A structural displacement monitoring method and system based on intelligent sensors

Abstract

The application relates to the technical field of underground structure displacement measurement, and discloses a structure displacement monitoring method and system based on an intelligent sensor. The method comprises the following steps: collecting initial inclination signals, acceleration signals and gyroscope static signals; generating a zero starting depth reference point; constructing a depth time sequence; executing real-time compensation adjustment by a fuzzy logic controller (FLC); constructing an equidistant sampling point sequence; and calculating an accumulated displacement amount based on a double-channel displacement fusion mechanism of weight curvature correction. Compared with the displacement estimation method relying on inclination sensor integration or acceleration sensor accumulation in the prior art, especially in the soft soil layer scene of multi-source signal drift and probe speed fluctuation, the technical problem that high-precision structure displacement measurement is difficult to realize is solved. Since the speed increment adjustment mechanism driven by the fuzzy logic controller is introduced, dynamic deviation correction and depth displacement decoupling modeling of multi-sensor data are realized, and the spatial analysis accuracy of the displacement monitoring result is improved.

Description

Technical Field

[0001] This invention relates to the field of underground structure displacement measurement technology, and in particular to a structural displacement monitoring method and system based on intelligent sensors. Background Technology

[0002] Currently, with the continuous development of urban underground space, projects such as subway tunnels, diaphragm walls, and foundation pit retaining structures have placed higher demands on the real-time monitoring accuracy of deep structural deformation. Displacement monitoring methods for underground structures mainly fall into three categories: The first category is displacement estimation methods based on inclinometer integration. This involves deploying inclinometer tubes and collecting inclinometer data along the depth direction, then calculating the horizontal displacement using angle integration. However, this method is highly dependent on the initial calibration accuracy of the sensor, and in soft soil layers or scenarios with poor signal stability, it is prone to cumulative integral drift errors, resulting in severe deviations in the end-point displacement. The second category is motion estimation methods based on accelerometer accumulation. This involves embedding MEMS inertial units in the inclinometer probe and calculating displacement through double integration of the acceleration signal. However, accelerometers are sensitive to weak disturbances and suffer from zero-bias drift during long-term operation. Especially in underground construction environments with significant vibration and noise interference, the stability and reliability of their displacement output cannot be guaranteed. The third category comprises extended methods that integrate multi-source information such as inertial navigation and gyroscope data. These methods attempt to enhance the robustness of displacement estimation by fusing attitude estimation and velocity. However, they often employ fixed filter parameters or linear feedback algorithms, which cannot cope with the complex coupling behavior of "velocity fluctuation—signal noise—nonlinear drift" during structural deformation. Especially in scenarios with uneven probe velocity, severe environmental magnetic interference, or deep soft soil heterogeneity, the adjustment capability and response speed of traditional algorithms are both limited. In summary, current technologies have significant shortcomings in multi-source data fusion, dynamic displacement correction, and adaptive control of slip states, making it difficult to meet the high precision, high stability, and anti-interference requirements for underground structure displacement monitoring.

[0003] Therefore, there is an urgent need for a structural displacement monitoring method based on intelligent sensors that can still achieve high-precision, stable and reliable real-time monitoring of structural displacement under conditions of multi-source noise drift, non-constant probe speed and non-linear structural deformation, so as to improve the engineering adaptability of underground structure safety assessment and risk early warning. Summary of the Invention

[0004] To address the aforementioned technical shortcomings, the purpose of this invention is to propose a structural displacement monitoring method based on intelligent sensors. This method aims to solve the technical problem that existing displacement estimation methods, which rely on tilt sensor integration or accelerometer accumulation, are particularly difficult to implement with high precision in soft soil scenarios with multi-source signal drift and probe velocity fluctuations.

[0005] To solve the above-mentioned technical problems, the present invention adopts the following technical solution: The present invention provides a structural displacement monitoring method based on intelligent sensors. The structure displacement monitoring method based on intelligent sensors includes: Step S10: Deploy inclinometer probes and intelligent sensor arrays inside the structure to be measured. Collect initial tilt angle signals, probe acceleration signals, and gyroscope static signals through the inclinometer probes and intelligent sensor arrays. Based on the initial tilt angle signals, probe acceleration signals, and gyroscope static signals, use a reprojection mechanism based on gravity vector offset verification to correct the initial depth readings and generate a zero starting depth reference point. ; Step S20: During the downward movement of the inclinometer probe, at a preset sampling period... Real-time zero-starting-depth reference point Integral remapping calibration is performed to form a depth time series; based on the depth time series data set, a two-way sliding window fitting interpolation method is used to perform a multi-scale velocity estimation task, and the velocity estimation time series V is output. Step S30: Obtain the target velocity reference value sequence Based on the velocity estimation time series V and the target velocity reference value series The speed deviation time series is calculated, and the motor control quantity is adjusted in real time by a fuzzy logic controller (FLC) driven by sliding window trend prediction based on the speed deviation time series, and the cumulative displacement increment of the probe is output. Step S40: When the cumulative displacement increment of the probe exceeds the preset displacement increment threshold, an effective spatial sampling event is immediately triggered; the effective spatial sampling event records the probe triaxial tilt angle value and probe position depth at the time of the event trigger, and writes this set of data into the equidistant sampling point sequence, and outputs the equidistant sampling point sequence; Step S50: For the equidistant sampling point sequence, a dual-channel displacement fusion mechanism based on weighted curvature correction is used to calculate the cumulative horizontal displacement and the cumulative vertical displacement. Finally, the calculation results of the horizontal displacement and the cumulative vertical displacement are uploaded to the terminal device for visualization display through a preset Bluetooth module.

