capabilities. In summary, current technology faces the pain points of "inaccurate measurement, imprecise compensation, and unreal-time monitoring": it is impossible to establish an accurate prediction model for pre-tightening force attenuation, it is difficult to implement graded dynamic compensation, and long-term monitoring is costly. This directly leads to the risk of loosening in key connection parts such as wind turbine towers and bridge nodes, resulting in high maintenance costs. To address the aforementioned issues, this invention provides a preload monitoring and compensation system for bolted connections in steel structures. The system extracts the individual bolt attenuation trends in real time using an attenuation feature identification module. A three-stage progressive strategy—"first-stage over-tensioning, second-stage trend correction, and third-stage closed-loop fine-tuning"—is implemented via a graded compensation decision module. Furthermore, temperature compensation and an adaptive step-size reduction algorithm are introduced to achieve the engineering goal of preload control accuracy of ±2% and convergence times ≤ 3. Summary of the Invention

[0003] To address the aforementioned technical issues, a preload monitoring and compensation system for steel structure bolt connections is provided. This technical solution solves the problems of the inability to establish an accurate preload attenuation prediction model, the difficulty in implementing graded dynamic compensation, and the high cost of long-term monitoring. This directly leads to the problem of loosening risks in key connection parts such as wind turbine towers and bridge nodes, resulting in high maintenance costs.

[0004] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A preload monitoring and compensation system for bolted connections in steel structures, including: The preload monitoring module uses an ultrasonic sensor placed on the end face of the bolt head to collect acoustic elastic signals in real time. Based on the pre-stored acoustic elastic signal-preload relationship, it calculates the preload value in real time and outputs preload timing data and real-time preload value. The attenuation feature identification module is used to receive the preload time series data, extract the attenuation feature parameters of the preload time series data, determine and output the attenuation category of the bolt; The graded compensation decision module receives the attenuation category, the externally input temperature environment, and the externally input target preload based on the target preload, and generates a first control command, a second control command, and corresponding first holding time and second holding time. The first control command is used to control the preload application mechanism to stretch the bolt to a first preload higher than the target preload and hold it for the first holding time. The second control command is used to control the preload application mechanism to apply an additional compensation amount determined based on the attenuation category after the first holding time ends and hold it for the second holding time. The fine-tuning module receives the real-time preload value during the second holding time and compares it with the target preload. When the deviation exceeds a preset threshold, it outputs a fine-tuning command. The fine-tuning command is used to control the preload application mechanism to increase or decrease the preload by a preset step size until the deviation falls into the target range. The execution module is used to receive the first control command, the second control command, or the fine-tuning command, and control the preload application mechanism to apply and maintain the preload.

[0005] Preferably, the preload monitoring module includes: An ultrasonic sensing unit is positioned on the end face of the bolt head for real-time acquisition of acoustoelastic signals. The calculation unit is used to calculate the acoustoelastic signal into a preload value in real time based on the pre-stored acoustoelastic signal-preload relationship; The timing recording unit is used to record the preload value in a time sequence, forming and outputting preload timing data; The real-time output unit is used to output the preload value calculated at the current moment as the real-time preload value.

[0006] Preferably, the preload monitoring module further includes: The storage unit is used to store the pre-stored acoustoelastic signal-preload relationship, which is obtained through a preload calibration test of the same batch of bolts in a temperature range of -40℃ to +80℃. The ultrasonic sensing unit is a magnetostrictive ultrasonic sensor, which is arranged on the end face of the bolt head and contacts the end face of the bolt through a coupling agent.

[0007] Preferably, the attenuation feature recognition module includes: The data receiving unit is used to receive and buffer the preload timing data; The feature extraction unit is used to extract attenuation feature parameters and attenuation trend from the preload time series data. The attenuation feature parameter is the rate of decrease of preload during the first holding time. The attenuation trend is extracted by linear fitting or exponential fitting. The category determination unit is used to determine and output the attenuation category of the bolt based on the attenuation characteristic parameters. When the preload decrease rate is greater than 1% / minute of the target preload, the fast attenuation type is output; otherwise, the slow attenuation type is output.

[0008] Preferably, the hierarchical compensation decision module includes: The first-level control unit is configured to generate the first control command, control the preload application mechanism to stretch the bolt to a first preload force higher than the target preload force and maintain it for the first holding time. The first-level control unit is also configured to monitor the attenuation trend in real time based on preload feedback during the first holding time. The second-level control unit is used to correct the additional compensation amount based on the attenuation trend, generate the second control command, and control the preload application mechanism to apply the corrected additional compensation amount and maintain the second holding time after the first holding time ends. The prediction compensation unit is used to calculate the pre-compensation amount and superimpose it on the second control command based on the actual residual preload at the end of the first holding time, the target preload, and the predicted attenuation amount during the second holding time. The predicted attenuation amount is obtained by extrapolation based on the attenuation trend.

