Compensation method and system for temperature influence in passive bolt monitoring system

Abstract

The invention discloses a temperature influence compensation method and system in a passive bolt monitoring system, and belongs to the technical field of structural health monitoring, and the method comprises the steps: obtaining a resonant frequency signal of a passive bolt monitoring sensor and temperature data, collected by a preset distributed temperature sensing network, of a plurality of monitoring points; reconstructing a two-dimensional temperature field cloud picture of the bolt connection area through interpolation calculation; extracting an equivalent comprehensive temperature rise value from the temperature field distribution, and performing preliminary compensation on the resonant frequency signal to generate a preliminary strain value; calculating a virtual thermal strain value generated by temperature influence; and performing decoupling operation on the initial strain value based on the virtual thermal strain value to obtain a compensated real mechanical strain value, and accurately separating the real mechanical strain from the total measurement strain by adopting a method of reconstructing a two-dimensional temperature field of a bolt connection area and calculating a decoupling virtual thermal strain by utilizing a finite element analysis micro-simulation model. Therefore, the accuracy and reliability of bolt pre-tightening force monitoring are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of structural health monitoring, in particular to a temperature influence compensation method and system in a passive bolt monitoring system. BACKGROUND

[0002] At present, bolt connection is one of the most basic and widely used connection methods in mechanical equipment, bridge structure, wind turbine and other large key engineering structures. The pre-tightening force of the bolt is directly related to the reliability, sealing and overall safety of the structure. Therefore, it is of great significance to monitor the bolt pre-tightening force for a long time, in real time and accurately. Passive bolt monitoring sensors show great application potential in this field due to their advantages such as no need for built-in power supply, long service life and maintenance-free. Such sensors usually reflect the pre-tightening force state of the bolt by measuring a stress-related physical quantity such as the change of the resonant frequency.

[0003] In the existing technology, in order to solve the problem of temperature change interfering with the measurement results of the passive bolt monitoring sensor, a single temperature sensor is generally arranged near the bolt. The temperature value of a measurement point is obtained by the temperature sensor, and then the resonant frequency signal collected is corrected according to the relationship curve or empirical formula between the sensor resonant frequency and the temperature previously calibrated in the laboratory, so as to try to eliminate the influence of temperature change.

[0004] However, the existing technical solution has obvious defects. In actual working conditions, due to the influence of complex factors such as sunlight, equipment heating and air flow, the temperature distribution of the bolt and its connecting parts is often extremely uneven, and there is a large temperature gradient. The temperature value of a single measurement point cannot represent the true thermal state of the entire bolt connection area. This compensation method only considers the influence of temperature on the electrical or magnetic properties of the sensor, but ignores the fact that temperature change will cause thermal expansion and contraction of the bolt itself and the connected parts, thereby generating an additional thermal strain. The thermal strain and the mechanical strain generated by the pre-tightening force are superimposed together, and only by correcting the sensor signal drift cannot effectively separate the two. SUMMARY

[0005] To solve the above problems, the present application provides a temperature influence compensation method and system in a passive bolt monitoring system, which adopts the method of reconstructing the two-dimensional temperature field of the bolt connection area and calculating the decoupling virtual thermal strain by using the finite element analysis micro-simulation model, so as to accurately separate the real mechanical strain from the total measured strain, thereby improving the accuracy and reliability of the bolt pre-tightening force monitoring.

[0006] The above object can be achieved by the following scheme: A method for compensating temperature influence in a passive bolt monitoring system, comprising: acquiring a resonant frequency signal of a passive bolt monitoring sensor and temperature data of a plurality of monitoring points collected by a preset distributed temperature sensing network; reconstructing a two-dimensional temperature field cloud map of a bolt connection area by interpolation calculation according to the temperature data of the plurality of monitoring points; extracting an equivalent comprehensive temperature rise value from the temperature field distribution, preliminarily compensating the resonant frequency signal according to a preset temperature-frequency drift relationship to generate a preliminary strain value; inputting the equivalent comprehensive temperature rise value into a preset finite element analysis micro-simulation model to calculate a virtual thermal strain value generated by temperature influence; and decoupling the preliminary strain value based on the virtual thermal strain value to obtain a compensated real mechanical strain value.

[0007] Optionally, the acquiring of the resonant frequency signal of the passive bolt monitoring sensor and the temperature data of the plurality of monitoring points collected by the preset distributed temperature sensing network comprises: controlling a wireless reader-writer to synchronously transmit an inquiry signal to the passive bolt monitoring sensor and the distributed temperature sensing network; synchronously receiving and demodulating a return signal from the passive bolt monitoring sensor to obtain the resonant frequency signal, and synchronously reading measurement values of each node in the distributed temperature sensing network to obtain the temperature data.

[0008] Optionally, the reconstructing of the two-dimensional temperature field cloud map of the bolt connection area by interpolation calculation according to the temperature data of the plurality of monitoring points comprises: acquiring position coordinates of each sensor in the distributed temperature sensing network and corresponding temperature readings thereof; and generating a two-dimensional temperature field cloud map covering an end surface of a bolt head, an end surface of a nut and a surface of a connected part by inverse distance weighted interpolation algorithm according to the position coordinates and the temperature readings.

[0009] Optionally, the extracting of the equivalent comprehensive temperature rise value from the temperature field distribution comprises: identifying temperature distribution of a bolt shank area according to the two-dimensional temperature field cloud map; and performing weighted average on the temperature of the bolt shank area according to a preset weight coefficient calculation rule, the weight coefficient being determined based on distances of each pixel point from a center of the bolt shank and heat capacities of components in which the pixel points are located, and outputting the equivalent comprehensive temperature rise value.

