A BIM-based construction method for large-span post-tensioned bonded prestressed beams

Abstract

This invention discloses a BIM-based construction method for large-span post-tensioned bonded prestressed beams, comprising: acquiring vibration signals of the working reinforcement and reference signals of the doubling reinforcement; inputting the doubling reference signals into an adaptive filtering algorithm with phase-preserving constraints to perform noise cancellation processing on the vibration signals of the working reinforcement, generating purified pressure wave signals; performing multi-scale decomposition on the purified signals, and extracting the propagation delay of the direct wave based on coherence analysis between components, generating a phase-delay feature matrix; calculating the real-time stress distribution field based on the geometric parameters of a pre-configured BIM 3D model, and synchronously updating it to the BIM 3D model. This invention solves the phase distortion problem under noise interference through physical reference channels and phase-preserving filtering, and improves the accuracy of time delay extraction through multi-path separation technology, achieving high-precision and robust real-time monitoring of prestress.

Description

Technical Field

[0001] This invention belongs to the field of civil engineering construction monitoring, and in particular, it is a BIM-based construction method for large-span post-tensioned bonded prestressed beams. Background Technology

[0002] Post-tensioned bonded prestressing technology is a core process for realizing large-span prestressed concrete structures, and its construction quality directly determines the long-term safety and service life of the structure. During the post-tensioning process, the prestressing tendons experience significant prestress loss from the tensioning end to the anchorage end due to factors such as friction with the duct wall and deviations in the prestressing tendon path, resulting in a deviation between the actual effective prestress distribution along the line and the design value. If this deviation exceeds the allowable range, it may affect the normal function of the structure, causing excessive deflection or cracks, or even induce structural damage. Therefore, real-time and accurate monitoring and feedback control of the actual stress distribution of the prestressing tendons during the construction phase is a crucial link in ensuring the construction quality and fundamental safety of large-span structures, and has engineering significance and technical value.

[0003] Currently, monitoring of post-tensioned prestressed concrete construction primarily relies on a combination of theoretical calculations and on-site measurements. At the theoretical calculation level, engineering practice generally uses pre-defined friction coefficients and duct deviation coefficients based on relevant specifications to estimate prestressing friction loss. At the on-site measurement level, mainstream techniques include installing pressure sensors at the tensioning and anchoring ends to monitor the tension force of the jacks and the retraction of the anchorages, and attaching strain gauges to the concrete surface of key beam sections to indirectly calculate the stress state of the prestressing tendons by measuring the concrete strain. Some research has begun to explore dynamic non-destructive testing techniques based on wave propagation principles, applying excitation to one end of the prestressing tendon and measuring its response signal at the other end to analyze the stress state. In terms of data processing, the collected data is mostly processed offline, and stress conversion relies on simplified linear elastic material assumptions. Regarding construction control, open-loop control is mainly based on a pre-defined tensioning scheme, using dual control indicators of tension force and elongation to determine whether tensioning is in place.

[0004] However, existing technologies still face several specific and profound technical bottlenecks when dealing with dynamic and nonlinear problems in complex construction sites. These bottlenecks mainly focus on the accurate perception of signal-level information and the accurate inversion of physical-level states. In summary, the core challenges faced by existing technologies are: the inability to preserve signal phase information, which is crucial for accurate measurement, under strong noise interference, and the difficulty in effectively distinguishing direct wave paths carrying core time delay information in complex wave fields. Summary of the Invention

[0005] The purpose of this invention is to provide a BIM-based construction method for large-span post-tensioned bonded prestressed beams to solve the aforementioned problems in the existing technology.

[0006] Technical solution: A BIM-based construction method for large-span post-tensioned bonded prestressed beams, comprising:

[0007] Collect the vibration signal of the working rib and the reference signal of the dummy rib;

[0008] The noise cancellation process of the working rib vibration signal is performed using the dumb rib reference signal to generate a purification pressure wave signal.

[0009] Analyze the purification pressure wave signal, extract the propagation delay, and generate a phase-delay feature matrix;

[0010] Based on the phase-delay feature matrix and the geometric parameters of the pre-configured BIM 3D model, the real-time stress distribution field is calculated by inversion and updated to the BIM 3D model.

[0011] Beneficial effects: This invention solves the phase distortion problem under noise interference by using a physical reference channel and phase-preserving filtering, and improves the accuracy of time delay extraction through multi-path separation technology, thus achieving high-precision and robust real-time monitoring of prestress. Attached Figure Description

[0012] Figure 1 This application provides a flowchart of the construction method for a large-span post-tensioned bonded prestressed beam based on BIM, which is an embodiment of the present application.

[0013] Figure 2 A flowchart illustrating the steps for applying phase-preserving constraints as provided in an embodiment of this application.

[0014] Figure 3 A flowchart illustrating the steps involved in determining the step size, as provided in an embodiment of this application.

[0015] Figure 4 A flowchart illustrating the steps for generating an adaptive sweep frequency excitation signal, as provided in an embodiment of this application. Detailed Implementation

[0016] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0017] It should be noted that the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion, for example, a process, method, system, product, or device that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or units that are not explicitly listed or that are inherent to such process, method, product, or device.

[0018] The study revealed that phase information is distorted during noise filtering. Dynamic measurement methods based on wave propagation are highly dependent on the accurate determination of wave packet propagation delay, which in turn depends on the precise analysis of the instantaneous phase of the signal. Construction sites are filled with strong common-mode noise generated by mechanical vibrations and electromagnetic equipment, necessitating filtering of the acquired signals. However, traditional filtering algorithms (such as standard LMS adaptive filtering or bandpass filtering) are designed to minimize signal amplitude errors. While filtering noise, they inevitably introduce nonlinear disturbances to the signal's phase spectrum, resulting in phase distortion. This distortion undermines the physical basis of subsequent delay calculations, creating a fundamental technical contradiction: filtering operations aimed at improving the signal-to-noise ratio precisely damage the most critical information required for measurement, preventing a breakthrough in measurement accuracy. Furthermore, there is the problem of interference from multi-path propagation in complex structures. When the excitation wave propagates within the complex acoustic waveguide composed of prestressed tendons, grouting, and corrugated pipes, in addition to the direct wave propagating along the shortest path, a large number of reflected waves are generated through interface reflection and scattering. These multipath signals are superimposed at the receiving end, causing the received waveform to exhibit multiple complex peaks. The simple time-domain peak detection method is very likely to misjudge the reflected wave, which may have stronger energy, as the first wave (i.e., the direct wave), thus causing serious errors in the calculation of propagation delay, which are directly transmitted to the final stress inversion result, resulting in a misjudgment of the structural state.

[0019] like Figure 1 As shown, a BIM-based construction method for large-span post-tensioned bonded prestressed beams is proposed, including the following steps:

[0020] The vibration signal of the working rib and the reference signal of the dormant rib are collected.

[0021] In this embodiment, the working reinforcement refers to the prestressed tendon that actually bears the tensile force in the prestressed structure and serves as the wave propagation medium. The dormant reinforcement refers to a prestressed tendon that is identical or substantially identical to the working reinforcement in terms of path, material, and diameter, but is intentionally not subjected to tensile force. Its function is to serve as a physical reference channel for measuring the environmental noise and vibration interference shared by the working reinforcement at the construction site, i.e., common-mode noise. In other words, the working reinforcement refers to the reinforcing steel that participates in the structural stress, such as the main reinforcement, distribution reinforcement, and stirrups in structural components such as beams, slabs, columns, and walls. The dormant reinforcement refers to the reinforcing steel that does not participate in the direct stress, such as the negative moment reinforcement in slabs, the stirrup reinforcement in the web of beams, and the structural reinforcement in columns.