[0009] Preferably, the first preload is 108%-112% of the target preload, and is determined according to the bolt specification: 110%-112% for M24 and below, 109%-111% for M27-M30, and 108%-110% for M33 and above; the first holding time is determined according to the ambient temperature: 8 minutes for below 0℃, 6 minutes for 0℃-30℃, and 5 minutes for above 30℃.

[0010] Preferably, the additional compensation amount is determined according to the following algorithm: The basic compensation is 3% of the target preload, and the attenuation type correction is +2% for fast attenuation type and -1% for slow attenuation type. The temperature correction is increased by 0.5% for every 10°C below the ambient temperature and decreased by 0.5% for every 30°C above the ambient temperature. The second holding time is dynamically adjusted based on the deviation between the actual residual preload and the target preload at the end of the first holding stage: 3 minutes when the deviation is less than 2%, 4 minutes when the deviation is 2%-4%, and 5 minutes when the deviation is greater than 4%.

[0011] Preferably, the fine-tuning module includes: The comparison unit is used to compare the real-time preload value with the target preload; The threshold determination unit is used to determine whether the deviation exceeds a preset threshold. The graded adjustment unit is used to output a first fine-tuning command when the deviation is first determined to exceed a preset threshold. The preset step size of the first fine-tuning command is determined according to the attenuation category. After a single adjustment, if the deviation still exceeds the preset threshold, a second fine-tuning command is output. The preset step size of the second fine-tuning command is 60% of that of the first fine-tuning command. If the deviation still exceeds the preset threshold for the third time, a third fine-tuning command is output and the bolt is marked as abnormal. The preset step size of the third fine-tuning command is 60% of that of the second fine-tuning command. The number of adjustments is limited and used to accumulate the number of adjustments. When the number of adjustments exceeds 3, the output of the adjustment command will stop and an alarm will be triggered.

[0012] Preferably, the preset threshold, preset step size, and target interval are determined according to the attenuation category classification: For the rapid decay type, the preset threshold is ±4%, the preset step size is 1% of the target preload, and the target range is ±3%; for the slow decay type, the preset threshold is ±2%, the preset step size is 0.5% of the target preload, and the target range is ±1.5%.

[0013] Preferably, the execution module includes: The instruction parsing unit is used to receive and parse the first control instruction, the second control instruction, or the fine-tuning instruction, and extract the target preload force, holding time, and adjustment step size. A preload loading unit is used to apply tensile load to bolts. The preload loading unit includes a hydraulic loading unit or an electric drive loading unit. The hydraulic loading unit is used for M30 and above specifications, and the electric drive loading unit is used for M27 and below specifications. The collaborative control unit is used to coordinate the loading sequence of each preload application mechanism during the synchronous tensioning of multiple bolts, so that the preload deviation of each bolt is within ±5%. The status recording unit is used to collect the oil pressure, displacement or motor current of the preload application mechanism in real time and output it to an external data recorder.

[0014] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention utilizes a magnetostrictive ultrasonic sensor positioned on the bolt head end face to collect acoustoelastic signals in real time. Combined with a pre-stored acoustoelastic-preload calibration relationship, it achieves real-time calculation and timing recording of the preload value. An attenuation feature identification module extracts the preload decrease rate within the first holding time and determines the attenuation type (fast / slow), enabling accurate identification and classification of individual bolt attenuation characteristics. A graded compensation decision module executes a multi-level compensation strategy of "first-level over-tensioning - second-level trend correction," extrapolating the compensation amount based on the attenuation trend to achieve feedforward compensation and dynamic correction of preload attenuation. Finally, a closed-loop fine-tuning with decreasing step size (60% increments) is implemented within the second holding time to achieve rapid convergence of preload deviation (convergence rate >95% within 3 iterations). Attached Figure Description

[0015] Figure 1 This is a system framework diagram of the present invention; Figure 2 This is a schematic diagram of the preload loading unit structure of the present invention. Detailed Implementation

[0016] The following description is intended to disclose the invention and enable those skilled in the art to implement it. The preferred embodiments described below are merely examples, and other obvious variations will occur to those skilled in the art.