[0010] Optionally, the inputting of the equivalent comprehensive temperature rise value into the preset finite element analysis micro-simulation model to calculate the virtual thermal strain value generated by temperature influence comprises: calculating a ratio of temperature variation to spatial variation of the two-dimensional temperature field cloud map to obtain temperature gradient information; loading the equivalent comprehensive temperature rise value and the temperature gradient information as boundary conditions to the preset finite element analysis micro-simulation model to output a thermal induced strain field of the bolt; and extracting an average strain value of the bolt shank from the thermal induced strain field to generate the virtual thermal strain value.

[0011] Optionally, the decoupling operation on the preliminary strain value based on the virtual thermal strain value to obtain a compensated real mechanical strain value comprises: performing decoupling operation on the preliminary strain value as a minuend and the virtual thermal strain value as a subtrahend to separate out the real mechanical strain value.

[0012] Optionally, after obtaining the real mechanical strain value, the method further comprises: calculating a current pre-tightening force of the bolt according to the real mechanical strain value and a preset elastic modulus of the bolt; judging whether the current pre-tightening force of the bolt satisfies a preset thermal stability condition; if yes, taking the real mechanical strain value currently calculated as a reference strain value; calculating an error between the reference strain value and a strain value obtained through a single temperature sensor compensation model; and performing reverse fine tuning on parameters of the temperature-frequency drift relationship and the finite element analysis micro-simulation model according to the error.

[0013] Optionally, the judgment of whether the current pre-tightening force of the bolt satisfies the preset thermal stability condition comprises: calculating an instantaneous change rate of the temperature data of the plurality of monitoring points and a maximum temperature difference therebetween; and if the instantaneous change rate is lower than a first threshold value and the maximum temperature difference is lower than a second threshold value, it is determined that the thermal stability condition is satisfied.

[0014] Optionally, the reverse fine tuning on the parameters of the temperature-frequency drift relationship or the finite element analysis micro-simulation model according to the error comprises: using a least square method to iteratively optimize coefficients in the temperature-frequency drift relationship and the finite element analysis micro-simulation model parameters with the objective of minimizing the error.

[0015] Based on the same inventive concept, the present application also provides a temperature influence compensation system in a passive bolt monitoring system, comprising: a multi-source acquisition module configured to acquire a resonant frequency signal of a passive bolt monitoring sensor and temperature data of a plurality of monitoring points acquired through a preset distributed temperature sensing network; a three-dimensional temperature field reconstruction module configured to reconstruct a two-dimensional temperature field cloud map of a bolt connection area through interpolation calculation according to the temperature data of the plurality of monitoring points; a temperature-frequency collaborative processing module configured to extract an equivalent comprehensive temperature rise value from the temperature field distribution, perform preliminary compensation on the resonant frequency signal according to a preset temperature-frequency drift relationship, and generate a preliminary strain value; a virtual thermal strain simulation module configured to input the equivalent comprehensive temperature rise value into a preset finite element analysis micro-simulation model to calculate a virtual thermal strain value caused by temperature influence; a mechanical strain decoupling module configured to perform decoupling operation on the preliminary strain value based on the virtual thermal strain value to obtain a compensated real mechanical strain value.

[0016] Compared with the prior art, the present application has the following advantages: 1. By finely reconstructing the temperature field of the bolt connection area and combining high-fidelity physical simulation model for calculation, the measurement error caused by temperature is deeply and comprehensively compensated. This method not only considers the signal drift of the sensor itself due to temperature, but more importantly, it can accurately quantify and eliminate the thermal strain caused by the thermal expansion and contraction of the bolt and the connected parts, so as to obtain the true mechanical strain and improve the accuracy of the bolt pretightening force measurement.

[0017] 2. The present application proposes an online self-adaptive calibration mechanism based on actual working conditions, so that the monitoring system has the ability of self-learning and optimization. The system can intelligently identify the thermal stability condition, and use the high-credibility measurement results under the condition as the reference to fine-tune the parameters of the core compensation model in reverse. This closed-loop feedback optimization mechanism ensures that the system can adapt to factors such as sensor aging and long-term changes in material properties, maintains high-precision measurement performance throughout the life cycle, and enhances the long-term reliability and robustness of the monitoring system.

[0018] 3. The distributed temperature sensing network and two-dimensional temperature field cloud reconstruction technology can comprehensively capture the complex and uneven temperature distribution of the bolt connection area. Compared with the traditional single-point temperature measurement method, this method provides accurate characterization of temperature gradient and local hot spot area, and provides more real and detailed boundary conditions for subsequent equivalent temperature rise extraction and finite element simulation analysis, fundamentally improving the physical reality of temperature influence analysis, and is an important foundation for high-precision compensation.

[0019] Other features and advantages of the present application will be set forth in the following description, and in part will become apparent to those skilled in the art from the description, or can be learned by practice of the present application. The objects and other advantages of the present application can be realized and obtained by the structures indicated in the specification, claims and drawings. BRIEF DESCRIPTION OF DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiment or prior art description. Obviously, the drawings in the following description are some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings.

[0021] Figure 1 is a flowchart of a temperature influence compensation method in a passive bolt monitoring system according to an embodiment of the present application.

[0022] Figure 2 is a two-dimensional temperature field cloud diagram and equivalent comprehensive temperature rise value extraction schematic diagram of a bolt connection area according to an embodiment of the present application.

