Vertical point displacement deviation control method and system

Abstract

The present application relates to the technical field of survey deviation control, and specifically includes a vertical point displacement deviation control method and system, the method comprising: deploying a three-axis laser displacement sensor at a target vertical point, collecting displacement deviation data and vibration frequency spectrum, and generating a deviation feature set; marking key nodes according to the building architecture, configuring a displacement compensation vector, and based on load mass distribution correction weight coefficients and action timing, driving a hydraulic leveling mechanism to control the displacement deviation of the target vertical point, solving the technical problem of insufficient displacement deviation accuracy of the vertical point of the building affected by foundation settlement and environmental factors, and achieving the technical effect of improving the safety and stability of the overall structure by deploying a high-precision three-axis laser displacement sensor, collecting displacement deviation data and environmental vibration frequency spectrum in real time, configuring a displacement compensation vector and dynamically adjusting weight coefficients to achieve accurate control of displacement deviation.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of survey deviation control, and particularly relates to a vertical point displacement deviation control method and system. BACKGROUND

[0002] With the continuous development of modern engineering and construction technology, the precision requirements for the displacement deviation of vertical points of buildings are higher and higher. In particular, in high-rise buildings, large bridges, reservoir dams and other projects, the stability control of parameters such as the verticality and horizontal displacement of buildings directly relates to the safety and service life of the project. Commonly, GPS measurement of vertical deviation needs to apply leveling technology to obtain the difference between the geodetic datum and the height anomaly difference. Generally, a precision level is used in combination, but due to the restriction of the precision of the instrument itself, the measurement precision is not high.

[0003] In summary, the prior art has the technical problem of insufficient precision of the displacement deviation of vertical points of buildings due to the influence of foundation settlement and environmental factors. SUMMARY

[0004] The present application provides a vertical point displacement deviation control system to solve the technical problem of insufficient precision of the displacement deviation of vertical points of buildings due to the influence of foundation settlement and environmental factors in the prior art.

[0005] In view of the above problems, the technical scheme of the present application is as follows:

[0006] In one aspect of the present application, a vertical point displacement deviation control method is provided, wherein the method comprises: deploying a three-axis laser displacement sensor at a target vertical point, collecting displacement deviation data in the vertical direction and environmental vibration frequency spectrum, and generating a deviation feature set; according to the engineering and construction architecture in which the reference plane corresponding to the target vertical point is located, marking pile foundation key nodes and building core nodes, and configuring a first displacement compensation vector associated with the target vertical point and the pile foundation key nodes, and a second displacement compensation vector associated with the target vertical point and the building core nodes in combination with the deviation feature set; based on the load mass distribution parameters of the engineering and construction architecture, correcting the weight coefficients and action timing of the first displacement compensation vector and the second displacement compensation vector; according to the first displacement compensation vector and the second displacement compensation vector, driving a hydraulic leveling actuator to correct and control the displacement deviation of the target vertical point in combination with the correction result.

[0007] In another aspect of the present application, a displacement deviation control system for a vertical point is provided, wherein the system comprises: a data acquisition module configured to deploy a three-axis laser displacement sensor at a target vertical point, acquire displacement deviation data in the vertical direction and environmental vibration spectrum, and generate a deviation feature set; a compensation vector configuration module configured to mark pile key nodes and building core nodes according to an engineering building framework in which a reference plane corresponding to the target vertical point is located, and configure a first displacement compensation vector associated with the target vertical point and the pile key nodes, and a second displacement compensation vector associated with the target vertical point and the building core nodes, in combination with the deviation feature set; a correction module configured to correct weight coefficients and action timing of the first displacement compensation vector and the second displacement compensation vector based on load mass distribution parameters of the engineering building framework; and a deviation correction control module configured to drive a hydraulic leveling actuator to correct the displacement deviation of the target vertical point in combination with the correction results of the first displacement compensation vector and the second displacement compensation vector.

[0008] In summary, the one or more technical solutions provided in the present application achieve the technical effects of improving the safety and stability of the overall structure by deploying a high-precision three-axis laser displacement sensor to acquire displacement deviation data in the vertical direction and environmental vibration spectrum in real time, configuring displacement compensation vectors and dynamically adjusting weight coefficients, and accurately controlling the displacement deviation. BRIEF DESCRIPTION OF DRAWINGS

[0009] Figure 1 A flowchart of a displacement deviation control method for a vertical point is provided for the present application.

[0010] Figure 2 A structural diagram of a displacement deviation control system for a vertical point is provided for the present application.