[0017] Reference Figure 1 As shown, the preload monitoring and compensation system for steel structure bolt connections includes: The preload monitoring module uses an ultrasonic sensor placed on the end face of the bolt head to collect acoustic elastic signals in real time. Based on the pre-stored acoustic elastic signal-preload relationship, it calculates the preload value in real time and outputs preload timing data and real-time preload value. The preload monitoring module includes: An ultrasonic sensing unit is positioned on the end face of the bolt head for real-time acquisition of acoustoelastic signals. The calculation unit is used to calculate the acoustoelastic signal into a preload value in real time based on the pre-stored acoustoelastic signal-preload relationship; wherein, during real-time calculation, the calculation unit first obtains the acoustic time difference signal of the current bolt through the ultrasonic sensing unit. Simultaneously, the real-time temperature is obtained through a thermocouple (response time 0.1s, accuracy ±0.5℃) pre-embedded in the center of the bolt head. Substituting the pre-stored calibration formula, the current preload F is calculated using algebraic transformation as follows: in, The effective stress cross-sectional area of ​​the bolt (approximately 561 mm² for an M30 bolt). 2 ), This is the acoustic time difference signal of the current bolt. The acoustoelastic coefficient (unit: ns / MPa, typical value is about 1.2×10-2 ns / MPa) reflects the sensitivity of a material to changes in sound velocity under stress; It is a temperature coefficient (unit: ns / ℃, typical value is about -0.8 ns / ℃), reflecting the sound velocity drift and thermal expansion effect caused by temperature changes; This is the stress-temperature coupling correction factor (unit: ns / (MPa·℃)), used to correct for the effect of temperature on the acoustoelastic coefficient. The nonlinear effects are mitigated, with C being the zero-point drift constant. A dynamic temperature compensation term is introduced during the calculation. When the ambient temperature deviates from the calibration reference temperature (usually 20℃) by 10℃, the system automatically calls the corresponding temperature correction coefficient to compensate for the acoustic time difference. The compensated preload calculation value is stored in a time sequence by the timing recording unit to form preload timing data. Simultaneously, the real-time preload value is output at a 100Hz refresh rate by the real-time output unit for closed-loop control by downstream modules. The algorithm achieves a solution accuracy of ±1.5% under operating conditions from -20℃ to +60℃, meeting engineering application requirements.

[0018] The timing recording unit is used to record the preload value in a time sequence, forming and outputting preload timing data; The real-time output unit is used to output the preload value calculated at the current moment as the real-time preload value.

[0019] The preload monitoring module also includes: The storage unit stores the pre-stored acoustoelastic signal-preload relationship, which is obtained through preload calibration tests on bolts from the same batch within a temperature range of -40℃ to +80℃. The acoustoelastic calibration tests are conducted on a dedicated calibration test bench equipped with a high-precision hydraulic tensile tester (accuracy ±0.5%FS) and a temperature-controlled environmental chamber (temperature control accuracy ±1℃). Calibration samples are taken from high-strength bolts from the same batch, with no fewer than three sets of samples, each set containing M24, M30, and M36 bolts. Two pieces of each specification were used, covering the range of bolt specifications applicable to the system. The calibration process employed a graded loading strategy: at seven temperature points—-40℃, -20℃, 0℃, 20℃, 40℃, 60℃, and 80℃—six levels of load were applied, representing 20%, 40%, 60%, 80%, 100%, and 120% of the target preload value, respectively. Each load level was maintained for 30 seconds to eliminate creep effects. The ultrasonic signal acquisition frequency was set to 100Hz, and continuous acquisition for 10 seconds was used to calculate the average value as the acoustic transit time characteristic value under this condition. The experimental data were fitted to a three-dimensional surface using the least squares method to establish the functional relationship between acoustic transit time (Δt), stress (σ), and temperature (T).

[0020] in, The acoustoelastic coefficient (typical value approximately 1.2 × 10⁻⁶) -5 MPa -1 ) The temperature coefficient (approximately -0.8 × 10⁻⁶) -6 / ℃), σ is the coupling correction coefficient, C is the zero-point drift constant, σ is the stress, and T is the temperature; this calibration relationship is stored in the storage unit of the preload monitoring module as a reference for subsequent real-time calculation.