[0023] Figure 3 is a strain value comparison curve before and after temperature compensation of an embodiment of the present application.

[0024] Figure 4 is a structural schematic diagram of a temperature influence compensation system in a passive bolt monitoring system of an embodiment of the present application. DETAILED DESCRIPTION

[0025] In order to make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme of the embodiments of the present application will be clearly and completely described below with reference to the drawings of the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative labor fall within the scope of protection of the present application.

[0026] With reference to Figure 1 An embodiment of the present application proposes a temperature influence compensation method in a passive bolt monitoring system, which adopts a method of reconstructing a two-dimensional temperature field of a bolt connection area and calculating decoupled virtual thermal strain by using a finite element analysis micro-simulation model, and can accurately separate real mechanical strain from total measured strain, thereby improving the accuracy and reliability of bolt pretightening force monitoring.

[0027] The method of the embodiment specifically includes: obtaining a resonant frequency signal of a passive bolt monitoring sensor and temperature data of a plurality of monitoring points collected through a preset distributed temperature sensing network; reconstructing a two-dimensional temperature field cloud map of a bolt connection area by interpolation calculation according to the temperature data of the plurality of monitoring points; extracting an equivalent comprehensive temperature rise value from the temperature field distribution, preliminarily compensating the resonant frequency signal according to a preset temperature-frequency drift relationship, and generating a preliminary strain value; inputting the equivalent comprehensive temperature rise value into a preset finite element analysis micro-simulation model to calculate a virtual thermal strain value generated by temperature influence; performing decoupling operation on the preliminary strain value based on the virtual thermal strain value to obtain a compensated real mechanical strain value.

[0028] Specifically, a dual temperature compensation mechanism based on physical simulation and measured data fusion is established, which abandons the traditional single-point temperature measurement mode. Through the distributed sensing network and interpolation algorithm, a refined two-dimensional temperature field of the bolt connection area is reconstructed, so as to obtain an equivalent comprehensive temperature rise value which can fully reflect the non-uniform heat distribution. The temperature influence is stripped in two steps. First, the preset temperature-frequency drift relationship is used to preliminarily correct the electrical characteristic drift of the sensor itself, and a mixed preliminary strain value containing the real mechanical strain and the structural thermal strain is obtained. Second, the equivalent comprehensive temperature rise value is loaded into the finite element analysis micro-simulation model consistent with the actual structure, and the virtual thermal strain value caused by the current temperature field is independently and accurately calculated. Finally, according to the strain linear superposition principle, the virtual thermal strain value is subtracted from the mixed preliminary strain value to realize the accurate decoupling of the thermal strain and the mechanical strain, and the real mechanical strain value free from temperature interference is finally obtained. The monitoring accuracy and reliability under complex and dynamic temperature environment are improved. By reconstructing the two-dimensional temperature field, this method can accurately capture the local temperature difference and temperature gradient caused by heat sources, air flow and other factors, and overcome the large error caused by single-point temperature measurement. More importantly, by introducing the finite element micro-simulation model to quantitatively calculate the thermal strain of the structure, this method fundamentally solves the problem that the traditional compensation method is difficult to accurately separate the influence of structural thermal expansion, and realizes comprehensive and deep compensation for the temperature effects of sensor drift and structural thermal deformation. Finally, the real mechanical strain value related only to the bolt pretightening force is output, thereby effectively avoiding the misjudgment of the pretightening force caused by temperature fluctuations, and providing solid technical support for ensuring the safety and stable operation of the key bolt connection structure.

[0029] Optionally, the acquisition of the resonant frequency signal of the passive bolt monitoring sensor and the temperature data of the plurality of monitoring points collected through the preset distributed temperature sensing network comprises: controlling the wireless reader to synchronously transmit an inquiry signal to the passive bolt monitoring sensor and the distributed temperature sensing network; synchronously receiving and demodulating the return signal from the passive bolt monitoring sensor to obtain the resonant frequency signal, and synchronously reading the measurement values of each node in the distributed temperature sensing network to obtain the temperature data.

[0030] Specifically, the wireless reader initiates the inquiry signal. The wireless reader is internally integrated with a high-precision clock module, which can uniformly schedule the transmission and reception timing of the signal. At the beginning of data collection, the wireless reader will synchronously trigger two parallel communication channels to transmit inquiry signals to the passive bolt monitoring sensor and the distributed temperature sensing network. For the passive bolt monitoring sensor, the inquiry signal is usually a broadband electromagnetic wave sweep signal. The sensor itself does not contain a power supply, and its operating energy is entirely derived from the electromagnetic wave transmitted by the wireless reader. When the frequency of the sweep signal matches the inherent resonance frequency of the sensor, the sensor will resonate and reflect or retransmit the signal with the maximum energy back to the wireless reader. The receiving module of the wireless reader captures this return signal and demodulates it. The demodulation process mainly analyzes the frequency spectrum of the return signal, finds the frequency corresponding to the energy peak point through fast Fourier transform, and thus accurately obtains the resonance frequency signal representing the current stress state of the bolt. For the distributed temperature sensing network, the inquiry signal is a digital instruction used to wake up all temperature sensing nodes in the network and request them to report measurement data. The network is composed of multiple independent micro temperature sensor nodes arranged around the key positions of the bolt connection area. Each node will immediately perform a temperature measurement after receiving the synchronization instruction from the wireless reader, and package the obtained temperature data into a digital signal for return through wireless means. The wireless reader synchronously receives the return data packets from all nodes and performs analysis to obtain the accurate temperature data of multiple monitoring points in the bolt connection area at the same time.