[0011] Explanation of reference signs: data acquisition module M100, compensation vector configuration module M200, correction module M300, deviation correction control module M400. DETAILED DESCRIPTION

[0012] Embodiment One

[0013] The present application will be described in detail below with reference to the accompanying drawings, as shown in Figure 1 The present application provides a displacement deviation control method for a vertical point, wherein the method comprises:

[0014] S1: at the target vertical point, deploy a three-axis laser displacement sensor, collect displacement deviation data in the vertical direction and environmental vibration spectrum, and generate a deviation feature set; S2: according to the engineering building framework where the reference plane corresponding to the target vertical point is located, mark the pile foundation key node and the building core node, and combine the deviation feature set to configure a first displacement compensation vector associated with the target vertical point and the pile foundation key node, and a second displacement compensation vector associated with the target vertical point and the building core node.

[0015] Specifically, deploying a three-axis laser displacement sensor at the target vertical point means installing a high-precision sensor that measures the displacement in three directions (X, Y, Z) at the same time at the key vertical point of the building, which can collect displacement deviation data in the vertical direction, record environmental vibration spectrum, and generate a deviation feature set, which is a multi-dimensional data set containing time series data of displacement deviation and vibration spectrum features, for subsequent analysis and compensation. Marking the pile foundation key node and the building core node means determining the key positions of the pile foundation and the core structure related to the vertical point on the reference plane of the building, which are the key points of the building structure stability. The displacement compensation vector represents the direction and size of the displacement adjustment to be applied.

[0016] Execution steps: Install a three-axis laser displacement sensor at the vertical point of the building, which can collect displacement deviation data in the vertical direction and environmental vibration spectrum in real time. For example, in a high-rise building, the sensor collects data at a sampling rate of 10Hz to generate a deviation feature set containing time series of displacement deviation and vibration spectrum features. According to the reference plane of the building, mark the pile foundation key nodes (such as the top and bottom of the pile foundation) and the building core nodes (such as the core tube and shear wall), which are the key points of the building structure stability and can reflect the overall displacement trend of the building.

[0017] Combine the deviation feature set to configure displacement compensation vectors for each target vertical point, pile foundation key node, and building core node. Further, through the displacement deviation time series and vibration spectrum features in the deviation feature set, calculate the first displacement compensation vector associated with the pile foundation key node and the second displacement compensation vector associated with the building core node. The first and second displacement compensation vectors can accurately describe the adjustment direction and size of the displacement deviation. Preferably, the high-precision measurement capability of the three-axis laser displacement sensor can capture small displacement deviations and environmental vibration spectrum, and the generation of the deviation feature set provides a data basis for subsequent intelligent compensation. The marking of the pile foundation key node and the building core node ensures that the configuration of the compensation vector is targeted, improving the accuracy of the correction control.

[0018] S3: correcting the weight coefficients and action timing of the first displacement compensation vector and the second displacement compensation vector based on the load mass distribution parameter of the engineering building framework; S4: driving the hydraulic leveling actuator to correct the displacement deviation of the target vertical point according to the first displacement compensation vector and the second displacement compensation vector combined with the correction result.

[0019] Specifically, the load mass distribution parameter refers to the mass distribution of the building at different structural points, usually in the form of a mass matrix or distribution diagram; the weight coefficient is a parameter in the displacement compensation vector, used to adjust the influence of the compensation vector; the action timing refers to the application order and duration of the compensation vector in time; the hydraulic leveling actuator is a mechanical device that can adjust the position of the building vertical point according to the control signal, usually used for correction control.

[0020] Execution steps: dynamically adjust the weight coefficients and action timing of the first displacement compensation vector and the second displacement compensation vector according to the load mass distribution parameter of the building; further, if a certain area of the building is heavily loaded, increase the weight coefficient of the compensation vector in that area, and adjust the action timing to ensure the compensation effect; if the load mass distribution parameter shows that the top of the building is heavily loaded, adjust the weight coefficient of the top compensation vector from 0.5 to 0.7, and advance the action timing by 10 milliseconds to respond to the displacement deviation more quickly.

[0021] According to the corrected displacement compensation vector, the hydraulic leveling actuator corrects the displacement deviation of the target vertical point; the hydraulic leveling actuator adjusts the position of the vertical point according to the direction and size of the compensation vector; the analysis of the load mass distribution parameter provides detailed information about the building load, so that the weight coefficient and action timing of the compensation vector can be dynamically adjusted according to the actual load situation, and the high-precision control capability of the hydraulic leveling actuator ensures the accuracy and stability of the correction operation.

[0022] Further, the method further comprises:

[0023] Symmetrically arranging a dual-frequency laser interferometer at the four corners of the reference plane; setting a first data synchronization constraint condition based on the dual-frequency laser interferometer; wherein under the condition of a preset sampling rate interval, the time alignment error limited by the first data synchronization constraint condition meets the preset stability requirement.