[0022] Specifically, this step involves optimizing sensor placement based on BIM data. For example, by reading the .ifc format file of the Building Information Model (BIM), the three-dimensional spatial curve equation P(s)=[x(s), y(s), z(s)] of the prestressing tendons and the geometric properties of the beam are obtained, where (x, y, z) are the three-dimensional spatial coordinates and s is the arc length parameter on the prestressing tendon path. Based on this, a curvature analysis algorithm can be used to identify potential stress concentration areas on the prestressing tendon path, such as locations where the curvature κ is greater than a certain threshold (e.g., 0.02 / m), and accelerometers can be preferentially placed at these key locations, as well as representative locations such as beam endpoints and spans. To enable the excitation signal to actively avoid strong noise frequencies at the construction site and improve the signal-to-noise ratio, a period of background noise signal (e.g., 30 seconds) can be collected using sensors before formal data acquisition. Fast Fourier Transform (FFT) analysis is performed on this noise signal to obtain the noise power spectrum N(f). Based on this power spectrum, an adaptive frequency sweep excitation signal can be generated. The generation logic is as follows: in frequency bands with large noise power spectrum amplitudes, the sweep rate is increased; in clean frequency bands with low noise, the sweep rate is slowed down, thereby injecting more signal energy into the high signal-to-noise ratio frequency band. The generated adaptive sweep excitation signal is applied to the jack or anchor at the tensioning end through a piezoelectric exciter; simultaneously, vibration signals x from multiple measuring points on the working rib are synchronously acquired at a high sampling rate (e.g., 10kHz). i (t) and the reference signal r at the corresponding position on the dumb rib. i (t). These collected geometric data, time-domain signals, noise spectra, etc., together constitute a structured multi-source dataset, providing input for subsequent processing.

[0023] The noise cancellation of the working rib vibration signal is performed using the reference signal of the dumb rib to generate a purified pressure wave signal.

[0024] In this embodiment, the purified pressure wave signal refers to the signal that primarily reflects the propagation characteristics of the excitation signal along the prestressed tendon after removing common-mode noise components from the original working tendon vibration signal. Specifically, an adaptive filtering algorithm is used, with the reference signal r(t) from the dormant tendon as the input to the adaptive filter. The filter continuously adjusts its weights to generate the best estimate of the common-mode noise. This estimate is subtracted from the working tendon signal x(t) to obtain the purified signal y(t). By combining differential measurement and adaptive processing, the traditional noise processing paradigm is changed, transforming noise from estimation to measurement, thus improving the accuracy of noise cancellation. Noise interference can be eliminated to the maximum extent without damaging the phase information of the useful signal.

[0025] Analyze the purification pressure wave signal, extract the propagation delay, and generate a phase-delay feature matrix.

[0026] In this embodiment, propagation delay refers to the time required for the purification pressure wave signal to propagate between two different measuring points on the working rib. The phase-delay characteristic matrix is ​​a matrix T=[τ] that records the propagation delay between all pairs of measuring points. ij ], where τ ij This represents the time delay of the signal propagating from measurement point i to measurement point j. Specifically, phase-locked packet tracking analysis is employed. Since the signal-to-noise ratio of the purified signal has been improved, its instantaneous phase can be accurately extracted using methods such as Hilbert transform. By analyzing the gradient of the phase with frequency (i.e., group delay), the precise propagation time of the wave packet between different measurement points can be calculated. To address the reflection and multipath problems that may occur when waves propagate in structures, multipath separation algorithms can also be included. For example, based on the coherence differences of signals from different paths, direct waves and reflected waves can be distinguished, and multipath information can be fused to obtain a robust propagation delay estimate.

[0027] Based on the phase-delay feature matrix and the geometric parameters of the pre-configured BIM 3D model, the real-time stress distribution field is calculated by inversion, and the real-time stress distribution field is updated to the BIM 3D model.

[0028] In this embodiment, the real-time stress distribution field refers to the continuous function or discrete vector σ(s,t) of the stress magnitude along the entire working rib path. Specifically, an inversion model is established based on the wave velocity-stress relationship. The propagation speed (wave velocity) of sound waves in a solid medium is related to the stress on the medium; generally, the greater the stress, the faster the wave velocity. Using this physical principle, a system of equations can be established, where each equation represents the relationship between a pair of measuring points: τ ij =∫ pathds / v(σ, T) represents the measured time delay, which is equal to the integral of the path length divided by the wave velocity. The wave velocity model v(σ, T) here preferably considers the coupling effect of stress σ and temperature T, exhibiting a nonlinear relationship. By solving this inversion equation set, the stress values ​​at each point on the prestressing tendon can be obtained. The calculated real-time stress distribution data is then mapped back into the BIM model, updating the corresponding attributes of the prestressing tendon elements (e.g., creating an attribute set named Pset_TendonStress), thus achieving digital twin visualization and recording of the construction status.

[0029] According to one aspect of this application, the acquisition of the dumb rib reference signal originates from a physical dumb rib that is identical to the working rib in path and material but is not subjected to tension. The physical dumb rib is specifically used as a noise reference channel to measure common-mode noise. The noise cancellation process specifically involves: inputting the dumb rib reference signal into an adaptive filtering algorithm to generate a dynamic estimate of the common-mode noise, and subtracting the dynamic estimate from the vibration signal of the working rib to generate a purified pressure wave signal.

[0030] In this embodiment, the motivation for setting up the dormant reinforcement is that, theoretically, the impact of common-mode noise such as environmental vibration and electromagnetic interference at the construction site on the working reinforcement and the dormant reinforcement is the same. Since the dormant reinforcement does not bear tensile force, the main component of its acquired signal r(t) is the common-mode noise n. common (t) and a portion of random noise n random_ref (t). The signal x(t) of the working rib is the propagating pressure wave signal s. prop (t), common-mode noise nc ommon (t) and random noise n random The superposition of (t). Therefore, the dummy signal r(t) provides the adaptive filter with a high-quality reference input that is highly correlated with the noise to be eliminated.

[0031] According to one aspect of this application, the adaptive filtering algorithm controls the generation of dynamic estimates by applying a phase-preserving constraint; specifically, the phase-preserving constraint minimizes the instantaneous phase difference between the working rib vibration signal and the finally generated purification pressure wave signal.

[0032] The objective function of traditional adaptive filtering algorithms (such as the LMS algorithm) is usually to minimize the mean square value of the filtering error, i.e., minE[|e(n)|]. 2The error e(n) is the difference between the desired signal and the filtered output. This optimization only focuses on the amplitude error, which often causes unpredictable phase shifts while eliminating noise, and phase information is crucial for subsequent time delay calculations. In this embodiment, to solve this problem, the objective function of the traditional algorithm is improved by introducing a phase preservation constraint. That is, when the algorithm iteratively updates its filtering weights, it must not only minimize the amplitude error, but also satisfy a constraint that the instantaneous phase of the processed purified signal y(t) should be as close as possible to the instantaneous phase of the original working signal x(t). This constitutes a dual-constraint optimization problem, which allows the phase characteristics of the signal to be preserved while reducing noise.

[0033] like Figure 2 As shown, according to one aspect of this application, the application of phase-preserving constraints specifically includes the following data processing steps:

[0034] The instantaneous phase difference is obtained by calculating the instantaneous phase difference between the working rib vibration signal and the purification pressure wave signal using Hilbert transform.

[0035] Based on the instantaneous phase difference, a phase preservation factor is constructed in the form of a complex exponential function of the phase difference;

[0036] The phase preservation factor is introduced as a multiplication term into the update rule of the time-varying filter weights in the adaptive filtering algorithm.

[0037] Specifically, Hilbert transforms are performed on the working rib signal x(t) and the current step's purified signal (filtered output) y(t) to obtain their analytic signals x. a (t) and y a (t). The argument of an analytic signal is its instantaneous phase. For example, Φ x (t)=angle(x a (t)) and Φ y (t)=angle(y a (t)), where Φ x (t) represents the instantaneous phase (argument) corresponding to the working rib signal, Φ y (t) represents the instantaneous phase corresponding to the current purification signal. Therefore, the instantaneous phase difference ΔΦ(t) = Φ x (t)-Φ y(t). To suppress endpoint effects, a window function (such as a Tukey window) can be applied to the signal before transformation. In some embodiments, to suppress endpoint effects generated when transforming finite-length signals, it is preferable to apply a Tukey window to the signal before transformation. The roll-off coefficient r of the Tukey window can be selected according to the transient characteristics of the signal. For example, for signals with strong impulses, a smaller r value (such as 0.1) can be selected, and for relatively stable signals, a larger r value (such as 0.25) can be selected to achieve a balance between suppressing endpoint effects and preserving the signal waveform. After transformation, the argument of the analytic signal is the instantaneous phase. Based on the calculated instantaneous phase difference, a complex phase preservation factor P(t) = exp(j·ΔΦ(t)) can be constructed, where j is the imaginary unit. This factor has a modulus of 1, and its phase directly reflects the phase difference between the two signals.