[0021] The ultrasonic sensing unit is a magnetostrictive ultrasonic sensor, arranged on the end face of the bolt head and in contact with the bolt end face through a coupling agent. The preload monitoring module uses a magnetostrictive ultrasonic sensor (EMAT) as the core detection element, specifically the MS-5000H industrial magnetostrictive ultrasonic probe. This probe has a built-in permanent magnet providing an axial bias magnetic field with a magnetic field strength set at 800 A / m to ensure the bolt material reaches the ideal magnetostrictive sensitivity range. The sensor front end has an integrated excitation and receiving coil winding, preferably with 80 turns and a wire diameter of 0.2 mm. Ultrasonic excitation with an adjustable frequency from 2 MHz to 10 MHz is achieved through a pulse transmission / reception circuit. During installation, the sensor is positioned at the center of the bolt head end face using a special fixture. The fixture employs a three-point positioning jaw structure to ensure a coaxiality error of less than 0.1 mm. The coupling agent is a high-temperature resistant silicone grease (operating temperature -40℃ to +150℃), with an application thickness controlled between 0.1 mm and 0.3 mm to ensure acoustic coupling stability and avoid acoustic time-of-flight drift introduced by thickness errors.

[0022] The attenuation feature identification module is used to receive the preload time series data, extract the attenuation feature parameters of the preload time series data, determine and output the attenuation category of the bolt; The attenuation feature identification module includes: The data receiving unit is used to receive and buffer the preload timing data; The feature extraction unit is used to extract attenuation feature parameters and attenuation trend from the preload time series data. The attenuation feature parameter is the rate of decrease of preload during the first holding time. The attenuation trend is extracted by linear fitting or exponential fitting. The category determination unit is used to determine and output the attenuation category of the bolt based on the attenuation characteristic parameters. When the preload decrease rate is greater than 1% / minute of the target preload, it outputs a rapid attenuation type; otherwise, it outputs a slow attenuation type. The attenuation category threshold is determined based on extensive experimental statistical data. The experiment selected 100 groups of 10.9 grade M30 high-strength bolts from the same batch. Under standard test conditions (temperature 20±2℃, humidity 60%RH), an initial preload of 600kN was applied, and the natural attenuation curve of the preload was recorded within the first 5 minutes. Statistical analysis showed that the sample attenuation rate exhibited a clear bimodal distribution: the rapid attenuation type samples (approximately 35%) mainly had an attenuation rate in the range of 0.8-1.5% / minute, with a median of 1.2% / minute; the slow attenuation type samples (approximately 65%) mainly had an attenuation rate in the range of 0.2-0.6% / minute, with a median of 0.4% / minute. By fitting a normal distribution and analyzing the 95% confidence interval, 1.0% / minute (based on a target preload of 600kN, i.e., 6kN / minute) was determined as the statistical dividing point between the two distributions, with a confidence level greater than 95%. Therefore, 1% / minute of the target preload was selected as the judgment threshold. When the decay rate output by the feature extraction unit is greater than this threshold, the category judgment unit outputs a fast decay type label; otherwise, it outputs a slow decay type label.

[0023] The physical mechanism underlying these differences in preload decay primarily stems from the microstructure and material properties of the bolt-connected interface. Rapid decay typically corresponds to bolts with high surface roughness (Ra>6.3μm) or those that are unphosphated. During initial loading, surface micro-protrusions undergo intense plastic deformation and embedding, leading to rapid relaxation of the preload within a short time. Slow decay, on the other hand, corresponds to bolts with low surface roughness (Ra<3.2μm) and that have undergone phosphated or coated treatment. The interface exhibits primarily elastic deformation with a lower creep rate. Furthermore, differences in the friction coefficient of the threaded pair (0.18-0.22 for rapid decay and 0.12-0.15 for slow decay) also result in different torque-to-preload conversion efficiencies, indirectly affecting the decay rate. By pre-identifying the decay category, the system can adjust the additional compensation amount in the second-level compensation stage to compensate for the differences in preload loss caused by different physical mechanisms.

[0024] The graded compensation decision module receives the attenuation category, the externally input temperature environment, and the externally input target preload based on the target preload, and generates a first control command, a second control command, and corresponding first holding time and second holding time. The first control command is used to control the preload application mechanism to stretch the bolt to a first preload higher than the target preload and hold it for the first holding time. The second control command is used to control the preload application mechanism to apply an additional compensation amount determined based on the attenuation category after the first holding time ends and hold it for the second holding time. The hierarchical compensation decision module includes: The first-level control unit is used to generate the first control command, control the preload application mechanism to stretch the bolt to a first preload higher than the target preload and maintain it for the first holding time. The first-level control unit is also configured to monitor the attenuation trend in real time based on preload feedback during the first holding time. The first-level control unit uses an incremental PID control algorithm to achieve precise application and maintenance of the preload, and the control parameters are tuned to a proportional coefficient. 0.8, integral coefficient The differential coefficient is 0.1. The value is 0.05, the sampling period is set to 50ms, and the preload feedback source in the control process adopts a dual-channel redundant design: the main channel is the magnetostrictive displacement sensor (1μm resolution) built into the preload application mechanism in the execution module, which calculates the clamping force through piston displacement, using the following formula:

[0025] Where k is the system stiffness coefficient; the auxiliary channel is the ultrasonic real-time solution value of the preload monitoring module, used for cross-validation. When the deviation between the two channels exceeds 2%, a sensor fault alarm is triggered and the system switches to single-channel control. During the first holding time, the first-level control unit compares the target preload with the feedback value in real time, and dynamically adjusts the servo valve opening through a PID algorithm to achieve smooth application of tensile force. At the same time, this unit also receives the attenuation trend slope λ (unit: kN / min) output by the attenuation feature identification module in real time, which is used for the prediction of the second-level compensation. The second-level control unit is used to correct the additional compensation amount based on the attenuation trend, generate the second control command, and control the preload application mechanism to apply the corrected additional compensation amount and maintain the second holding time after the first holding time ends; wherein, the second-level control unit corrects the additional compensation amount based on the attenuation trend monitored during the first holding period; the basic additional compensation amount Set at 3% of the target preload, this value is derived from extensive engineering practice statistics (which can compensate for more than 80% of conventional relaxation losses); a correction formula is introduced to address the attenuation characteristics of specific bolts:

[0026] ,in The measured slope of the attenuation trend. The reference slope is taken as 0.4% of the target preload / min (the median of the slow decay type), and k is the correction factor (determined to be 0.6-0.8 through calibration tests). Additional compensation based on the base; when a fast decay type is detected ( When the compensation is relatively large, the corrected additional compensation amount can reach 5%; for the slow decay type, it is reduced to 2%, achieving precise adaptation of the compensation amount.

[0027] The prediction compensation unit is used to calculate a pre-compensation amount and superimpose it onto the second control command based on the actual residual preload at the end of the first holding time, the target preload, and the predicted attenuation amount during the second holding time. The predicted attenuation amount is obtained by extrapolation based on the attenuation trend. The prediction compensation unit further introduces feedforward control to eliminate the predicted attenuation during the second holding time. The formula for calculating the pre-compensation amount is:

[0028] in, This unit is based on the actual residual preload at the end of the first stage. Preload for target, The predicted attenuation amount is extrapolated to the attenuation trend and then superimposed onto the second control command.

[0029] The first preload is 108%-112% of the target preload, and is determined according to the bolt specification: 110%-112% for M24 and below, 109%-111% for M27-M30, and 108%-110% for M33 and above; the first holding time is determined according to the ambient temperature: 8 minutes for below 0℃, 6 minutes for 0℃-30℃, and 5 minutes for above 30℃.

[0030] Regarding the over-tensioning ratio of the first preload, a comparative experiment selected 30 samples from each of three schemes: 105%, 110%, and 115%. The dispersion (standard deviation) of the residual preload after the second stage of retention was statistically analyzed: ±4.2% for the 105% group, ±1.8% for the 110% group, and ±3.5% for the 115% group. This demonstrates that 110% is the point of minimum dispersion, thus determining the over-tensioning range to be 108%-112%. For different bolt specifications, through compliance calculation (C=L / AE), it was found that for M24 bolts (L=120mm), With a flexibility of 0.34×10-6 mm / N, it can withstand high over-tension (110%-112%), while M33 bolts (L=180mm) can withstand high over-tension. Due to its high stiffness and stress sensitivity, a lower proportion (108%-110%) is adopted for the material with a flexibility of 0.26×10-6mm / N to avoid the risk of yielding. The temperature correction coefficient is determined through five sets of temperature comparison tests: the same electrical signal is applied in environments of -20℃, 0℃, 20℃, 40℃, and 60℃, and the preload drift is measured to be -8%, -4%, 0%, +4%, and +8%, respectively. The fitted temperature sensitivity coefficient is 0.5% / 10℃, that is, for every 10℃ deviation of the ambient temperature from the reference, the compensation amount increases or decreases by 0.5% to offset the effects of low-temperature shrinkage or high-temperature creep of the material.

[0031] The additional compensation amount is determined according to the following algorithm: The basic compensation is 3% of the target preload, and the attenuation type correction is +2% for fast attenuation type and -1% for slow attenuation type. The temperature correction is increased by 0.5% for every 10°C below the ambient temperature and decreased by 0.5% for every 30°C above the ambient temperature. The second holding time is dynamically adjusted based on the deviation between the actual residual preload and the target preload at the end of the first holding stage: 3 minutes when the deviation is less than 2%, 4 minutes when the deviation is 2%-4%, and 5 minutes when the deviation is greater than 4%.