[0031] Optionally, the two-dimensional temperature field cloud map of the bolt connection area is reconstructed by interpolation calculation according to the temperature data of the plurality of monitoring points, comprising: obtaining the position coordinates of each sensor in the distributed temperature sensing network and the corresponding temperature readings; generating a two-dimensional temperature field cloud map covering the bolt head end face, the nut end face and the connected component surface through the inverse distance weighted interpolation algorithm according to the position coordinates and the temperature readings.

[0032] Specifically, the basic data, i.e., the accurate two-dimensional spatial position coordinates of each sensor in the distributed temperature sensing network pre-calibrated and stored in the deployment stage, and the corresponding real-time temperature readings of each sensor synchronously collected in the monitoring process, are obtained. The inverse distance weighted interpolation algorithm is used to calculate the temperature value of any unarranged sensor position in the target area. For the estimated temperature T(x, y) of any point in the two-dimensional temperature field, , wherein, is the temperature estimation value of the interpolation point to be solved; n is the total number of sensors in the distributed temperature sensing network; a real temperature measurement value representative of the i-th sensor, obtained by the step of synchronous acquisition; a Euclidean distance between the point (x, y) to be determined and the i-th sensor position p is a preset power index for adjusting the decay rate of the weight with the distance, usually taking a value of 2, which can be calibrated and optimized according to the heat transfer characteristics of the actual application scene. By traversing and calculating the dense mesh of the entire bolt connection area, including the bolt head end face, the nut end face and the connected component surface, each mesh node is assigned a temperature value calculated by interpolation. Through the pseudo-color rendering technology, an intuitive two-dimensional temperature field cloud chart is generated, and different color gradients in the chart clearly show the temperature level and distribution trend of the entire connection area.

[0033] Optionally, the extracting an equivalent comprehensive temperature rise value from the temperature field distribution comprises: According to the two-dimensional temperature field cloud chart, the temperature distribution of the bolt shank area is identified; According to a preset weight coefficient calculation rule, the temperature of the bolt shank area is weighted and averaged, the weight coefficient is determined based on the distance of each pixel point from the center of the bolt shank and the heat capacity of the component, and the equivalent comprehensive temperature rise value is output.

[0034] Specifically, based on the pre-established geometric model, the area corresponding to the bolt shank is accurately identified and circled in the two-dimensional temperature field cloud chart. This area is the core part of the bolt under the tensile pre-tightening force, and the temperature change is directly related to the accuracy of stress monitoring. A weighted average strategy is used to calculate the equivalent comprehensive temperature rise value. A preset weight coefficient calculation rule is established, and the weight coefficient of each pixel point or discrete grid point in the area is determined by two key physical quantities. One is the distance of the point to the geometric center of the bolt shank, and the other is the material heat capacity of the component where the point is located. Specifically, the closer the position to the center axis of the bolt shank, the more direct the influence of the temperature change on the axial thermal expansion of the bolt, and a higher weight is given; the components in contact with the bolt, such as the connected components, have larger heat capacity, have greater influence on the temperature field distribution of the bolt, and the temperature points in the contact area with these components also have higher weight. For the equivalent comprehensive temperature rise value there is: , wherein, is the final output equivalent comprehensive temperature rise value; the summation symbol ∑ represents the traversal calculation of all n discrete points in the bolt shank area; is the temperature value of the i-th discrete point in the two-dimensional temperature field cloud chart; is the reference baseline temperature in the initial state or a certain stable working condition; is the weight coefficient of the ith discrete point, whose value is determined by the distance between the point and the center of the bolt shank and the thermal capacity of the component where the point is located, etc. Figure 2 Fig. 2 shows a two-dimensional temperature field cloud map of a bolted joint area and an extraction diagram of an equivalent comprehensive temperature rise value, representing the logic of two-dimensional temperature field cloud map reconstruction and equivalent temperature rise value extraction, and realizing accurate quantification of non-uniform temperature field through inverse distance weighted interpolation and regional weighted average.

[0035] Optionally, the equivalent comprehensive temperature rise value is input into a preset finite element analysis micro-simulation model to calculate a virtual thermal strain value caused by temperature influence, including: calculating the proportion of temperature change and spatial change of the two-dimensional temperature field cloud map to obtain temperature gradient information; loading the equivalent comprehensive temperature rise value and the temperature gradient information as boundary conditions to the preset finite element analysis micro-simulation model to output the thermal induced strain field of the bolt; extracting the average strain value of the bolt shank from the thermal induced strain field to generate the virtual thermal strain value.