[0038] Furthermore, to increase robustness, a confidence weight α(t) can be introduced, so that the effect of this factor is weakened when the phase difference fluctuates drastically. For example, α(t) = exp(-σ ΔΦ (t) / σ0), where σ ΔΦ (t) is the standard deviation of the phase difference within the local window, and σ0 is the standardization coefficient. At this point, the phase preservation factor is P(t) = α(t)·exp(j·ΔΦ(t)). The constructed phase preservation factor P(n) (discrete form) is multiplied into the weight update formula of the adaptive filter. For example, for the LMS algorithm, its traditional update rule is w(n+1) = w(n) + μ·e(n)·r(n), where w(n) is the filter weight vector at step n, w(n+1) is the updated filter weight vector at step n+1, μ is the learning rate or step size coefficient, e(n) is the error signal at step n, and r(n) is the reference signal sequence from the dummy signal. With the introduction of phase-preserving constraints, the update rule can be improved to w(n+1) = w(n) + μ(n)·e(n)·r(n)·P(n), where μ(n) is the time-varying step size used in the nth iteration, and P(n) is the phase-preserving factor matrix used in the nth iteration. Alternatively, in a more complex dual-constraint model, the update rule can be decomposed into the expression for the magnitude error e. amp and phase error e phase The common response, such as w(n+1)=w(n)+μ(n)·[e amp (n)·r(n)+λ·e phase [(n)·r(n)ΘP(n)], where Θ represents element-to-element multiplication and λ is the phase constraint strength coefficient, used to balance the proportion of amplitude and phase error in weight update.

[0039] like Figure 3As shown, according to one aspect of this application, the adaptive filtering algorithm uses an adaptive step size to update its time-varying filtering weights, and the process of determining the step size specifically includes:

[0040] The variance of the filtering error sequence generated by the adaptive filtering algorithm is calculated in real time to obtain the error variance value;

[0041] Based on the preset inverse function relationship, the value of the adaptive step size is determined by the error variance value, so that the step size is inversely proportional to the error variance.

[0042] In this embodiment, the motivation for using an adaptive step size (or variable step size) is that the noise intensity at the construction site is not constant. When the noise suddenly increases, the filtering error will increase, requiring a smaller step size to ensure stable convergence of the algorithm; when the environment is relatively quiet and the filtering error is small, a larger step size can be used to accelerate the convergence speed and tracking capability. The specific implementation is as follows: In each iteration, the variance var(e(nL:n)) of the filtering error e(n) of the most recent L samples is calculated. The step size μ(n) is updated according to the inverse function relationship. An exemplary relationship is μ(n) = μ0 / [1 + γ·var(e(nL:n))], where μ0 is the initial or maximum step size, and γ is a positive constant used to control the sensitivity of the step size to variance changes. Optionally, to further improve environmental adaptability, the step size calculation can also consider the power spectrum of the current noise. For example, an environmental adjustment factor β(n) = 1 + γ′·mean(N(f current ±Δf)), where N(f) current ) represents the noise power near the main frequency of the current signal, γ′ is the noise sensitivity factor, and Δf is the frequency offset window width. This factor is then multiplied into the step size calculation formula, such as μ(n) = μ0·β(n) / [1 + var(n) / var0], where var0 is the baseline variance. The step size adjustment considers both the filter's output error and the input noise characteristics of the external environment, making the adaptive process more intelligent and efficient.

[0043] In the preferred scheme for calculating error variance using a multi-scale sliding window, the selection of the multi-scale sliding window length (e.g., L1=50, L2=200, L3=1000 samples) is based on the fact that different time scales can capture different dynamic characteristics of the error sequence. The short-scale window (L1) is sensitive to error fluctuations caused by sudden strong disturbances or transient signal changes; the medium-scale window (L2) reflects the algorithm's convergence performance under normal conditions; and the long-scale window (L3) characterizes the algorithm's performance in a quasi-steady state. The weighting coefficients (e.g., 0.5, 0.3, 0.2) are determined based on engineering experience or offline simulation experiments, with the basic principle being to assign higher weights to the short-scale error variance. This is because in real-time adaptive processes, a rapid response to recent errors is usually more important than consideration of the long-term steady state, helping the algorithm recover from unstable states more quickly.

[0044] This embodiment solves the problem of phase information distortion during noise filtering by introducing a physical dormant rib parallel to the working rib as a noise reference channel and applying phase-preserving constraints in the adaptive filtering algorithm. On the one hand, the dormant rib transforms the uncertain common-mode noise, which needs to be estimated by the algorithm, into a physical quantity that can be measured with high fidelity, providing an extremely accurate reference input for adaptive subtraction. On the other hand, the phase-preserving constraint introduces a penalty term based on instantaneous phase difference into the filter's weight update rule, forcing the algorithm to maintain the consistency of the phase characteristics of the purified signal with the original signal while minimizing amplitude error. This enables the system to effectively filter common-mode noise in strong interference environments such as vibration and electromagnetic interference at construction sites, while ensuring that the signal phase carrying the core propagation delay information is not distorted, providing a high-quality, high-fidelity input signal for subsequent high-precision stress inversion.

[0045] In a preferred embodiment, to more rigorously solve the dual-constraint problem involving both amplitude minimization and phase preservation, an optimization framework based on the Alternating Direction Multiplier Method (ADMM) can be employed. Specifically, the process involves constructing a Lagrangian objective function L encompassing all optimization objectives: L = ||xw⊕r|| 2 +λ1•(angle(y)-angle(x)) 2 +λ2•∣∣▽w∣∣ 2The algorithm consists of three terms: the first term is the traditional amplitude error term; the second term is the phase preservation constraint term, which aims to minimize the phase difference between the purified signal y and the original signal x; and the third term is the weight smoothing constraint, used to prevent drastic oscillations in the filter weights w and improve algorithm stability. r is the reference signal, ⊕ is the convolution operator, and ▽ is the gradient. λ1 and λ2 are two Lagrange multipliers, i.e., regularization parameters, used to balance the relative importance of the three objectives. Their initial values ​​can be set empirically, for example, both to 0.1, and updated based on the dual residuals during iteration. The ADMM algorithm decomposes the complex optimization problem into a series of easier-to-solve subproblems and solves them through alternating iterations. Each iteration cycle includes: fixing the phase term and other variables, solving for the optimal filter output that minimizes the amplitude error; fixing the amplitude term and other variables, solving for the optimal phase that minimizes the phase error; and updating the Lagrange multipliers. The ADMM iteration process continues until the algorithm meets the convergence criterion. A feasible convergence criterion is that when the norms of both the original residual and the dual residual are less than a preset small threshold (e.g., 10), the convergence criterion is achieved. -4 Alternatively, when the number of iterations reaches a preset maximum value (e.g., 10 iterations in the inner loop), the algorithm is considered to have converged, and the current optimal filtering result is output as the purified pressure wave signal.