[0006] Preferably, in step S10, the initial tilt angle output signal includes a horizontal initial tilt angle output signal. and vertical initial tilt angle output signal The probe acceleration signal includes the horizontal probe acceleration signal. and vertical probe acceleration signal The static signal of the gyroscope includes the static signal of the horizontal gyroscope. and vertical gyroscope static signal .

[0007] Preferably, in step S10, the initial depth reading is corrected using a reprojection mechanism based on gravity vector offset verification, based on the initial tilt angle signal, probe acceleration signal, and gyroscope static signal, to generate a zero starting depth reference point. The steps specifically include: First, an ideal gravity vector signal is constructed based on the initial tilt angle signal, probe acceleration signal, and gyroscope static signal. Then, the initial attitude offset angle is calculated using the cosine similarity inverse solution method based on the ideal gravity vector signal and the preset standard Earth gravity vector signal. Next, an initial attitude offset angle threshold is set. Based on the initial attitude offset angle and the initial attitude offset angle threshold, an initial rotation matrix is ​​constructed using the perturbation vector iteration method. The initial rotation matrix is ​​used to rotate the tilt attitude of the current inclinometer probe in the local coordinate system to the ideal vertical direction. Then, the gravity vector residual is calculated based on the initial rotation matrix and the ideal gravity vector signal. ; Repeatedly iterate to calculate the gravity vector residual until... If the residual convergence threshold is less than the preset value, a corrected rotation matrix is ​​output. Finally, the initial depth reading is reprojected onto the global coordinate system based on the corrected rotation matrix, and the zero starting depth reference point is output. .

[0008] Preferably, in step S20, during the downward movement of the inclinometer probe, a preset sampling period is used. Real-time zero-starting-depth reference point The steps include performing integral remapping calibration to form a depth time series; and using a bidirectional sliding window fitting interpolation method to perform a multi-scale velocity estimation task based on the depth time series data set, outputting a velocity estimation time series V. Specifically, these steps include: Step S201: Acquire data at a preset sampling period during the downward movement of the inclinometer probe. Depth data groups at intervals Set a sliding window with a length of 2w+1 for the center of the depth data group. Sliding window Internal data is defined as ;in, For a moment The zero starting depth reference point; For a moment The zero starting depth reference point; For a moment The zero-starting depth reference point; n is the length of the depth data group; Step S202: For sliding windows Perform bidirectional polynomial fitting to construct a local depth trend model. Based on local depth trend model The forward depth window sequence and the backward depth window sequence are extracted; Step S203: Based on the extracted forward depth window sequence and backward depth window sequence, construct the target velocity estimate at time t using the Savitzky-Golay filtering fitting method. It outputs the velocity estimation time series V.

[0009] Preferably, in step S30, a target velocity reference value sequence is obtained. Based on the velocity estimation time series V and the target velocity reference value series The steps of calculating the speed deviation time series, and adjusting the motor control quantity in real time using a fuzzy logic controller (FLC) driven by sliding window trend prediction based on the speed deviation time series, and outputting the cumulative displacement increment of the probe, specifically include: Step S301: Obtain the target velocity reference value sequence Based on the velocity estimation time series V and the target velocity reference value series Calculate the velocity deviation time series; calculate the trend acceleration series based on the velocity deviation time series using the smoothing differential method. Step S302: Using the velocity deviation time series and trend acceleration series as input variables, establish a two-dimensional fuzzy rule table for the fuzzy logic controller (FLC); the two-dimensional fuzzy rule table includes the input fuzzy set and the output control command; Step S303: Adjust the motor control quantity in real time based on the output control command and update the position of the inclinometer probe in real time, and finally output the cumulative displacement increment of the probe.