[0036] Specifically, the core of the process of calculating the virtual thermal strain value by the finite element analysis micro-simulation model is to construct a digital model highly consistent with the actual bolted joint structure in terms of geometric size, material property and constraint condition, i.e. the preset finite element analysis micro-simulation model. The accurate three-dimensional geometry and mesh division of the bolt, nut and connected component are defined in advance, and the real material parameters such as elastic modulus, Poisson's ratio and thermal expansion coefficient are assigned. The two-dimensional temperature field cloud map is processed to extract the boundary conditions applied to the simulation model, including calculating the proportion of temperature change with spatial position in the two-dimensional temperature field cloud map, i.e. temperature gradient information. The temperature gradient describes the rate of change of temperature in space and reflects the non-uniform heating of the bolted joint area. The equivalent comprehensive temperature rise value and the temperature gradient information are loaded as thermal load boundary conditions to the finite element analysis micro-simulation model. The equivalent comprehensive temperature rise value is the overall temperature change, and the temperature gradient information is accurately applied to the surface of the corresponding components in the form of field distribution, completely reproducing the real non-uniform temperature field. After loading the thermal load, the finite element solver is started for thermal analysis. In order to separately calculate and isolate the structure deformation and strain caused by temperature change, no mechanical pre-tightening force is applied to the model in this simulation process. The solver calculates the displacement and strain of each node and element in the model according to the thermal expansion coefficient of the material and the applied temperature field. The complete three-dimensional thermal induced strain field of the bolt at each position due to thermal effects is output. From the three-dimensional thermal induced strain field data, the axial strain component of the bolt shank region is extracted, and the axial strain values of all nodes in the region are integrated or arithmetically averaged to obtain the virtual thermal strain value caused by thermal effects, which accurately quantifies the average axial strain of the bolt shank caused by thermal expansion and contraction under the current non-uniform temperature field.

[0037] Optionally, the decoupling operation based on the virtual thermal strain value on the preliminary strain value to obtain the compensated real mechanical strain value includes: The preliminary strain value is taken as the minuend, and the virtual thermal strain value is taken as the subtrahend to perform decoupling operation to separate the real mechanical strain value.

[0038] Specifically, within the elastic deformation range, the total strain of the bolt is the linear superposition result of the mechanical strain generated by the mechanical stress and the thermal strain caused by the temperature change. Through decoupling operation, the mechanical strain part is accurately separated from the mixed total strain. Based on the preliminary strain value and the virtual thermal strain value, the preliminary strain value is obtained according to the resonant frequency signal of the passive bolt monitoring sensor, and the preset temperature-frequency drift relationship is used for preliminary compensation, which represents the apparent total strain physically felt by the bolt rod, including the thermal strain component generated by the thermal expansion and contraction of the bolt structure itself and the mechanical strain component generated by the pre-tightening force; the virtual thermal strain value is the average strain of the bolt rod calculated by the thermal effect after applying the accurate two-dimensional temperature field cloud map as the boundary condition through the high-precision finite element analysis micro-simulation model. The preliminary strain value is taken as the minuend, and the virtual thermal strain value is taken as the subtrahend, and the two are subtracted to obtain the compensated real mechanical strain value. The operation can be represented by the following formula: , Wherein, is the final compensated real mechanical strain value, which directly reflects the current pre-tightening force state of the bolt; is the preliminary strain value obtained by preliminary temperature compensation, which represents the total apparent strain of the bolt; is the virtual thermal strain value calculated by the finite element analysis micro-simulation model. As shown in Figure 3 the strain value comparison curve before and after temperature compensation represents the strain value change before and after temperature compensation. The decoupled real mechanical strain value is stable and is not affected by temperature fluctuations, which verifies the effectiveness of the thermal strain separation.

[0039] Optionally, after obtaining the real mechanical strain value, it further includes: calculating the current pre-tightening force of the bolt according to the real mechanical strain value and the preset elastic modulus of the bolt; judging whether the current pre-tightening force of the bolt meets the preset thermal stability condition; if yes, taking the current calculated real mechanical strain value as the reference strain value; calculating the error between the reference strain value and the strain value obtained by the single temperature sensor compensation model; performing reverse fine tuning on the temperature-frequency drift relationship and the parameters of the finite element analysis micro-simulation model according to the error.

[0040] Specifically, after the temperature compensation of the single measurement is completed, the current preload of the bolt is calculated according to Hooke's law using the real mechanical strain value. The calculation formula is as follows: , wherein F is the calculated current preload of the bolt; E is the elastic modulus of the bolt material pre-calibrated and stored; A is the effective cross-sectional area of the bolt shank; and ε is the compensated real mechanical strain value obtained by decoupling operation. The current working condition is continuously monitored and judged whether it meets a preset thermal stability condition. The condition defines a time window in which the external temperature disturbance is minimal and the internal temperature distribution of the bolt connection tends to be uniform and constant. In this state, it can be reasonably assumed that the real mechanical preload of the bolt remains unchanged in a short time. If it is judged that the thermal stability condition is met, the real mechanical strain value calculated at this moment is marked and stored as a high-credibility reference strain value. A compensation model serving as a comparison reference, such as a model for linear compensation based on only a single temperature sensor reading, is called to process the original data collected at the same time and obtain a comparison strain value. By calculating the difference between the reference strain value and the comparison strain value, a quantified error is obtained. The error is used as a feedback signal for reverse fine-tuning.

[0041] Optionally, the judging whether the current preload of the bolt meets the preset thermal stability condition comprises: calculating the instantaneous change rate of the temperature data of the plurality of monitoring points and the maximum temperature difference therebetween; if the instantaneous change rate is lower than a first threshold value and the maximum temperature difference is lower than a second threshold value, it is determined that the thermal stability condition is met.