[0046] In a specific embodiment, assuming that at a certain discrete time n=1, the following input is given: working rib signal x(1)=1.2, which can be understood as the real pressure wave signal s(1)=0.5 and common-mode noise n c (1) = 0.7 superposition; the dummy reference signal r(1) = 0.72, this value is relative to the common-mode noise n c (1) The measured value contains a small measurement error; the weight of the adaptive filter at the previous moment is w(1)=0.9; the parameters used in the algorithm are set as follows: initial step size μ0=0.1, step size adjustment sensitivity γ=0.5, phase constraint strength λ=0.2. The calculation process is as follows: Calculate the filter output: y(1)=w(1)·r(1)=0.9·0.72=0.648, which is the filter's estimate of common-mode noise. Calculate the error signal: filter error e(1)=x(1)-y(1)=1.2-0.648=0.552, which is the preliminary clean signal. Calculate the instantaneous phase and phase error (conceptual demonstration): to simplify the calculation, the actual Hilbert transform is not performed, but it is assumed that the following instantaneous phase value is obtained through the transformation: instantaneous phase of the working rib signal: Φ x (1) = 0.8π (radians); Instantaneous phase of the filter output: Φ y (1) = 0.75π (radians) phase error e phase (1)=Φ x (1)-Φ y(1) = 0.05π. Calculate the amplitude error: Amplitude error e amp (1) = |x(1)| - |y(1)| = 1.2 - 0.648 = 0.552 (In this simplified example, the amplitude error equals the filtering error); Calculate the adaptive step size μ(2): First, the error variance is needed. Assume that the error variance of the most recent L samples is calculated as var(e) = 0.04. According to the formula: μ(2) = μ0 / [1 + γ·var(e)] = 0.1 / [1 + 0.5·0.04] = 0.1 / 1.02 ≈ 0.098 It can be seen that due to the existence of error, the step size has been slightly reduced from the initial 0.1 to 0.098. Calculate the phase preservation factor P(1): Assume that the confidence weight α(1) = 0.95 is obtained through phase difference dynamic analysis. The phase difference ΔΦ(1) can be approximately understood as e phase (1). P(1)=α(1)·exp(j·e phase (1))=0.95·exp(j·0.05π)=0.95·(cos(0.05π)+j·sin(0.05π))≈0.95·(0.9877+j·0.1564)≈0.9383+j·0.1486. ​​Update filter weights w(2): Use an update rule that includes both amplitude and phase errors (element-wise multiplication is ignored for simplification): w(2)=w(1)+μ(2)·[e amp (1)+λ·e phase (1)]·r(1)·P(1)*(In actual algorithms, the conjugate of the reference signal and the phase factor P(1)* is usually used); Assuming that the role of P(1)* in weight update is mainly reflected in its phase correction, here is to demonstrate the numerical calculation process and simplify the physical meaning of this step, only the calculation is performed: weight increment Δw=0.098·[0.552+0.2·0.05π]·0.72≈0.098·[0.552+0.0314]·0.72≈0.098·0.5834·0.72≈0.0411; w(2)=w(1)+Δw=0.9+0.0411=0.9411. It can be seen that in a single iteration of n=1, the filter weight is updated from 0.9 to 0.9411. This new weight is closer to 1.0, meaning that in the next iteration, the filter's estimate of the noise, y(2), will be closer to the true noise value n. c (1) = 0.7. This process will continue until the weights converge, minimizing the energy of the filtering error e(n) (i.e., the cleaned signal) and minimizing the phase difference between its phase and the original signal.

[0047] like Figure 4As shown, according to one aspect of this application, the vibration signal of the working rib and the reference signal of the dormant rib are acquired by applying an adaptive frequency sweep excitation signal. The generation process of the adaptive frequency sweep excitation signal specifically includes:

[0048] Before applying the adaptive frequency sweep excitation signal, the background noise signal of the construction site is collected in advance;

[0049] Spectral analysis is performed on the background noise signal to generate a noise power spectrum, which is used to indicate the peak frequency band of the environmental noise.

[0050] An adaptive frequency sweep excitation signal is generated based on the noise power spectrum. The instantaneous frequency of the adaptive frequency sweep excitation signal will change the rate at which it passes through the noise peak frequency band during the frequency sweep process.

[0051] In this embodiment, the adaptive frequency sweep excitation signal is a non-uniform frequency sweep signal, and its frequency sweep speed dynamically changes according to the spectral distribution of environmental noise. Traditional fixed-frequency or linear frequency sweep signals may unfortunately fall into the strong noise frequency band generated by the operation of large equipment on site, resulting in severe signal contamination. This embodiment actively avoids this by strategically channeling signal energy into relatively clean frequency bands. Specifically, a short period before the tensioning operation begins (e.g., 30 seconds) is used to collect a segment of the background noise signal n(t) from the site using the accelerometer to be used for measurement. To obtain a stable noise spectrum estimate, the collected signal can be segmented, for example, with 2048 sampling points per segment, a 50% overlap between segments, and each segment is windowed (e.g., a Hanning window) and subjected to FFT calculation to obtain the power spectral density (PSD) of each segment. i (f). By averaging the power spectral density of all segments, a smooth and representative noise power spectral density N(f) is obtained: N(f) = (1 / M)∑ i=1 M |FFT(n) i (t))∣ 2 / N; where n i (t) represents the background noise after weighting by the window function for the i-th segment; N is the normalization factor, and M is the total number of segments. The Hannin window function is preferred because it achieves a good balance between good frequency resolution and low sidelobe leakage. A 50% overlap rate is chosen to ensure the statistical stability of the spectrum estimation while reducing the loss of data edge information due to windowing. The noise power spectral density N(f) obtained in the previous step is read and analyzed to identify noise peaks. An exemplary method is to calculate the noise mean μ across the entire spectrum. N and standard deviation σ N and all that satisfy the condition N(f)>μ N +2σ NThe frequency point f is identified as the noise peak frequency f peak These frequencies constitute the range that needs to be avoided.

[0052] According to one aspect of this application, the process of generating an adaptive frequency sweep excitation signal based on the noise power spectrum specifically includes:

[0053] Based on the noise power spectrum, a frequency avoidance weight vector is constructed, where the weight value corresponding to the noise peak frequency band is lower.

[0054] The time-varying instantaneous sweep rate is obtained by multiplying the reference sweep rate by the frequency avoidance weight vector.

[0055] Integrate the instantaneous sweep rate to generate a sweep function; and synthesize an adaptive sweep excitation signal based on the sweep function.

[0056] Specifically, a frequency avoidance weighting function W(f) is constructed based on the noise power spectral density N(f). The goal of this function is to assign low weights to frequencies with high noise and high weights to frequencies with low noise. An exemplary function form is an sigmoid function, such as W(f) = 1 / [1 + exp((N(f) - N...]. threshold ) / k′)], where N threshold Here, k is a preset noise threshold, and k' is a coefficient that controls the steepness of the transition band. Thus, when N(f) is much larger than the threshold, W(f) approaches 0; when N(f) is much smaller than the threshold, W(f) approaches 1. Combining the values ​​of this function at discrete frequency points forms the frequency avoidance weight vector W. A baseline constant sweep rate k0 is set. During the sweep process, the sweep rate is dynamically adjusted by querying the weight value W(f(t)) corresponding to the current instantaneous frequency f(t). The time-varying instantaneous sweep rate can be calculated as k(t) = k0 × W(f(t)), where k0 is the baseline sweep rate, a constant. This means that when the instantaneous frequency of the signal is about to enter the noise peak region, the corresponding W(f(t)) decreases, resulting in a decrease in the instantaneous rate k(t). This should be understood as a larger frequency change per unit time, i.e., a faster sweep across the noise region. When in a clean frequency band with less noise, W(f(t)) is larger, k(t) is also larger, and the signal stays in this band for a relatively longer time, injecting more energy. The instantaneous frequency f(t) is obtained by integrating the instantaneous sweep rate k(t): f(t) = f0 + ∫0 t The function f(t) is called the sweep frequency function. Substituting this sweep frequency function into the sine function, we can synthesize the adaptive sweep frequency excitation signal s(t) = A·sin(2π∫0^k f(t)dτ), where f0 is the starting frequency of the sweep, τ is the time variable, and d is the differential element of the integral operator. t f(τ)dτ), where A is the amplitude of the excitation signal. This signal has the ability to intelligently avoid noise.

[0057] This embodiment achieves a paradigm shift from passive noise reduction to active interference immunity by pre-sensing the background noise spectrum of the construction site before applying excitation and generating an adaptive frequency sweep excitation signal that actively avoids noise peak frequency bands, thus improving the source quality of signal acquisition. It innovatively applies the spectrum sensing concept from the field of cognitive radio to structural excitation. Specifically, it obtains a site noise map through FFT analysis, dynamically adjusts the frequency sweep rate, quickly sweeps through noisy frequency bands to reduce energy injection, and appropriately lingers in clean frequency bands with low noise to increase energy injection. In the complex electromagnetic and mechanical vibration environment of prestressed construction, this ensures that the excitation energy is always effectively coupled to the frequency band with the highest signal-to-noise ratio, improving the signal-to-noise ratio of the initial acquired signal. This not only directly improves the stability and accuracy of all subsequent processing steps but also enhances the overall system's anti-interference robustness.