[0010] Preferably, in step S30, the input fuzzy set includes a speed deviation input set and a trend acceleration input set; the output control command includes a speed control set, which includes "strong deceleration", "medium deceleration", "hold", "medium acceleration" and "strong acceleration"; the speed control set is used to represent the adjustment direction and intensity of the motor drive.

[0011] Preferably, step S50, which involves calculating the cumulative horizontal displacement and cumulative vertical displacement using a weighted curvature correction-based dual-channel displacement fusion mechanism for the equally spaced sampling point sequence, specifically includes: For the i-th equally spaced sampling point in the equally spaced sampling point sequence First, the initial horizontal displacement increment of the first channel is calculated using the angle measurement method. and the original vertical displacement increment Then, for equally spaced sampling points... The local curvature correction term is calculated using the second-order finite difference method. Based on this local curvature correction term and the angle measurement method, the corrected horizontal displacement increment of the second channel is calculated. and corrected vertical displacement increment ; Curvature-aware fusion weights are set based on local curvature correction terms, and the proportions of the first and second channels in the final fusion result are dynamically adjusted based on the curvature-aware fusion weights to output the fusion displacement increment result. The cumulative horizontal displacement and cumulative vertical displacement of each depth point are obtained by accumulating the fusion displacement increment result segment by segment.

[0012] The present invention also provides a structural displacement monitoring system based on intelligent sensors, comprising: Step S50, which involves calculating the cumulative horizontal and vertical displacements using a weighted curvature correction-based dual-channel displacement fusion mechanism for the equally spaced sampling point sequence, specifically includes: For the i-th equally spaced sampling point in the equally spaced sampling point sequence First, the initial horizontal displacement increment of the first channel is calculated using the angle measurement method. and the original vertical displacement increment Then, for equally spaced sampling points... The local curvature correction term is calculated using the second-order finite difference method. Based on this local curvature correction term and the angle measurement method, the corrected horizontal displacement increment of the second channel is calculated. and corrected vertical displacement increment ; Curvature-aware fusion weights are set based on local curvature correction terms, and the proportions of the first and second channels in the final fusion result are dynamically adjusted based on the curvature-aware fusion weights to output the fusion displacement increment result. The cumulative horizontal displacement and cumulative vertical displacement of each depth point are obtained by accumulating the fusion displacement increment result segment by segment.

[0013] The present invention also provides a structural displacement monitoring device based on intelligent sensors, comprising: a memory, a processor, and a structural displacement monitoring program based on intelligent sensors stored in the memory and executable on the processor. When the structural displacement monitoring program based on intelligent sensors is executed by the processor, a structural displacement monitoring method based on intelligent sensors is implemented.

[0014] The present invention also provides a computer program product, including a structural displacement monitoring program based on intelligent sensors, wherein the structural displacement monitoring program based on intelligent sensors implements the structural displacement monitoring method based on intelligent sensors when executed by a processor.

[0015] The beneficial effects of this invention are as follows: By integrating a mechanical sensor array with a fuzzy logic controller (FLC) through a linkage control mechanism, this invention achieves dynamic compensation for probe speed fluctuations and signal drift, effectively improving the stability and anti-interference capability of structural displacement monitoring in nonlinear deformation environments such as soft soil layers.

[0016] This invention combines the acquired tilt angle, acceleration, and gyroscope signals to construct a zero-starting depth reference point based on gravity vector reprojection, which solves the problem of cumulative error amplification caused by initial state error in traditional displacement measurement and improves the depth analysis accuracy of structural deformation. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart illustrating the first embodiment of a structural displacement monitoring method based on intelligent sensors according to the present invention.

[0019] Figure 2 This is a schematic diagram of a device for a structural displacement monitoring method based on intelligent sensors according to the present invention. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. 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 are within the scope of protection of the present invention.

[0021] Example 1: As Figure 1 The diagram shown is a flowchart of the first embodiment of the structural displacement monitoring method based on intelligent sensors of the present invention, which presents the first embodiment of the structural displacement monitoring method based on intelligent sensors of the present invention.

[0022] In the first embodiment, the structural displacement monitoring method based on smart sensors includes: Step S10: Deploy inclinometer probes and intelligent sensor arrays inside the structure to be measured. Collect initial tilt angle signals, probe acceleration signals, and gyroscope static signals through the inclinometer probes and intelligent sensor arrays. Based on the initial tilt angle signals, probe acceleration signals, and gyroscope static signals, use a reprojection mechanism based on gravity vector offset verification to correct the initial depth readings and generate a zero starting depth reference point. ; It should be noted that the "reprojection mechanism based on gravity vector offset verification" in this step refers to the following: The acquired initial tilt angle signal, three-axis acceleration signal, and gyroscope static signal are fused together. By establishing a deviation projection relationship between the theoretical model and the measured model of the three-dimensional gravity vector, an error suppression mechanism is constructed, including attitude calculation, sensor zero drift compensation, and depth vector correction. This mechanism not only includes the extraction and calibration of low-frequency steady-state components in the sensor output but also covers compensation for instantaneous angular velocity offset caused by the initial downward disturbance of the probe, ensuring the repeatability and stability of the final corrected zero-starting depth reference point.