[0042] Specifically, the multi-point temperature data collected by the distributed temperature sensing network in continuous time series is utilized. The instantaneous change rate of temperature and the maximum temperature difference between monitoring points are evaluated in parallel. The instantaneous change rate of temperature represents the stability of the temperature field in the bolted connection area in the time dimension. For each monitoring point in the distributed temperature sensing network, the difference between the current temperature value and the temperature value at the last sampling time is calculated, and then divided by the sampling time interval to obtain the instantaneous temperature change rate of the point. By traversing all monitoring points, the maximum absolute value in these change rates is found as the representative instantaneous change rate of the entire area. The maximum temperature difference between monitoring points represents the uniformity of the temperature field in the bolted connection area in the spatial dimension. At the same sampling time, the temperature data of all monitoring points is read, and the highest temperature value and the lowest temperature value are found from them. The difference between the two is calculated to obtain the maximum temperature difference at the current time. It directly reflects the degree of non-uniformity or the size of the temperature gradient of the temperature field in space. The instantaneous change rate of temperature and the maximum temperature difference between monitoring points are compared with two pre-set threshold values. The first threshold value is the upper limit for the instantaneous change rate, and the second threshold value is the upper limit for the maximum temperature difference. These two threshold values are pre-determined through experimental calibration or theoretical analysis according to the thermal characteristics, material properties and measurement accuracy requirements of the specific application scenario. Only when the maximum instantaneous change rate is lower than the first threshold value and the maximum temperature difference is also lower than the second threshold value, it is determined that the current working condition meets the pre-set thermal stability condition. If either condition is not met, it is considered to be in a thermal non-stable state.

[0043] Optionally, the step of adjusting the parameters of the temperature-frequency drift relationship or the finite element analysis micro-simulation model according to the error comprises: The least squares method is used to minimize the error, and the coefficients in the temperature-frequency drift relationship and the parameters of the finite element analysis micro-simulation model are iteratively optimized.

[0044] Specifically, the least squares method is used to minimize the error calculated under the thermal stability condition, and the temperature-frequency drift relationship and the finite element analysis micro-simulation model are iteratively optimized. When the thermal stability condition is met, the high-credibility reference strain value is obtained at the same time, and the comparative strain value is also obtained. The difference between the two constitutes the error that needs to be minimized. The goal of the least squares method is to find a set of optimal model parameters, so that the error square sum between the model output calculated by these parameters and the observation target reaches the minimum. The objective function of this optimization process can be represented as: , Where J(P) is the objective function to be minimized, i.e. the sum of squared errors; P represents a vector containing all the model parameters to be optimized; δ(P) is the error between the strain values calculated by the full compensation process and the reference strain values under the condition of parameters P, and the summation symbol Σ represents the accumulation calculation of multiple data points collected during the thermal stabilization period to enhance the robustness of optimization. The parameters P to be optimized specifically include two parts. The first part is the empirical coefficient in the preset temperature-frequency drift relationship. For example, if the relationship is a polynomial fitting, then P contains the coefficients of each term of the polynomial. The second part is the key material property parameters in the finite element analysis micro-simulation model, such as the thermal expansion coefficients and elastic moduli of the bolts and connected parts. Optimization algorithms such as gradient descent or Newton's method will start an iterative loop, in each iteration, the algorithm will fine-tune the parameters P and perform a complete strain calculation process again to get a new error δ(P). By calculating the gradient of the objective function J(P) with respect to each parameter, the algorithm can determine the parameter adjustment direction and step size that makes J(P) decrease the fastest, and update the parameters. This process is repeated until the value of the objective function J(P) converges to a preset number of iterations, at which time the obtained parameters P are the current optimal solution, and the original model is updated using the optimized parameters.

[0045] Based on the same inventive concept, as shown in Figure 4 The present application also provides a compensation system for temperature influence in a passive bolt monitoring system, the system comprising: A multi-source acquisition module for acquiring the resonant frequency signal of the passive bolt monitoring sensor and the temperature data of multiple monitoring points collected through a preset distributed temperature sensing network; A three-dimensional temperature field reconstruction module for reconstructing a two-dimensional temperature field cloud map of the bolt connection area by interpolation calculation according to the temperature data of the multiple monitoring points; A temperature-frequency collaborative processing module for extracting an equivalent comprehensive temperature rise value from the temperature field distribution, performing preliminary compensation on the resonant frequency signal according to a preset temperature-frequency drift relationship, and generating a preliminary strain value; A virtual thermal strain simulation module for inputting the equivalent comprehensive temperature rise value into a preset finite element analysis micro-simulation model to calculate a virtual thermal strain value caused by temperature influence; A mechanical strain decoupling module for performing decoupling operation on the preliminary strain value based on the virtual thermal strain value to obtain a compensated real mechanical strain value.

[0046] Embodiment 1: In order to verify the feasibility of the application in implementation, the application is applied to the tower foundation bolt connection monitoring of a certain large land-based wind turbine generator. The tower drum and the foundation of the wind turbine generator are connected by high-strength bolt flange, and the stability of the bolt pretightening force is crucial to the structural safety of the unit. However, the temperature of the unit in the wild environment changes dramatically, and the sunshine temperature difference and seasonal temperature difference will cause great error in the bolt monitoring data. The traditional single-point temperature compensation method is difficult to accurately separate the thermal effect, and often causes false alarm of pretightening force loosening.

[0047] In this embodiment, the passive bolt monitoring system described in the application is deployed on a group of key tower foundation bolts. It includes passive bolt monitoring sensors installed on the bolt head, and a distributed temperature sensing network uniformly arranged along the flange. A wireless reader / writer is installed on the inner wall of the tower drum for synchronous acquisition of sensing data, and the temperature compensation method described in the application is executed through a back-end computing platform.

[0048] In order to verify the beneficial effects of the application, the wind turbine generator is continuously monitored for six months, and data under various working conditions such as from winter low temperature to summer high temperature and intra-day temperature fluctuation are recorded. The following is the data analysis and effect verification results under certain working conditions.