[0032] The fine-tuning module receives the real-time preload value during the second holding time and compares it with the target preload. When the deviation exceeds a preset threshold, it outputs a fine-tuning command. The fine-tuning command is used to control the preload application mechanism to increase or decrease the preload by a preset step size until the deviation falls into the target range. The fine-tuning module includes: The comparison unit is used to compare the real-time preload value with the target preload; The threshold determination unit is used to determine whether the deviation exceeds a preset threshold. The graded adjustment unit is used to output a first fine-tuning command when the deviation is first determined to exceed a preset threshold. The preset step size of the first fine-tuning command is determined according to the attenuation category. After a single adjustment, if the deviation still exceeds the preset threshold, a second fine-tuning command is output. The preset step size of the second fine-tuning command is 60% of that of the first fine-tuning command. If the deviation still exceeds the preset threshold for the third time, a third fine-tuning command is output and the bolt is marked as abnormal. The preset step size of the third fine-tuning command is 60% of that of the second fine-tuning command. The number of adjustments is limited and used to accumulate the number of adjustments. When the number of adjustments exceeds 3, the output of the adjustment command will stop and an alarm will be triggered.

[0033] The preset threshold, preset step size, and target interval are determined according to the attenuation category classification: For the rapid decay type, the preset threshold is ±4%, the preset step size is 1% of the target preload, and the target range is ±3%; for the slow decay type, the preset threshold is ±2%, the preset step size is 0.5% of the target preload, and the target range is ±1.5%.

[0034] The graded adjustment unit employs an adaptive step-size reduction strategy to achieve rapid convergence. Comparative experiments determined the step-size reduction coefficient to be 60%. The experiment selected 100 groups of M30 bolts and compared fine-tuning using a fixed step size (100% target preload) and a decreasing step size (100%→60%→36%). The results showed that the fixed step-size strategy required an average of 5.2 adjustments to fall within the target range and exhibited oscillation and overshoot. In contrast, the 60% decreasing strategy achieved a 96% convergence rate within 3 adjustments, reducing the average number of adjustments to 2.1, effectively avoiding overshoot and oscillation. The mathematical basis for this strategy is that a larger initial adjustment length (corresponding to a larger decay)... The standard step size determined by the category can quickly approach the target. The second step size is reduced to 60% to suppress overshoot. The third step size is further reduced to 36% (60%×60%) to achieve fine correction. If the deviation still exceeds the preset threshold after the third adjustment, it is judged as an abnormal working condition (which may be due to thread jamming, severe surface damage or sensor failure). The number of times the unit limits the number of times the abnormality is triggered, the audible and visual alarm is activated (red warning light + buzzer), and the abnormal bolt number and suggested handling measures (retighten / replace bolt / check sensor) are displayed on the human-machine interface. At the same time, the data of the abnormal bolt is marked in red and stored in the abnormal database for manual review. Fine-tuning parameters are configured differently based on the attenuation type: For fast-attenuating bolts, since their preload naturally decays quickly, setting an overly strict threshold would lead to frequent adjustments or even oscillations. Therefore, a wider threshold of ±4%, a larger step size of 1% of the target preload (approximately 6kN@M30), and a target range of ±3% are set, allowing for some deviation to maintain system stability. For slow-attenuating bolts, which have good natural stability, high-precision control can be implemented. A strict threshold of ±2%, a fine step size of 0.5% of the target preload (approximately 3kN@M30), and a target range of ±1.5% are set to ensure that the final preload is close to the design value. This mapping relationship is pre-written into the system configuration table through calibration tests. When the attenuation feature identification module outputs the category identifier, the fine-tuning module automatically calls the corresponding parameter group to achieve graded control of "coarse adjustment - fine adjustment - precision adjustment".

[0035] The execution module is used to receive the first control command, the second control command, or the fine-tuning command, and control the preload application mechanism to apply and maintain the preload.