[0023] Understandably, the gravity vector reprojection mechanism described above can effectively compensate for sensor zero drift and initial angle offset caused by non-ideal conditions such as non-vertical structural arrangement, initial probe oscillation, and friction interference from soft soil pipe walls. This makes the depth axis sequence constructed with this correction result as a reference more accurate, providing a more reliable spatial reference point for subsequent velocity integral estimation and displacement reconstruction, and improving the consistency of depth positioning and displacement analysis accuracy throughout the entire inclinometer link.

[0024] It should be understood that traditional structural displacement measurement methods often rely on the initial depth point provided by a single tilt sensor, electronic ruler, or roller encoder. In soft soil environments or complex geological scenarios, these methods are highly susceptible to interference from factors such as probe swaying, uncertain initial slip distance, and uneven pipe wall contact surfaces, leading to depth axis drift or accumulated displacement estimation errors, especially significant errors in the 0-0.5m range near the starting point. This invention, however, integrates data from multiple mechanical sensors to construct a multi-source attitude reference surface and verifies its projection against the standard gravity direction. This achieves joint correction of multi-dimensional angular errors and gravity direction offsets, resulting in stronger physical consistency of the initial depth reference point.

[0025] For example, in a diaphragm wall displacement monitoring project, the initial offset error measured by initializing the depth axis using an inclination sensor averaged ±6.4cm. Furthermore, due to the uncontrollable probe installation posture, there was a repeatability difference of over 30% in the initial points. However, after processing using the reprojection mechanism described in this invention, the initial depth error of multiple monitoring points under three repeated deployments was controlled within ±1.2cm, with an average deviation of 0.7cm. The stability of the initial points was improved by approximately 80%, significantly increasing the overlap of the displacement curve across different measurement cycles and avoiding the problem of false displacement jumps in the initial data.

[0026] Step S20: During the downward movement of the inclinometer probe, at a preset sampling period... Real-time zero-starting-depth reference point Integral remapping calibration is performed to form a depth time series; based on the depth time series data set, a two-way sliding window fitting interpolation method is used to perform a multi-scale velocity estimation task, and the velocity estimation time series V is output. It should be noted that the "integral remapping calibration process" in this step refers to performing discrete integral iterations on the zero-starting depth reference point generated in the previous stage within the continuous time domain, and correcting the integral error by combining the real-time acceleration information of the probe, so as to achieve synchronous mapping between depth and time. The so-called "bidirectional sliding window fitting interpolation method" refers to performing local polynomial fitting on the forward and backward data windows centered at t at each time t, and then using the derivative of the fitting function to estimate the instantaneous velocity at that time, in order to reduce the deviation of unidirectional fitting at velocity abrupt change points. This method includes low-pass filtering of acceleration noise, adaptive adjustment of the time window length, and real-time updating of velocity smoothing constraints.

[0027] Understandably, through this combined integral remapping and bidirectional sliding window processing method, this step enables self-calibration of the depth sequence and continuous reconstruction of the velocity curve under non-uniform probe motion, ensuring the velocity estimation results are immune to short-term disturbances and maintain long-term trends. This mechanism can maintain the stability of the velocity estimation even when the time sampling frequency is constant but the spatial sliding distance varies, providing a high-confidence reference signal for the subsequent fuzzy logic controller input.

[0028] It should be understood that traditional methods for calculating probe velocity using unidirectional difference or fixed-step integration are prone to velocity curve jitter or delayed response under conditions of uneven probe sliding speed and large frictional variations, leading to the accumulation of displacement calculation errors. This invention achieves mutual compensation of local trends through a bidirectional sliding window fitting interpolation method, allowing forward data to reflect inertial extension and backward data to reflect damped convergence, thus mathematically forming a symmetrically weighted velocity estimation model and significantly reducing the impact of dynamic noise on the velocity calculation results.

[0029] Step S30: Obtain the target velocity reference value sequence Based on the velocity estimation time series V and the target velocity reference value series The speed deviation time series is calculated, and the motor control quantity is adjusted in real time by a fuzzy logic controller (FLC) driven by sliding window trend prediction based on the speed deviation time series, and the cumulative displacement increment of the probe is output. It should be noted that the sliding window trend prediction-driven fuzzy logic controller in this step refers to extracting the trend characteristics of the speed deviation sequence within a continuous time window, calculating the trend increment of speed change in the short term through linear regression prediction or exponential smoothing methods, and using speed deviation and trend increment as the two input variables of the fuzzy controller. The fuzzy controller has a built-in two-dimensional fuzzy rule table, with "speed deviation" and "deviation change rate" as input fuzzy sets. The membership function of the fuzzy set can adopt a triangular or trapezoidal distribution. The output fuzzy set corresponds to the adjustment level of the motor control quantity. After defuzzification, the output is fed back to the motor drive unit in real time to realize the dynamic adaptive adjustment of the probe's sliding speed.