[0049] On July 15, 2023, a sunny summer day, from 10:00 to 14:00, the sun's direct radiation caused the temperature of the flange to rise sharply on the sunny side, while the temperature of the shady side was relatively low, forming a non-uniform temperature field. By synchronously transmitting inquiry signals with the wireless reader / writer, the resonant frequency signals of the bolt monitoring sensors were obtained, and the temperature data of each node in the distributed temperature sensing network were synchronously read.

[0050] According to the temperature data collected from multiple monitoring points, the two-dimensional temperature field cloud map covering the bolt head, nut and flange surface is successfully reconstructed by using the inverse distance weighted interpolation algorithm. The cloud map clearly shows that the temperature on the sunny side reaches 55℃, while the temperature on the shady side is only 32℃. Then the bolt shank area is identified from the temperature field cloud map, and the equivalent comprehensive temperature rise value of 18.5℃ is calculated by the weighted average rule based on distance and heat capacity. This temperature rise value is used to preliminarily compensate the resonant frequency signal to generate a preliminary strain value.

[0051] At the same time, the equivalent comprehensive temperature rise value and the temperature gradient information extracted from the cloud map are loaded as thermal load into the pre-set finite element analysis micro-simulation model. The model accurately simulates the geometry and material properties of the bolt, nut and flange. The calculated virtual thermal strain value caused by the non-uniform temperature field is +35με (microstrain). Finally, the virtual thermal strain value is subtracted from the preliminary strain value to obtain the compensated real mechanical strain value.

[0052] In addition, at 4:00 on August 20, 2023, it was monitored that the working condition met the preset thermal stability condition: the instantaneous change rate of the temperature of each measuring point was less than 0.1℃ / min, and the maximum temperature difference between each other was less than 0.5℃. It was determined that the thermal stability condition was met at this moment, and the real mechanical strain value calculated at the moment was taken as the high-credibility reference strain value. By comparing the reference value with the value obtained by the single temperature sensor compensation model, it was calculated that there was an error of 4με between the two. Then, by using the least square method, the coefficients in the temperature-frequency drift relationship and the material thermal expansion coefficient of the finite element model were inversely fine-tuned to minimize the error, and the online self-calibration was completed.

[0053] The data show that the method can effectively overcome the interference of the complex temperature environment. Under the sunshine temperature difference working condition on July 15, the traditional single-point temperature compensation method incorrectly interprets the thermal expansion as a decrease in pre-tightening force, and calculates that the bolt pre-tightening force decreases by 12%, triggering a false alarm. However, the present application successfully avoids misjudgment by accurately reconstructing the two-dimensional temperature field and decoupling the finite element simulation, and the real mechanical strain value calculated by the present application changes little, indicating that the bolt pre-tightening force remains stable. The online self-calibration function also ensures that the accuracy of long-term monitoring will not be attenuated due to equipment aging or environmental changes.

[0054] Table 1 Comparison table of wind turbine generator set bolt monitoring data .

[0055] Table 2 Online adaptive calibration effect data table .

[0056] The above table 1 and table 2 record the actual application data of the present application in the monitoring of the wind turbine generator set bolt, and details the performance of the present application in high-precision compensation and adaptive optimization.

[0057] The data in table 1 show that the advantage of the present application is greater when facing a severe non-uniform temperature field. The traditional method cannot capture the temperature gradient, resulting in a serious lack of evaluation of the thermal effect, thereby producing the illusion of a significant decrease in pre-tightening force. However, the present application accurately calculates and strips the virtual thermal strain of +35με by reconstructing the two-dimensional temperature field and finite element simulation, and the final real mechanical strain value shows that the bolt state is stable, effectively avoiding the waste of maintenance resources and unnecessary shutdown inspection.

[0058] The data in Table 2 proves the self-learning and self-optimization capability of the present application. The ideal calibration opportunity, i.e. the thermal stable condition, can be intelligently identified in actual operation, and the result of the high-precision model is used as a reference to inversely fine-tune the core algorithm parameters. After calibration, the error is significantly reduced, and the material parameters of the finite element model are closer to the actual physical characteristics. This online closed-loop calibration mechanism ensures accuracy and reliability throughout the entire life cycle.

[0059] It should be noted that the electrical connection between the above-mentioned various units does not necessarily mean direct connection of the line, indirect connection mode, as long as the purpose of the present application is achieved, which is applicable to the embodiments of the present application. The above-described is only an exemplary embodiment of the present application, and cannot limit the scope of the present application.

[0060] That is, any equivalent changes and modifications made in accordance with the teachings of the present application are still within the scope of the present application. Other embodiments of the present application will be readily apparent to those skilled in the art upon considering the description and practice of the disclosure. The present application is intended to cover any variations, uses, or adaptive changes to the present application that follow the general principles of the present application and include common knowledge or conventional technical means in the art not described in the present application.

Claims

1. A method for compensating for temperature effects in a passive bolt monitoring system, characterized in that, include: Acquire the resonant frequency signal of the passive bolt monitoring sensor and the temperature data of multiple monitoring points collected through a preset distributed temperature sensing network; Based on the temperature data from the multiple monitoring points, a two-dimensional temperature field cloud map of the bolted connection area is reconstructed through interpolation calculation; The equivalent comprehensive temperature rise value is extracted from the temperature field distribution, and the resonant frequency signal is initially compensated according to the preset temperature-frequency drift relationship to generate the initial strain value; The equivalent comprehensive temperature rise value is input into a preset finite element analysis micro-simulation model to calculate the virtual thermal strain value caused by temperature influence. Based on the virtual thermal strain value, the preliminary strain value is decoupled to obtain the compensated real mechanical strain value.