[0036] The execution module includes: The instruction parsing unit is used to receive and parse the first control instruction, the second control instruction, or the fine-tuning instruction, and extract the target preload force, holding time, and adjustment step size. A preload loading unit is used to apply tensile load to bolts. The preload loading unit includes a hydraulic loading unit or an electric drive loading unit. The hydraulic loading unit is used for M30 and above specifications, and the electric drive loading unit is used for M27 and below specifications. Combination Figure 2 As shown, the preload loading unit, serving as the force output terminal of the execution module, adopts a dual-channel parallel architecture design. The loading mode is automatically switched according to the bolt specifications via a mode selection switch. When dealing with large bolts of M30 and above, the system activates the hydraulic loading subunit. This unit consists of a servo hydraulic pump station and an ultra-high pressure tensioner. The pump station provides a stable 70MPa oil source, which, after precise adjustment by the servo valve, enters the tensioner cylinder, pushing the piston upwards to tension the bolt through the sleeve structure, achieving the application of large tonnage (typically >500kN) preload. When dealing with small to medium-sized bolts of M27 and below, the system switches to the electric drive loading subunit. This unit uses a combination of a servo motor and a precision planetary reducer to convert the high-speed, low-torque output of the motor into low-speed, high-torque output. A CNC wrench head applies precise torque to the bolt, achieving control of small tonnage bolts (typically <400kN) through the torque-preload conversion relationship. Both subunits share the same control signal interface, but their physical output characteristics differ: the hydraulic subunit primarily uses displacement control (calculating preload through piston stroke), suitable for large-diameter bolts with high stiffness and sensitive deformation; the electric drive subunit primarily uses torque control (using torque sensor feedback), suitable for small to medium-diameter bolts requiring rapid loading and unloading. M30 and M27 serve as a dividing line due to the significant difference in the effective stress cross-sectional area of ​​the bolts (Aeff = 561 mm² for M30). 2 The Aeff of the M27 is 459 mm. 2 This boundary ensures that both loading methods operate within their optimal economic performance range, avoiding insufficient torque when using electric wrenches for large bolts and excessive redundancy when using hydraulic tensioners for small bolts.

[0037] The collaborative control unit is used to coordinate the loading sequence of each preload application mechanism during the synchronous tensioning of multiple bolts, so that the preload deviation of each bolt is within ±5%. The status recording unit is used to collect the oil pressure, displacement or motor current of the preload application mechanism in real time and output it to an external data recorder.

[0038] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention. The scope of protection claimed by the appended claims and their equivalents is defined.

Claims

1. A preload monitoring and compensation system for bolted connections in steel structures, characterized in that, include: The preload monitoring module uses an ultrasonic sensor placed on the end face of the bolt head to collect acoustic elastic signals in real time. Based on the pre-stored acoustic elastic signal-preload relationship, it calculates the preload value in real time and outputs preload timing data and real-time preload value. The attenuation feature identification module is used to receive the preload time series data, extract the attenuation feature parameters of the preload time series data, determine and output the attenuation category of the bolt; The graded compensation decision module receives the attenuation category, the externally input temperature environment, and the externally input target preload based on the target preload, and generates a first control command, a second control command, and corresponding first holding time and second holding time. The first control command is used to control the preload application mechanism to stretch the bolt to a first preload higher than the target preload and hold it for the first holding time. The second control command is used to control the preload application mechanism to apply an additional compensation amount determined based on the attenuation category after the first holding time ends and hold it for the second holding time. The fine-tuning module receives the real-time preload value during the second holding time and compares it with the target preload. When the deviation exceeds a preset threshold, it outputs a fine-tuning command. The fine-tuning command is used to control the preload application mechanism to increase or decrease the preload by a preset step size until the deviation falls into the target range. The execution module is used to receive the first control command, the second control command, or the fine-tuning command, and control the preload application mechanism to apply and maintain the preload.

2. The preload monitoring and compensation system for steel structure bolt connections according to claim 1, characterized in that, The preload monitoring module includes: An ultrasonic sensing unit is positioned on the end face of the bolt head for real-time acquisition of acoustoelastic signals. The calculation unit is used to calculate the acoustoelastic signal into a preload value in real time based on the pre-stored acoustoelastic signal-preload relationship; The timing recording unit is used to record the preload value in a time sequence, forming and outputting preload timing data; The real-time output unit is used to output the preload value calculated at the current moment as the real-time preload value.

3. The preload monitoring and compensation system for steel structure bolt connections according to claim 2, characterized in that, The preload monitoring module also includes: The storage unit is used to store the pre-stored acoustoelastic signal-preload relationship, which is obtained through a preload calibration test of the same batch of bolts in a temperature range of -40℃ to +80℃. The ultrasonic sensing unit is a magnetostrictive ultrasonic sensor, which is arranged on the end face of the bolt head and contacts the end face of the bolt through a coupling agent.