[0058] According to one aspect of this application, the process of analyzing a purified pressure wave signal to extract its propagation delay includes:

[0059] Multi-scale decomposition of the purification pressure wave signal yields a narrowband component signal set;

[0060] Analyze the coherence between signals in the narrowband component signal set;

[0061] Based on coherence, signal components are classified into direct wave paths and reflected wave paths; propagation delay is extracted from the direct wave paths.

[0062] In this embodiment, multi-scale decomposition is a signal processing technique that decomposes a broadband signal into a series of narrowband components with different center frequencies and bandwidths. For example, wavelet packet decomposition can be used. The purified pressure wave signal y... i (t) Performing, for example, 5-level wavelet packet decomposition, yields signal components across multiple frequency bands. Based on the energy or signal-to-noise ratio of each band, several (e.g., 3) frequency bands with the highest energy concentration can be automatically selected and reconstructed into a narrowband component signal set {y}. i (b) (t)}, b=1,2,3. The direct wave path refers to the path with the least energy loss and shortest propagation time from the excitation point to the measurement point. The reflected wave path refers to the path of the signal reaching the measurement point after being reflected once or multiple times by structural boundaries (such as the interface between concrete and steel reinforcement, beam edges, etc.). The propagation delay of the direct wave most directly reflects the stress state between two points. However, in the received signal, the direct wave is often superimposed with various reflected waves, making direct identification difficult. By decomposing the signal into different frequency bands and utilizing the differences in frequency domain coherence of signals from different paths, it becomes possible to distinguish them.

[0063] Further analysis of relevance and classification paths includes:

[0064] Calculate the cross-power spectrum and self-power spectrum between different signals in a narrowband component signal set;

[0065] Based on the cross-power spectrum and the self-power spectrum, the coherence function used to characterize coherence is calculated.

[0066] Clustering algorithms are applied to the values ​​of the coherence function, and signal components are automatically classified into direct wave paths and reflected wave paths based on a preset coherence threshold.

[0067] Specifically, the cross-power spectrum S between different narrowband component signals from different measurement points (e.g., measurement point i and measurement point j) is calculated. ij (f)=E[X i *(f)X j (f)] and their respective power spectra S ii (f), S jj (f); where X i *(f) represents the conjugate frequency domain expression of the narrowband signal at frequency f at the i-th measurement point, X j (f) represents the frequency domain expression of the narrowband signal at frequency f at the j-th measurement point. To improve the estimation accuracy, the Welch method can be used, which calculates the signal by averaging piecewise periodograms. Based on this, the coherence function is calculated: γ 2 (f)=∣S ij (f)∣ 2 / [S ii (f)S jj (f)]. The range of the coherence function is between [0, 1]. The closer the value is to 1, the stronger the linear correlation between the two signals at that frequency. Direct wave signals, due to their single propagation path, typically maintain high coherence between their different frequency components. However, reflected waves, having undergone complex paths, experience chaotic phase relationships between their different frequency components, leading to decreased coherence. This characteristic can be used to perform cluster analysis on the calculated coherence function values. For example, the K-means clustering algorithm can be used to automatically group the coherence function values. Different clusters can be defined by preset coherence thresholds. An exemplary classification rule is: Direct wave cluster: Average coherence γ 2 Signal components >0.8; First-order reflection clusters: Average coherence in 0.4 < γ 2 Signal components with values ​​between 0.8 and 0.8; multiple reflections / noise clusters: average coherence γ 2Signal components with a value <0.4. In this way, each narrowband signal component is tagged with a path classification label indicating whether it is a direct wave or a reflected wave. It should be noted that before calculating the cross-power spectrum and self-power spectrum, accurate analytical signal construction of each narrowband component signal is required. This process should strictly follow methods based on Bedrosian's theorem verification and frequency domain processing to ensure the accuracy of subsequent coherence analysis.

[0068] Furthermore, the process of extracting the propagation delay specifically includes:

[0069] The initial path delay is extracted for the direct wave path and the reflected wave path, respectively.

[0070] Based on the path classification results, a path confidence weight is assigned to each path delay, with the weight of the path belonging to the direct wave path being higher than that of the path belonging to the reflected wave path.

[0071] For the path delay of all paths, a weighted fusion calculation is performed based on the path credibility weight corresponding to the path delay to obtain the final propagation delay.

[0072] In this embodiment, for each signal component classified in the previous step, the initial path delay τ is calculated using the phase gradient method or the cross-correlation method. i Based on its path classification label, assign a path credibility weight w to it. path，i For example, the weight of the direct wave path can be set to 1.0, the weight of the single-reflection wave path to 0.5, and the weight of the multiple-reflection wave path to 0.1. To obtain an accurate and robust final propagation delay, the delay information of all paths needs to be fused. A preferred fusion method is maximum likelihood estimation or weighted least squares. The final propagation delay τ final τ can be calculated using the following formula: final =∑(w i 2 ·τ i ) / ∑w i 2 , where w i Let w be the credibility weight of the i-th path. Further, the weight w... i It can be a comprehensive weight that considers not only path type but also the determinism of the measurement. For example, w i =w path，i ·exp(-σ τ，i / σ0), where σ τ，iσi represents the uncertainty of the i-th time delay measurement (e.g., the variance in its calculation process), and σ0 is a normalization constant. This method fully utilizes all acquired signal information, but highlights the most reliable direct wave information through weighting, effectively suppressing the interference of reflected waves on the time delay measurement, obtaining a high-precision final propagation time delay, and laying a solid foundation for subsequent stress inversion.

[0073] In some embodiments, the allocation of path credibility weights specifically involves assigning a fixed, empirical path credibility weight w to different categories of paths after path classification is completed. path，i A feasible allocation scheme is as follows: for paths classified as direct-access wave clusters, their weight w is... path，i The weight is set to 1.0; for paths classified as single-reflection clusters, the weight is set to 0.5; for paths classified as multiple-reflection clusters, the weight is set to 0.1. This reflects the decreasing reliability trend of different path signals in representing the actual propagation delay.

[0074] This embodiment solves the interference problem of multipath propagation in complex structures by introducing a multipath separation and fusion method based on coherence analysis. Specifically, the broadband received signal is decomposed into multiple narrowband components through multi-scale decomposition. Utilizing the difference in the stability of the phase relationship between the frequency components of the signals propagating along different paths, a coherence function is calculated to identify signals with highly stable phase relationships (high coherence) as direct waves, while those with chaotic relationships (low coherence) are identified as reflected waves. When fusing time delay information, different confidence weights are assigned to different paths, highlighting the contribution of the direct wave. This embodiment enables the effective differentiation of the first wave signal, which is usually weak in energy but crucial, from those masked by strong reflected waves in complex acoustic waveguides composed of prestressed tendons, grouting, and corrugated pipes. The extracted propagation time delay accurately corresponds to the shortest physical path between two points, providing a key guarantee for the accuracy of the final stress inversion.

[0075] According to one aspect of this application, the real-time stress distribution field is calculated by inversion, including:

[0076] Additional temperature data distributed along the path of the prestressing tendons were collected.

[0077] A wave velocity-stress-temperature model is established, in which wave velocity is expressed as a nonlinear function of stress and a function of temperature.

[0078] Based on the model, an inversion equation set with stress as the unknown is established, where one side of the equation is the propagation delay from the phase-delay characteristic matrix, and the other side is the path integral of the model using geometric parameters and temperature data.

[0079] Solve the inversion equations to obtain the real-time stress distribution field.