[0030] Understandably, the fuzzy logic control mechanism based on sliding window trend prediction described above can correct the motor's drive current and control duty cycle in real time under conditions such as probe speed fluctuations, sudden changes in friction, or unstable sliding inertia, thereby maintaining the speed stability and controllability of the inclinometer probe during its descent. Compared to traditional proportional-integral (PI) or proportional-integral-derivative (PID) algorithms that rely solely on linear error feedback, the fuzzy controller in this scheme can make forward-looking adjustments based on the trend of speed deviation changes, improving the response speed and stability of speed control.

[0031] It should be understood that traditional methods using fixed motor speed or simple PI closed-loop control cannot identify in real time the differences in frictional resistance and changes in sliding inertia of the probe at different depths. This can easily lead to sudden increases or decreases in speed in soft soil layers or sections with abrupt changes in tilt angle, thus affecting the smoothness of the speed integral results and the accuracy of displacement calculation. This invention provides forward information for fuzzy logic control through a sliding window trend prediction algorithm, enabling the controller to perform pre-compensation adjustments before the speed deviation fully manifests, fundamentally reducing the disturbance of nonlinear speed fluctuations to the displacement calculation link.

[0032] Step S40: When the cumulative displacement increment of the probe exceeds the preset displacement increment threshold, an effective spatial sampling event is immediately triggered; the effective spatial sampling event records the probe triaxial tilt angle value and probe position depth at the time of the event trigger, and writes this set of data into the equidistant sampling point sequence, and outputs the equidistant sampling point sequence; It should be noted that the "effective spatial sampling event" in this step refers to the data recording action automatically triggered once the cumulative displacement increment of the probe reaches a preset sampling interval (e.g., every 5cm or every 10cm) during the probe's sliding process. The sampling content includes the triaxial tilt angle values ​​at the instant the event is triggered (usually acquired by an attitude sensor or IMU module) and the current probe depth position; the triaxial tilt angles include attitude angles around the X, Y, and Z axes, typically represented as Roll, Pitch, and Yaw values; the depth position is obtained by continuous accumulation using the aforementioned integral remapping mechanism. The results of each sampling event will be organized into structured data items and added to the sampling point sequence for subsequent displacement reconstruction and profile analysis tasks.

[0033] Understandably, using a "cumulative displacement threshold" as the trigger condition for sampling effectively avoids the non-uniform spatial distribution problem caused by traditional sampling methods based on a "fixed time period" under non-uniform speeds. Especially when the probe's sliding speed fluctuates, if sampling triggering is based solely on time rather than spatial intervals, it may lead to an alternating distribution of dense and sparse regions, thus affecting the spatial resolution and accuracy of subsequent displacement curves. Employing a displacement-based sampling control mechanism ensures that sampling is always conducted at equal intervals in space, contributing to the construction of a displacement curve sequence with good continuity and physical consistency.

[0034] It should be understood that traditional underground displacement measurement techniques are mostly based on isochronous sampling, which assumes that the probe slip velocity is constant or negligible. However, this assumption is difficult to hold under complex geological conditions, resulting in severely uneven spacing between sampling points at different depths. Consequently, the reconstructed structural deformation profile contains jumps or offset errors. The sampling mechanism proposed in this invention, based on a "cumulative displacement increment threshold triggering" approach, significantly improves the spatial representativeness of sampling points and the structural continuity of the reconstructed model from the perspective of spatial uniformity. It also enhances the sensitivity to changes in structural details, making it particularly suitable for soft soil layers, high deformation zones, or shallowly buried non-uniform geological environments.

[0035] For example, in a subway section settlement monitoring scenario, the traditional timed sampling scheme (1 second / time) resulted in a sampling interval of 15cm in the probe acceleration phase and only 3cm in the deceleration phase, leading to extremely uneven sampling density along the depth axis. However, by adopting the equal-interval triggering mechanism of this invention, the sampling point spacing is stably controlled at around 5cm, and the sampling density stability is improved to over 93%. The lateral displacement curve constructed in the end is approximately 42% lower in fitting error at the structural transition section compared to the traditional scheme, achieving high-resolution, low-noise reconstruction of the underground structure deformation profile.