4. The preload monitoring and compensation system for steel structure bolt connections according to claim 1, characterized in that, The attenuation feature identification module includes: The data receiving unit is used to receive and buffer the preload timing data; The feature extraction unit is used to extract attenuation feature parameters and attenuation trend from the preload time series data. The attenuation feature parameter is the rate of decrease of preload during the first holding time. The attenuation trend is extracted by linear fitting or exponential fitting. The category determination unit is used to determine and output the attenuation category of the bolt based on the attenuation characteristic parameters. When the preload decrease rate is greater than 1% / minute of the target preload, the fast attenuation type is output; otherwise, the slow attenuation type is output.

5. The preload monitoring and compensation system for steel structure bolt connections according to claim 4, characterized in that, The hierarchical compensation decision module includes: The first-level control unit is configured to generate the first control command, control the preload application mechanism to stretch the bolt to a first preload force higher than the target preload force and maintain it for the first holding time. The first-level control unit is also configured to monitor the attenuation trend in real time based on preload feedback during the first holding time. The second-level control unit is used to correct the additional compensation amount based on the attenuation trend, generate the second control command, and control the preload application mechanism to apply the corrected additional compensation amount and maintain the second holding time after the first holding time ends. The prediction compensation unit is used to calculate the pre-compensation amount and superimpose it on the second control command based on the actual residual preload at the end of the first holding time, the target preload, and the predicted attenuation amount during the second holding time. The predicted attenuation amount is obtained by extrapolation based on the attenuation trend.

6. The preload monitoring and compensation system for steel structure bolt connections according to claim 5, characterized in that, The first preload is 108%-112% of the target preload, and is determined according to the bolt specification: 110%-112% for M24 and below, 109%-111% for M27-M30, and 108%-110% for M33 and above; the first holding time is determined according to the ambient temperature: 8 minutes for below 0℃, 6 minutes for 0℃-30℃, and 5 minutes for above 30℃.

7. The preload monitoring and compensation system for steel structure bolt connections according to claim 5, characterized in that, The additional compensation amount is determined according to the following algorithm: The basic compensation is 3% of the target preload, and the attenuation type correction is +2% for fast attenuation type and -1% for slow attenuation type. The temperature correction is increased by 0.5% for every 10°C below the ambient temperature and decreased by 0.5% for every 30°C above the ambient temperature. The second holding time is dynamically adjusted based on the deviation between the actual residual preload and the target preload at the end of the first holding stage: 3 minutes when the deviation is less than 2%, 4 minutes when the deviation is 2%-4%, and 5 minutes when the deviation is greater than 4%.

8. The preload monitoring and compensation system for steel structure bolt connections according to claim 1, characterized in that, The fine-tuning module includes: The comparison unit is used to compare the real-time preload value with the target preload; The threshold determination unit is used to determine whether the deviation exceeds a preset threshold. The graded adjustment unit is used to output a first fine-tuning command when the deviation is first determined to exceed a preset threshold. The preset step size of the first fine-tuning command is determined according to the attenuation category. After a single adjustment, if the deviation still exceeds the preset threshold, a second fine-tuning command is output. The preset step size of the second fine-tuning command is 60% of that of the first fine-tuning command. If the deviation still exceeds the preset threshold for the third time, a third fine-tuning command is output and the bolt is marked as abnormal. The preset step size of the third fine-tuning command is 60% of that of the second fine-tuning command. The number of adjustments is limited and used to accumulate the number of adjustments. When the number of adjustments exceeds 3, the output of the adjustment command will stop and an alarm will be triggered.

9. The preload monitoring and compensation system for steel structure bolt connections according to claim 8, characterized in that, The preset threshold, preset step size, and target interval are determined according to the attenuation category classification: For the rapid decay type, the preset threshold is ±4%, the preset step size is 1% of the target preload, and the target range is ±3%; for the slow decay type, the preset threshold is ±2%, the preset step size is 0.5% of the target preload, and the target range is ±1.5%.

10. The preload monitoring and compensation system for steel structure bolt connections according to claim 1, characterized in that, The execution module includes: The instruction parsing unit is used to receive and parse the first control instruction, the second control instruction, or the fine-tuning instruction, and extract the target preload force, holding time, and adjustment step size. A preload loading unit is used to apply tensile load to bolts. The preload loading unit includes a hydraulic loading unit or an electric drive loading unit. The hydraulic loading unit is used for M30 and above specifications, and the electric drive loading unit is used for M27 and below specifications. The collaborative control unit is used to coordinate the loading sequence of each preload application mechanism during the synchronous tensioning of multiple bolts, so that the preload deviation of each bolt is within ±5%. The status recording unit is used to collect the oil pressure, displacement or motor current of the preload application mechanism in real time and output it to an external data recorder.

Images