[0080] In this embodiment, the wave velocity-stress-temperature model refers to a mathematical expression that describes the quantitative relationship between the propagation velocity v of the elastic wave in the prestressing tendon, the stress σ inside the prestressing tendon, and the temperature T of its surrounding environment. The inversion equation set refers to a set of equations with the stress values ​​of discrete nodes on the prestressing tendon as unknowns and the measured propagation time delay between each node as knowns. Specifically, while performing time delay measurements, real-time temperature data T(s) at multiple measuring points along the tendon path are simultaneously collected using temperature sensors (such as fiber optic grating sensors or thermocouples) embedded in the corrugated pipe or attached to the surface of the prestressing tendon. Based on acoustoelastic theory and experimental data, a nonlinear wave velocity-stress-temperature coupling model of the following form can be established: v(σ, T) = v0(T)[1 + α(T)σ + β(T)σ] 2 [; where v0(T), α(T), β(T) are temperature-related parameters in the model, determined through real-time calibration; v(σ, T) is a nonlinear function of stress σ and temperature T. This model shows that wave velocity not only has a nonlinear (quadratic) relationship with stress σ, but its coefficients v0(T), α(T), and β(T) are also functions of temperature T. Physically, the propagation delay between two measuring points is equal to the integral of the path length between the two points with respect to the wave velocity. Based on this, an equation can be established for any pair of measuring points i and j: τ ij =∫ si sj ds / v(σ(s),T(s)); where, τ ij The measured propagation delay is a known quantity. The geometric parameters of the integration path (such as curve length s) can be accurately extracted from the BIM model, and the temperature T(s) is a real-time measured value. The unknown quantity is the stress distribution function σ(s). This is achieved by discretizing the stiffener path into several nodes and then using the stress values ​​σ at these nodes to represent the stress function σ(s). k Expressed using interpolation functions (such as linear or spline interpolation), the above integral equation is transformed into a system of nonlinear algebraic equations of the form F(σ) = 0. Since this system of equations is often ill-posed, direct solution can lead to results that are extremely sensitive to measurement errors. Therefore, a regularization method is preferred for solving it. For example, the Levenberg-Marquardt algorithm with Tikhonov regularization can be used. Its goal is to solve the minimization problem: min |Aσ - τ measured || 2 +λ∣∣Dσ∣∣ 2 The first term is the data fitting term, representing the difference between the computation delay and the measurement delay; the second term is the regularization term (or smoothing constraint term), used to ensure the stress distribution obtained from the solution; σ has physical smoothness, D is the difference operator, A is the propagation response matrix, and τmeasured The propagation delay is obtained through actual measurement via signal processing. The value of the regularization parameter λ can be adaptively determined using methods such as the L-curve. After several iterations (e.g., 20 iterations), a stable nodal stress vector σ can be obtained. nodes .

[0081] In some embodiments, the specific implementation steps of the L-curve method are as follows: traverse a series of discrete regularization parameter λ values, solve the regularized least squares problem for each λ value, and calculate the norm (or smoothness norm) of the corresponding solution ||Dσ|| 2 The norm of the residuals ||Aσ-τ measured || 2 Plotting these two sets of values ​​on a logarithmic graph as the y-axis and x-axis, respectively, will form an L-shaped curve. The λ value corresponding to the corner of this curve is the optimal regularization parameter, as it achieves the best balance between solution smoothness and data fit. The initial damping factor in the Levenberg-Marquardt algorithm is typically set to a small positive number, such as 10. -3 This factor is adaptively adjusted during the iteration process based on the amount of decrease calculated at each step to ensure stable convergence of the algorithm.

[0082] Furthermore, the process of establishing the wave velocity-stress-temperature model further includes real-time calibration of the temperature-related parameters in the model, specifically:

[0083] Based on the currently collected temperature data, and by fitting historical measurement data, the current values ​​of temperature-related parameters in the model are determined, and a real-time calibration parameter set is generated.

[0084] A wave velocity-stress-temperature model was constructed using a real-time calibration parameter set.

[0085] In this embodiment, real-time calibration refers to updating and correcting the parameters in the wave velocity model based on the latest field environmental data (mainly temperature) before each stress inversion calculation, rather than using fixed parameters measured under laboratory conditions. The acoustic properties of materials (such as sound velocity and acoustoelastic coefficient) are highly sensitive to temperature. During a long construction period, the ambient temperature may change by several or even tens of degrees, causing the model parameters to drift. Real-time calibration is precisely to compensate for this error introduced by temperature changes. Specifically, based on historical measurement data (e.g., multiple sets of time delay and stress data measured at different construction stages and temperatures), a specific functional relationship between the model parameters v0(T), α(T), and β(T) and temperature T can be fitted. An exemplary relationship is as follows: Specifically: v0(T) = v 00 [1-αT [(T-T0)],α(T)=α0exp(-β) T |T-T0|), β(T)=β0[1+γ T (T-T0) 2 ], where v 00 α T T0, α0, β T ,β0,γ T It is a material constant obtained by least-squares fitting of historical data; specifically, v 00 For the wave velocity at reference temperature T0, α T α0 represents the linear decrease rate of wave velocity with temperature; β represents the stress sensitivity at the reference temperature. T γ is the temperature suppression factor; β0 is the reference second-order stress sensitivity at the reference temperature, and γ T This represents the temperature enhancement factor. During a new stress inversion, the current temperature distribution T(s) is measured. Substituting this temperature value into the above formula allows the calculation of the parameter set {v0(T), α(T), β(T)} specific to the current temperature environment. Using this parameter set specific to the current temperature environment to construct the wave velocity-stress-temperature model for the current calculation improves the accuracy of the inversion.

[0086] In some embodiments, the engineering feasibility details for the real-time calibration section are further explained as follows: To ensure the reliability of the fitting results, the historical measurement data used for parameter fitting should meet a certain minimum sample size requirement, for example, containing at least 30 sets of valid data points collected under different operating conditions (covering at least a temperature variation range of 10°C and a maximum stress variation range of 30%). When fitting historical data to determine temperature-related parameters, a clear goodness-of-fit criterion is required. A feasible criterion is that the coefficient of determination (R-squared) of the fitted model should be greater than 0.95, so that the established parameter-temperature relationship model can highly explain the variability of the data.

[0087] This embodiment solves the problem of inaccurate stress inversion caused by traditional methods that rely on simplified linear models and static parameters by constructing a coupled model that simultaneously considers the nonlinear effects of stress and the influence of temperature, and by calibrating the temperature-related parameters in the model in real time. Specifically, this nonlinear model (e.g., containing a quadratic stress term) is closer to the physical nature of sound wave propagation in stressed metal than the traditional linear model. More importantly, through real-time calibration, the system can dynamically update key parameters in the model, such as zero-point wave velocity and acoustoelastic coefficient, using field-measured temperature data, and compensate in real time for disturbances to the material's physical properties caused by changes in the construction environment temperature (such as diurnal temperature variation). This constitutes a key closed loop from accurate time delay measurement to accurate stress inversion, enabling the entire technology chain to ultimately output stress distribution results that are highly consistent with the actual physical state, thus improving accuracy.

[0088] Furthermore, when the nonlinear inversion process calculates the stress vector σ of discrete nodes along the prestressed tendon path... nodes Then, the following steps will be performed:

[0089] Step 1: Stress Field Continuity and Verification. To facilitate visualization in the BIM model and subsequent continuous field analysis, it is first necessary to convert the discrete nodal stress vector σ... nodes This is transformed into a continuous stress distribution function σ(s,t) along path s. Specifically, cubic spline interpolation or a similar smoothing-preserving interpolation algorithm can be used to transform σ. nodes Interpolation is performed to obtain the stress value at any point along the entire prestressing tendon path. After obtaining the continuous stress field σ(s,t), a verification step can be performed to ensure the physical rationality and accuracy of the inversion results. One optional verification method is to check the smoothness of the stress gradient, i.e., calculate ▽σ to ensure its value is within a reasonable range, for example, less than the ratio of the maximum allowable stress to the beam span, to avoid physically meaningless stress abrupt changes. Another verification method is to perform virtual time delay inversion, i.e., using the obtained stress field σ(s,t) and wave velocity model, to inversely calculate the theoretical propagation time delay τ between each measuring point. calc and compare it with the measured time delay τ measured Compare them. If the relative error, for example |τ calc -τ measured | / τ measuredIf the result is less than a preset threshold (e.g., 5%), the inversion result is considered reliable. Cubic spline interpolation, specifically, preferably uses non-knot boundary conditions. This condition provides two additional constraints by forcing the third derivatives of the first and second polynomial segments near the endpoints to be equal, thus eliminating the need to manually specify the derivative values ​​at the endpoints. This typically produces a visually and mathematically smoother and more natural overall interpolation curve.