[0036] Step S50: For the equidistant sampling point sequence, a dual-channel displacement fusion mechanism based on weighted curvature correction is used to calculate the cumulative horizontal displacement and the cumulative vertical displacement. Finally, the calculation results of the horizontal displacement and the cumulative vertical displacement are uploaded to the terminal device for visualization display through a preset Bluetooth module.

[0037] It should be noted that the "dual-channel displacement fusion mechanism based on weighted curvature correction" in this step refers to obtaining preliminary displacement estimates from the tilt integral channel and the acceleration integral channel respectively, and then fusion them by assigning different weights to the outputs of the two channels based on the curvature change characteristics of each point in the equally spaced sampling point sequence. Specifically, for regions with gentle curvature, the global trend estimate of the acceleration integral is more trusted; while in regions with abrupt curvature changes, the response weight to the local changes reflected by the tilt integral is increased, forming an adaptive fusion strategy that takes into account both global stability and local sensitivity. After curvature weighting correction, the calculation results of the horizontal and vertical displacements are sent to a mobile terminal device (such as a tablet or engineering phone) via a low-power Bluetooth module and displayed graphically, realizing the visual monitoring of the structural deformation trend.

[0038] Understandably, the dual-channel fusion mechanism overcomes the accuracy bottleneck faced by traditional displacement estimation schemes that rely on a single sensor path. The tilt integration path is prone to error accumulation but sensitive to local abrupt changes; the acceleration integration path suffers from noise amplification but possesses long-term smoothness. Combining the two and adaptively weighting them based on curvature changes significantly reduces distortion caused by sensor drift or data noise, improving the robustness and spatial resolution of displacement estimation. In particular, real-time data transmission via Bluetooth significantly enhances the convenience and engineering applicability of measurement results, reducing delays and errors during manual reading.

[0039] Example 2: Furthermore, the present invention provides a structural displacement monitoring system based on intelligent sensors, employing a structural displacement monitoring method based on intelligent sensors as described in the above embodiments, which can solve the technical problem of structural displacement monitoring based on intelligent sensors. Compared with the prior art, the beneficial effects of the structural displacement monitoring system based on intelligent sensors provided by the present invention are the same as those of the structural displacement monitoring method based on intelligent sensors provided in the above embodiments, and other technical features of the structural displacement monitoring system based on intelligent sensors are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.

[0040] Example 3: This invention provides a structural displacement monitoring device based on intelligent sensors. Please refer to... Figure 2A structural displacement monitoring device based on intelligent sensors includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, which are executed by the at least one processor to enable the at least one processor to perform a structural displacement monitoring method based on intelligent sensors as described in Embodiment 1 above. The structural displacement monitoring device based on intelligent sensors in this embodiment of the invention may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital radio receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Description), PMPs (Portable Media Players), vehicle terminals (e.g., vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. A structural displacement monitoring device based on intelligent sensors is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of the invention. A structural displacement monitoring device based on intelligent sensors may include a processing device 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory 1002 or a program loaded from a storage device 1003 into a random access memory 1004. Random access memory 1004 also stores various programs and data required for the operation of a smart sensor-based structural displacement monitoring device. Processing device 1001, read-only memory 1002, and random access memory 1004 are interconnected via bus 1005. I / O interface 1006 is also connected to the bus. Typically, the following systems can be connected to I / O interface 1006: input devices 1007 including, for example, touchscreens, touchpads, keyboards, mice, image sensors, microphones, accelerometers, gyroscopes, etc.; output devices 1008 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 1003 including, for example, magnetic tapes, hard disks, etc.; and communication devices 1009. Communication device 1009 allows a smart sensor-based structural displacement monitoring device to communicate wirelessly or wiredly with other devices to exchange data. Although a smart sensor-based structural displacement monitoring device with various systems is shown in the figure, it should be understood that it is not required to implement or possess all the systems shown. More or fewer systems can be implemented alternatively.

[0041] Example 4: This invention also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the above-described method for monitoring structural displacement based on an intelligent sensor. The computer program product provided by this invention can solve the technical problem of monitoring structural displacement based on an intelligent sensor. Compared with the prior art, the beneficial effects of the computer program product provided by this invention are the same as those of the above-described method for monitoring structural displacement based on an intelligent sensor, and will not be repeated here.

[0042] In particular, according to the embodiments disclosed in this invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from read-only memory 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this invention.

[0043] It should be understood that the various parts disclosed in this invention can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics may be combined in any suitable manner in one or more embodiments or examples.