2. The compensation method for temperature influence in a passive bolt monitoring system according to claim 1, characterized in that, The acquisition of the resonant frequency signal of the passive bolt monitoring sensor and the temperature data of multiple monitoring points collected through a preset distributed temperature sensing network includes: The wireless reader is controlled to synchronously transmit interrogation signals to the passive bolt monitoring sensor and the distributed temperature sensing network; The system synchronously receives and demodulates the return signal from the passive bolt monitoring sensor to obtain the resonant frequency signal, and synchronously reads the measured values ​​of each node in the distributed temperature sensing network to obtain temperature data.

3. The compensation method for temperature influence in a passive bolt monitoring system according to claim 1, characterized in that, The step of reconstructing a two-dimensional temperature field cloud map of the bolted connection area based on the temperature data from the multiple monitoring points through interpolation calculation includes: Obtain the position coordinates of each sensor in the distributed temperature sensing network and its corresponding temperature readings; Based on the location coordinates and temperature readings, a two-dimensional temperature field cloud map covering the bolt head end face, nut end face and the surface of the connected parts is generated by an inverse distance weighted interpolation algorithm.

4. The method for compensating for temperature effects in a passive bolt monitoring system according to claim 1, characterized in that, Extracting the equivalent comprehensive temperature rise value from the temperature field distribution includes: Based on the two-dimensional temperature field cloud map, identify the temperature distribution in the bolt shank region; According to the preset weighting coefficient calculation rules, the temperature of the bolt shank region is weighted and averaged. The weighting coefficient is determined based on the distance between each pixel and the center of the bolt shank and the heat capacity of the component. The equivalent comprehensive temperature rise value is then output.

5. The method for compensating for temperature effects in a passive bolt monitoring system according to claim 1, characterized in that, The step of inputting the equivalent comprehensive temperature rise value into a preset finite element analysis micro-simulation model to calculate the virtual thermal strain value caused by temperature influence includes: Calculate the ratio of temperature change to spatial change in a two-dimensional temperature field cloud map to obtain temperature gradient information; The equivalent comprehensive temperature rise and temperature gradient information are used as boundary conditions to load the thermally induced strain field of the bolts into the preset finite element analysis micro-simulation model. The average strain value of the bolt shank is extracted from the thermally induced strain field to generate a virtual thermal strain value.

6. The method for compensating for temperature effects in a passive bolt monitoring system according to claim 1, characterized in that, The step of decoupling the preliminary strain value based on the virtual thermal strain value to obtain the compensated true mechanical strain value includes: The preliminary strain value is used as the minuend and the virtual thermal strain value is used as the subtrahend for decoupling operations to separate the real mechanical strain value.

7. The method for compensating for temperature effects in a passive bolt monitoring system according to claim 1, characterized in that, After obtaining the actual mechanical strain value, the process also includes: Calculate the current preload of the bolt based on the actual mechanical strain value and the preset elastic modulus of the bolt; Determine whether the current preload of the bolt meets the preset thermal stability conditions; If the conditions are met, the currently calculated actual mechanical strain value will be used as the reference strain value. Calculate the error between the reference strain value and the strain value obtained through the single temperature sensor compensation model; Based on the error, the temperature-frequency drift relationship and the parameters of the finite element analysis micro-simulation model are fine-tuned in reverse.

8. The method for compensating for temperature effects in a passive bolt monitoring system according to claim 7, characterized in that, The determination of whether the current preload of the bolt meets the preset thermal stability condition includes: Calculate the instantaneous rate of change of temperature data at the multiple monitoring points and the maximum temperature difference between them; If the instantaneous rate of change is lower than the first threshold and the maximum temperature difference is lower than the second threshold, then the thermal stability condition is determined to be met.

9. The method for compensating for temperature effects in a passive bolt monitoring system according to claim 7, characterized in that, The step of fine-tuning the temperature-frequency drift relationship and the parameters of the finite element analysis micro-simulation model based on the error includes: Using the least squares method, with the goal of minimizing the error, the coefficients in the temperature-frequency drift relationship and the parameters of the finite element analysis micro-simulation model are iteratively optimized.

10. A temperature-effect compensation system for a passive bolt monitoring system, applied to a temperature-effect compensation method for a passive bolt monitoring system as described in any one of claims 1-9, characterized in that, The system includes: The multi-source acquisition module is used to acquire the resonant frequency signal of the passive bolt monitoring sensor and the temperature data of multiple monitoring points collected through a preset distributed temperature sensing network. The three-dimensional temperature field reconstruction module is used to reconstruct a two-dimensional temperature field cloud map of the bolt connection area based on the temperature data of the multiple monitoring points through interpolation calculation; The temperature-frequency co-processing module is used to extract the equivalent comprehensive temperature rise value from the temperature field distribution, perform preliminary compensation on the resonant frequency signal according to the preset temperature-frequency drift relationship, and generate preliminary strain value. The virtual thermal strain simulation module is used to input the equivalent comprehensive temperature rise value into a preset finite element analysis micro-simulation model and calculate the virtual thermal strain value generated by the temperature effect. The mechanical strain decoupling module is used to perform decoupling calculations on the preliminary strain value based on the virtual thermal strain value to obtain the compensated real mechanical strain value.

Images