[0090] Step Two: BIM Attribute Update and 3D Visualization. The validated real-time stress distribution field σ(s,t) will be synchronously updated to the BIM model, achieving real-time refresh of the digital twin model. Specifically, this process is implemented through a programmable interface (API). The stress field data σ(s,t) is mapped to the corresponding prestressed tendon element in the BIM model and updated as a new physical attribute in the element's IFC (Industry Foundation Classes) attribute set. For example, a custom attribute set named Pset_TendonStress can be created, and the stress distribution data can be written into it. To provide intuitive feedback to on-site engineers and managers, the system can perform 3D visualization rendering based on the updated BIM model. Through algorithms such as trilinear interpolation, continuous stress cloud maps can be generated on the BIM prestressed tendon 3D model, where different colors represent different stress levels (e.g., red indicates high stress areas, and blue indicates low stress areas). This makes the distribution of prestress within the structure immediately clear.

[0091] Step 3: Model Prediction and Construction Control Parameter Optimization. The ultimate goal of this embodiment is not only monitoring, but also to guide and optimize the construction process through accurate monitoring data, forming a closed-loop feedback. Specifically, the system compares the measured real-time stress distribution field σ(s,t) with the design stress distribution field σ stored in the BIM model. d By comparing the two, the deviation field Δσ(s) = σ(s,t) - σ(s) is calculated. d (s). This deviation field intuitively reflects the gap between the current construction status and the design requirements. Based on this deviation field Δσ(s), the system uses either Model Predictive Control (MPC) or the classic PID (Proportional-Integral-Derivative) control algorithm to calculate the adjustments required for the next tensioning operation. For example, when using the PID algorithm, the correction amount ΔF of the tension force can be calculated using the following formula: ΔF=K p •Δσ+K i •∫Δσdt+K d •dΔσ / dt; where K p K i K dThese are the proportional, integral, and differential coefficients, respectively. The purpose of this calculation is to minimize the current stress deviation after the next tensioning or holding operation. The result is a series of optimized construction control parameters, such as the recommended jack tension F for the next stage. i Tensioning speed v i Duration t i Etc., forming an optimized control parameter set {F} i v i , t i Similar to stress data, optimized control parameter sets are also written back into the BIM model, for example, stored in an attribute set named Pset_ConstructionControl. The intelligent tensioning equipment on site can directly read these optimized instructions from the BIM model for operation, or site engineers can manually adjust them based on these suggestions. This achieves a complete closed loop from real-time monitoring to data analysis, intelligent decision-making, and precise control, improving the quality, safety, and automation level of prestressed construction.

[0092] In some embodiments, the control parameter optimization calculation specifically involves: if a PID control algorithm is used, its parameter K... p K i K d Tuning is crucial. A commonly used engineering method is the Ziegler-Nichols tuning method based on the system's step response. This involves applying a step tension, measuring the system's response curve, calculating an initial set of PID parameters based on the curve's characteristics (such as delay time and rise time), and then fine-tuning these parameters. If a more advanced model predictive control (MPC) algorithm is used, its prediction and control time domains need to be defined. In the application scenario of this invention, the prediction time domain can be set to 10 to 20 calculation steps in the future to predict the short-term evolution trend of stress; the control time domain can be set to 2 to 5 calculation steps to optimize the control input for a few key future steps. This setting ensures both prediction accuracy and computational efficiency, meeting the requirements of real-time control.

[0093] According to one aspect of this application, a system for implementing a BIM-based construction method for large-span post-tensioned bonded prestressed beams may include the following hardware configuration:

[0094] BIM Data Server: Stores BIM models (.ifc format) of large-span prestressed beams and provides a data interface for processing modules to read geometric and attribute information.

[0095] Central processing unit: This can be a high-performance industrial computer or server, used to run all the data processing and inversion algorithms described in this invention.

[0096] Excitation signal generation and application module: includes an arbitrary waveform generator for generating adaptive frequency sweep signals; and a piezoelectric exciter mechanically coupled to the tension jack for converting electrical signals into small-amplitude mechanical vibrations.

[0097] Sensors and Data Acquisition Module: Accelerometer: Preferably a piezoelectric IEPE accelerometer, such as the PCBPiezotronics 352C33 model, which features high sensitivity, wide frequency response, and good signal-to-noise ratio. The sensor is fixed to the tensioning end of the prestressed tendon or a specially opened observation hole using epoxy resin adhesive or a magnetic base. Temperature Sensor: Preferably a fiber Bragg grating (FBG) sensor, which can be connected in series on a single optical fiber to achieve distributed temperature measurement along the tendon path. Multi-channel synchronous data acquisition card: For example, the NI (National Instruments) PXIe-4499, which can support simultaneous sampling of up to 24 channels with a maximum sampling rate of 204.8 kS / s.

[0098] In this embodiment, the recommended ranges for the system's key operating parameters are as follows: The sampling rate is preferably 10kHz to 20kHz; a lower sampling rate may fail to capture high-frequency details of the signal, while an excessively high sampling rate will generate unnecessary data redundancy. The sweep frequency range is preferably 0.5kHz to 5kHz, which covers the commonly used frequency bands of excitation signals in most civil structure inspections. For arrangement along the beam length, the sensor spacing is preferably 1 meter to 5 meters; too small a spacing is economical, while too large a spacing may miss local stress anomalies. The temperature measurement range is -10℃ to 60℃, covering most temperatures encountered in construction environments. The stress measurement range is 0 to 1500MPa, covering the entire stress range of prestressed high-strength steel bars from initial tension to final locking.

[0099] The system's workflow and data interface are defined as follows: The central processing unit (CPU) reads model data from the BIM server and generates a sensor layout scheme and initial excitation parameters. Instructions are sent to the excitation signal generator and data acquisition card via Ethernet interface. The acquisition card transmits the acquired acceleration and temperature data (e.g., in TDMS format) back to the CPU. Different algorithm modules within the CPU (such as noise cancellation, phase analysis, and stress inversion) sequentially process the data and write the final stress field and optimized control parameters back to the BIM data server via API interface to update model attributes. The entire process is scheduled uniformly by the CPU, enabling millisecond-level real-time response.

[0100] According to one aspect of this application, before extracting the propagation delay, an improved path integral method is used for phase expansion, and the expansion integer k(τ) is determined by minimizing the phase derivative jump.

[0101] In this embodiment, when the wrapped phase Φ of the narrowband component signal is obtained through Hilbert transform... w After Φ(t), since its range is restricted to the interval [-π, π], a phase jump phenomenon will occur. It is necessary to recover its true, continuous phase Φ(t) through a phase unwrapping algorithm, which is a prerequisite for accurate group delay calculation. Traditional phase unwrapping algorithms (e.g., simply detecting whether the phase difference between adjacent points exceeds π and compensating accordingly) are very sensitive to low signal-to-noise ratios and prone to errors, leading to incorrect slopes in the unwrapped phase and thus causing significant deviations in the group delay calculation. This embodiment uses an improved path integral method to enhance the robustness of the algorithm. Specifically, before performing phase extraction, the narrowband component signal y... i (b) (t) The condition of Bedrosian's Theorem is applied to verify that its spectrum does not overlap with the spectrum of the Hilbert transform kernel, which is a prerequisite for constructing an effective analytic signal. After passing the verification, the analytic signal z is accurately constructed by zero-padding the signal spectrum in the frequency domain and applying Hermitian symmetry. i (b) (t)=y i (b) (t)+j·H[y i (b) [(t)], where H[·] represents the Hilbert transform. Continuous phase Φ(t) and wrapper phase Φ w The relationship between Φ(t) and Φ(t) is given by Φ(t) = Φ(t) w The expression is given by: dΦ / dt = dΦ(t) + 2πk(t), where k(t) is an integer that varies with time. Differentiating this expression, we get dΦ / dt = dΦ. w / dt +2π(dk / dt). Since the actual instantaneous frequency (i.e., the phase derivative) should be gradually changing, the jumps in the phase derivative mainly come from dΦ. w / dt is the pulse at the wrapping point. The improved path integral method has the following expansion formula: Φ(t)=Φ(0)+∫0 t [dΦ w / dτ+2πk′(τ)]dτ Here, k′(τ) is not a direct integer, but represents an integer multiple of 2π that needs to be compensated at time τ. The principle for determining k′(τ) is to minimize the derivative jump of the final expanded phase Φ(t). An algorithm is used to find a set of integer k... i (corresponding to discrete time point i), such that the objective function J = ∑ i ∣(Φ i+1 -Φ i )-(Φ i -Φi-1 )∣ 2 Minimization. Essentially, this is an optimization problem of minimizing the second-order difference (i.e., the approximation of the derivative). This problem can be solved efficiently using dynamic programming or least squares algorithms, thereby determining an integer compensation value k that best preserves the continuity of the phase slope for each point in time where phase wrapping occurs. i Even when local noise in the signal causes phase jitter, this algorithm can find a globally optimal expansion path, avoiding the error accumulation of traditional algorithms, thus obtaining an accurate and smooth continuous phase sequence Φ. i (b) (t) provides a high-quality input for subsequent calculation of group delay by differentiation.