[0044] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A structural displacement monitoring method based on intelligent sensors, characterized in that, The methods include: Step S10: Deploy inclinometer probes and intelligent sensor arrays inside the structure to be measured. Collect initial tilt angle signals, probe acceleration signals, and gyroscope static signals through the inclinometer probes and intelligent sensor arrays. Based on the initial tilt angle signals, probe acceleration signals, and gyroscope static signals, use a reprojection mechanism based on gravity vector offset verification to correct the initial depth readings and generate a zero starting depth reference point. ; Step S20: During the downward movement of the inclinometer probe, at a preset sampling period... Real-time zero-starting-depth reference point Integral remapping calibration is performed to form a depth time series; based on the depth time series data set, a two-way sliding window fitting interpolation method is used to perform a multi-scale velocity estimation task, and the velocity estimation time series V is output. Step S30: Obtain the target velocity reference value sequence Based on the velocity estimation time series V and the target velocity reference value series The speed deviation time series is calculated, and the motor control quantity is adjusted in real time by a fuzzy logic controller (FLC) driven by sliding window trend prediction based on the speed deviation time series, and the cumulative displacement increment of the probe is output. Step S40: When the cumulative displacement increment of the probe exceeds the preset displacement increment threshold, an effective spatial sampling event is immediately triggered; the effective spatial sampling event records the probe triaxial tilt angle value and probe position depth at the time of the event trigger, and writes this set of data into the equidistant sampling point sequence, and outputs the equidistant sampling point sequence; Step S50: For the equidistant sampling point sequence, a dual-channel displacement fusion mechanism based on weighted curvature correction is used to calculate the cumulative horizontal displacement and the cumulative vertical displacement. Finally, the calculation results of the horizontal displacement and the cumulative vertical displacement are uploaded to the terminal device for visualization display through a preset Bluetooth module.

2. The structural displacement monitoring method based on intelligent sensors as described in claim 1, characterized in that, In step S10, the initial tilt angle output signal includes the horizontal initial tilt angle output signal. and vertical initial tilt angle output signal The probe acceleration signal includes the horizontal probe acceleration signal. and vertical probe acceleration signal The static signal of the gyroscope includes the static signal of the horizontal gyroscope. and vertical gyroscope static signal .

3. The structural displacement monitoring method based on intelligent sensors as described in claim 1, characterized in that, In step S10, the initial depth reading is corrected using a reprojection mechanism based on gravity vector offset verification, based on the initial tilt angle signal, probe acceleration signal, and gyroscope static signal, to generate a zero starting depth reference point. The steps specifically include: First, an ideal gravity vector signal is constructed based on the initial tilt angle signal, probe acceleration signal, and gyroscope static signal. Then, the initial attitude offset angle is calculated using the cosine similarity inverse solution method based on the ideal gravity vector signal and the preset standard Earth gravity vector signal. Next, an initial attitude offset angle threshold is set. Based on the initial attitude offset angle and the initial attitude offset angle threshold, an initial rotation matrix is ​​constructed using the perturbation vector iteration method. The initial rotation matrix is ​​used to rotate the tilt attitude of the current inclinometer probe in the local coordinate system to the ideal vertical direction. Then, the gravity vector residual is calculated based on the initial rotation matrix and the ideal gravity vector signal. ; Repeatedly iterate to calculate the gravity vector residual until... If the residual convergence threshold is less than the preset value, a corrected rotation matrix is ​​output. Finally, the initial depth reading is reprojected onto the global coordinate system based on the corrected rotation matrix, and the zero starting depth reference point is output. .

4. The structural displacement monitoring method based on intelligent sensors as described in claim 1, characterized in that, In step S20, during the downward movement of the inclinometer probe, a preset sampling period is used. Real-time zero-starting-depth reference point Perform integral remapping calibration to generate a depth time series; The steps for performing a multi-scale velocity estimation task based on a deep time series dataset using a bidirectional sliding window fitting interpolation method, and outputting a velocity estimation time series V, specifically include: Step S201: Acquire data at a preset sampling period during the downward movement of the inclinometer probe. Depth data groups at intervals Set a sliding window with a length of 2w+1 for the center of the depth data group. Sliding window Internal data is defined as ;in, For a moment The zero starting depth reference point; For a moment The zero starting depth reference point; For a moment The zero-starting depth reference point; n is the length of the depth data group; Step S202: For sliding windows Perform bidirectional polynomial fitting to construct a local depth trend model. Based on local depth trend model The forward depth window sequence and the backward depth window sequence are extracted; Step S203: Based on the extracted forward depth window sequence and backward depth window sequence, construct the target velocity estimate at time t using the Savitzky-Golay filtering fitting method. It outputs the velocity estimation time series V.