[0102] In one specific embodiment, the experiment was conducted at the construction site of an actual long-span continuous beam bridge, with a 50-meter-long prestressed tendon as the test object. The BIM-based construction method for long-span post-tensioned bonded prestressed beams of this application (denoted as the "method") was compared with a traditional monitoring method based on fixed-frequency excitation and strain gauge measurement (denoted as the "conventional method"). After tensioning, the actual stress value of the prestressed tendon at the mid-span position was measured to be 1155 MPa using a high-precision pressure sensor. The stress value calculated by the conventional method using strain gauges bonded to the concrete surface was 1020 MPa, with a relative error of (1155-1020) / 1155≈11.7%. The stress value obtained by the method through inversion was 1148 MPa, with a relative error of (1155-1148) / 1155≈0.6%. It can be seen that the measurement accuracy of the method is much higher than that of the conventional method. A high-power welding machine was used near the tensioning platform to simulate a strong electromagnetic noise environment. The signal-to-noise ratio (SNR) of the purified / processed signal is used as the evaluation metric. Under interference-free conditions: the SNR of the signal acquired by the traditional method is approximately 25 dB, while the SNR of the signal acquired by this method (after noise cancellation) is approximately 45 dB. Under welding machine interference: the SNR of the signal acquired by the traditional method drops sharply to approximately 8 dB, with the signal waveform almost completely submerged by noise. This method, due to the use of a physical rib reference channel and phase-preserving adaptive noise cancellation, maintains a processed SNR of approximately 35 dB, ensuring the reliability of subsequent analysis. It can be seen that this method has the ability to resist common-mode noise interference. On an industrial computer with the same hardware configuration (Intel i7-10700 CPU, 32GB RAM), 10 seconds of acquired data (sampling rate 10kHz) were processed. Traditional method (mainly filtering and peak detection): total time approximately 0.8 seconds. This method (including all adaptive excitation generation, noise cancellation, phase analysis, and stress inversion steps): total time approximately 2.5 seconds. Although this method has higher computational complexity, its processing time is much shorter than the data acquisition time, which can meet the real-time requirements of construction monitoring (typically with a response time in minutes). Furthermore, considering the improved accuracy and robustness it brings, this increase in computational cost is worthwhile.

[0103] The preferred embodiments of the present invention have been described in detail above. However, the present invention is not limited to the specific details in the above embodiments. Within the scope of the technical concept of the present invention, various equivalent transformations can be made to the technical solutions of the present invention, and these equivalent transformations all fall within the protection scope of the present invention.

Claims

1. A BIM-based construction method for large-span post-tensioned bonded prestressed beams, characterized in that, include: The vibration signals of the working reinforcing bars and the reference signals of the dormant reinforcing bars are collected; where the working reinforcing bars refer to the reinforcing bars that participate in the structural stress; and the dormant reinforcing bars refer to the reinforcing bars that do not participate in the direct stress. The noise cancellation process of the working rib vibration signal is performed using the dumb rib reference signal to generate a purification pressure wave signal. Analyze the purification pressure wave signal, extract the propagation delay, and generate a phase-delay feature matrix; Based on the phase-delay feature matrix and the geometric parameters of the pre-configured BIM 3D model, the real-time stress distribution field is calculated by inversion and updated to the BIM 3D model. Among them, the acquisition of the dummy bar reference signal originates from a physical dummy bar that is the same as the working bar in terms of path and material but is not subject to tension. The physical dummy bar is specifically used as a noise reference channel to measure common mode noise. The process of generating a purification pressure wave signal includes: inputting the dumb rib reference signal into an adaptive filtering algorithm to generate a dynamic estimate of the common-mode noise, and subtracting the dynamic estimate from the working rib vibration signal to generate the purification pressure wave signal. The adaptive filtering algorithm uses an adaptive step size to update its time-varying filtering weights. The process for determining the step size is as follows: The variance of the filtering error sequence generated by the adaptive filtering algorithm is calculated in real time to obtain the error variance value; Based on the preset inverse function relationship, the value of the adaptive step size is determined by the error variance value, so that the step size is inversely proportional to the error variance value.

2. The method according to claim 1, characterized in that, The adaptive filtering algorithm controls the generation of dynamic estimates by applying phase-preserving constraints; The phase preservation constraint specifically aims to minimize the instantaneous phase difference between the working rib vibration signal and the final generated purification pressure wave signal.

3. The method according to claim 2, characterized in that, The application of phase-preserving constraints specifically includes: The instantaneous phase difference is obtained by calculating the instantaneous phase difference between the working rib vibration signal and the purification pressure wave signal using Hilbert transform. Construct a phase-preserving factor in the form of a complex exponential function of instantaneous phase difference; The phase preservation factor is introduced as a multiplication term into the update rule of the time-varying filter weights in the adaptive filtering algorithm.

4. The method according to claim 1, characterized in that, The vibration signal of the working rib and the reference signal of the dormant rib are acquired by applying an adaptive frequency sweep excitation signal. The specific generation process includes: Before applying the excitation, the background noise signal of the construction site should be collected in advance; Perform spectral analysis on the background noise signal to generate a noise power spectrum; An adaptive frequency sweep excitation signal is generated based on the noise power spectrum. Its instantaneous frequency changes the rate at which it passes through the noise peak frequency band during the frequency sweep process.

5. The method according to claim 4, characterized in that, Generate an adaptive frequency sweep excitation signal based on the noise power spectrum, including: Based on the noise power spectrum, a frequency avoidance weight vector is constructed, where the weight value corresponding to the noise peak frequency band is lower than a preset threshold. The time-varying instantaneous sweep rate is obtained by multiplying the preset baseline sweep rate with the frequency avoidance weight vector. Integrate the instantaneous sweep rate to generate the sweep function; An adaptive sweep frequency excitation signal is synthesized based on a sweep frequency function.

6. The method according to claim 1, characterized in that, Extracting propagation delay includes: Multi-scale decomposition of the purification pressure wave signal yields a narrowband component signal set; Analyze the coherence between signals in the narrowband component signal set; Based on coherence, signal components are classified into direct wave paths and reflected wave paths; Extract the propagation delay from the direct wave path.

7. The method according to claim 6, characterized in that, The analysis of coherence and the classification of signal components into direct wave paths and reflected wave paths specifically include: Calculate the cross-power spectrum and self-power spectrum between different signals in a narrowband component signal set; Based on the cross-power spectrum and the self-power spectrum, the coherence function used to characterize coherence is calculated. Clustering algorithms are applied to the values ​​of the coherence function, and signal components are automatically classified into direct wave paths and reflected wave paths based on a preset coherence threshold.

8. The method according to claim 7, characterized in that, Extracting the propagation delay from the direct wave path includes: The initial path delay is extracted for the direct wave path and the reflected wave path, respectively. Based on the path classification results, a path reliability weight is assigned to each path delay; the weight of a path belonging to a direct wave path is higher than the weight of a path belonging to a reflected wave path. For the path delay of all paths, a weighted fusion calculation is performed based on the corresponding path credibility weight to obtain the final propagation delay.

Images