5. The structural displacement monitoring method based on intelligent sensors as described in claim 1, characterized in that, In step S30, a target velocity reference value sequence is obtained. Based on the velocity estimation time series V and the target velocity reference value series The steps of calculating the speed deviation time series, and adjusting the motor control quantity in real time using a fuzzy logic controller (FLC) driven by sliding window trend prediction based on the speed deviation time series, and outputting the cumulative displacement increment of the probe, specifically include: Step S301: Obtain the target velocity reference value sequence Based on the velocity estimation time series V and the target velocity reference value series Calculate the velocity deviation time series; calculate the trend acceleration series based on the velocity deviation time series using the smoothing differential method. Step S302: Using the velocity deviation time series and trend acceleration series as input variables, establish a two-dimensional fuzzy rule table for the fuzzy logic controller (FLC); the two-dimensional fuzzy rule table includes the input fuzzy set and the output control command; Step S303: Adjust the motor control quantity in real time based on the output control command and update the position of the inclinometer probe in real time, and finally output the cumulative displacement increment of the probe.

6. The structural displacement monitoring method based on intelligent sensors as described in claim 5, characterized in that, In step S30, the input fuzzy set includes the speed deviation input set and the trend acceleration input set; the output control command includes the speed control set, which includes "strong deceleration", "medium deceleration", "hold", "medium acceleration" and "strong acceleration"; the speed control set is used to represent the adjustment direction and intensity of the motor drive.

7. The structural displacement monitoring method based on intelligent sensors as described in claim 1, characterized in that, Step S50, which involves calculating the cumulative horizontal and vertical displacements using a weighted curvature correction-based dual-channel displacement fusion mechanism for the equally spaced sampling point sequence, specifically includes: For the i-th equally spaced sampling point in the equally spaced sampling point sequence First, the initial horizontal displacement increment of the first channel is calculated using the angle measurement method. and the original vertical displacement increment Then, for equally spaced sampling points The local curvature correction term is calculated using the second-order finite difference method. Based on this local curvature correction term and the angle measurement method, the corrected horizontal displacement increment of the second channel is calculated. and corrected vertical displacement increment ; Curvature-aware fusion weights are set based on local curvature correction terms, and the proportions of the first and second channels in the final fusion result are dynamically adjusted based on the curvature-aware fusion weights to output the fusion displacement increment result. The cumulative horizontal displacement and cumulative vertical displacement of each depth point are obtained by accumulating the fusion displacement increment result segment by segment.

8. A structural displacement monitoring system based on intelligent sensors, applied to the structural displacement monitoring method based on intelligent sensors according to any one of claims 1 to 7, characterized in that, The intelligent sensor-based structural displacement monitoring system includes: The reprojection reference correction module is used to deploy inclinometer probes and a group of intelligent sensors inside the structure under test. It acquires initial tilt angle signals, probe acceleration signals, and gyroscope static signals through these probes and sensors. Based on these signals, a reprojection mechanism based on gravity vector offset verification is used to correct the initial depth readings, generating a zero-starting depth reference point. ; The integral remapping and multi-scale velocity estimation module is used to estimate the velocity during the descent of the inclinometer probe at a preset sampling period. Real-time zero-starting-depth reference point Integral remapping calibration is performed to form a depth time series; based on the depth time series data set, a two-way sliding window fitting interpolation method is used to perform a multi-scale velocity estimation task, and the velocity estimation time series V is output. The fuzzy drive speed deviation compensation control module is used to obtain the target speed reference value sequence. Based on the velocity estimation time series V and the target velocity reference value series The speed deviation time series is calculated, and the motor control quantity is adjusted in real time by a fuzzy logic controller (FLC) driven by sliding window trend prediction based on the speed deviation time series, and the cumulative displacement increment of the probe is output. The trigger-type equidistant sampling and recording module is used to immediately trigger an effective spatial sampling event when the cumulative displacement increment of the probe exceeds the preset displacement increment threshold. The effective spatial sampling event records the probe's triaxial tilt angle and probe position depth at the time of the event triggering, and writes this set of data into the equidistant sampling point sequence, outputting the equidistant sampling point sequence. The dual-channel displacement fusion and Bluetooth upload module is used to calculate the cumulative horizontal displacement and cumulative vertical displacement for the equidistant sampling point sequence using a weighted curvature correction-based dual-channel displacement fusion mechanism. Finally, the calculation results of the horizontal displacement and cumulative vertical displacement are uploaded to the terminal device for visualization display via a preset Bluetooth module.

9. A structural displacement monitoring device based on intelligent sensors, characterized in that, The intelligent sensor-based structural displacement monitoring device includes: a memory, a processor, and an intelligent sensor-based structural displacement monitoring program stored in the memory and executable on the processor. When the intelligent sensor-based structural displacement monitoring program is executed by the processor, it implements the intelligent sensor-based structural displacement monitoring method according to any one of claims 1 to 7.

10. A computer program product, characterized in that, The computer program product includes a structure displacement monitoring program based on intelligent sensors, which, when executed by a processor, implements a structure displacement monitoring method based on intelligent sensors according to any one of claims 1 to 7.

Images