[0024] Specifically, the dual-frequency laser interferometer is a high-precision measuring instrument that can emit two different frequencies of laser at the same time for measuring physical quantities such as displacement and vibration; the reference plane is a reference surface of a building for positioning and measuring the positions of other points; the data synchronization constraint condition refers to the condition of ensuring the time alignment of different sensor data during data acquisition; the time alignment error refers to the time difference between different sensor data, and the stability requirement refers to the accuracy requirement that the time alignment error after data synchronization must meet.

[0025] The execution steps are: symmetrical arrangement of dual-frequency laser interferometers at the four corners of the reference plane to ensure that each corner can be accurately measured, symmetrical arrangement can provide comprehensive displacement and vibration data, improve the accuracy and reliability of measurement, for example, on a square reference plane with a side length of 10 meters, a dual-frequency laser interferometer is arranged at each corner, which can cover the measurement requirements of the entire plane; based on the dual-frequency laser interferometer, the first data synchronization constraint condition is set to ensure that the time alignment error meets the preset stability requirement within the preset sampling rate interval, preferably, the preset sampling rate is 100Hz, and the time alignment error needs to be controlled within 1ms to ensure the synchronization and stability of the data, through accurate time alignment, the data deviation between different sensors is eliminated, and the overall measurement accuracy is improved; the high-precision measurement capability of the dual-frequency laser interferometer can provide accurate displacement and vibration data, and the data synchronization constraint condition ensures the time consistency of multi-sensor data, thereby improving the measurement accuracy and reliability of the entire system.

[0026] Further, the method further comprises:

[0027] Based on the reference plane, a capacitive micro-displacement probe is arranged along the vertical axis direction; based on the capacitive micro-displacement probe, a second data synchronization constraint condition is set; wherein, under the condition of a preset resolution interval, the frequency domain aliasing suppression ratio limited by the second data synchronization constraint condition meets the preset accuracy requirement.

[0028] Specifically, the capacitive micro-displacement probe detects small displacement through capacitance change; the reference plane is a reference surface of a building for positioning and measuring the positions of other points; the data synchronization constraint condition refers to the condition of ensuring the time alignment of different sensor data during data acquisition; the frequency domain aliasing suppression ratio refers to the ability to suppress aliasing phenomenon in frequency domain analysis, and the aliasing phenomenon is that high-frequency signals are misjudged as low-frequency signals due to insufficient sampling rate; the preset accuracy requirement refers to the accuracy requirement that the frequency domain aliasing suppression ratio after data synchronization must meet.

[0029] The execution step is: on the reference plane, a capacitive micro displacement probe is arranged along the vertical axis direction to measure the micro displacement change in the vertical direction and output high-resolution displacement data; based on the capacitive micro displacement probe, a second data synchronization constraint condition is set to ensure that the frequency domain aliasing suppression ratio meets the preset accuracy requirement in the preset resolution interval, preferably, the preset resolution is 0.1 microns, and the frequency domain aliasing suppression ratio needs to reach more than 40 dB to ensure the reliability and accuracy of the data; through accurate frequency domain aliasing suppression, the influence of high-frequency noise can be eliminated, and the accuracy of the data is improved; the high-resolution measurement capability of the capacitive micro displacement probe can capture the micro displacement change, and the data synchronization constraint condition ensures the time consistency of the multi-sensor data, thereby improving the measurement accuracy and reliability of the entire system.

[0030] Further, a first displacement compensation vector associated with the target vertical point and the pile foundation key node is configured, and the method comprises:

[0031] The reference plane is pre-divided into a plurality of first pre-division space grids by the target vertical point and the pile foundation key node, and a first displacement compensation vector is configured based on the plurality of first pre-division space grids and the deviation feature set in a multi-scale association.

[0032] Specifically, the target vertical point is a key point in the building that needs to be controlled for displacement deviation; the pile foundation key node is an important node in the building pile foundation structure, which is usually closely related to the displacement deviation of the vertical point; the reference plane is a reference plane of the building, which is used to locate and measure the positions of other points; the space grid pre-division refers to dividing the reference plane into a plurality of small space units to more accurately analyze and process data; the deviation feature set is a multi-dimensional data set containing time series data and vibration frequency spectrum features of displacement deviation; the multi-scale association refers to analyzing the correlation between data in different scales (such as time scale and space scale); the displacement compensation vector is a mathematical model for rectification, which represents the displacement adjustment direction and size that needs to be applied.

[0033] The execution step is: by the target vertical point and the pile key node, the reference plane is pre-divided in space grid to obtain a plurality of first pre-division space grids, for example, in a high-rise building, the reference plane can be divided into a 10*10 grid, each grid unit represents a specific space area, and this division can help to more accurately locate and analyze the source of displacement deviation; by analyzing the displacement deviation time sequence and vibration frequency spectrum characteristics in the deviation feature set, combined with the position information of the space grid, the displacement compensation demand of each grid unit is calculated, further, if the displacement deviation of a certain grid unit is 3mm, and the vibration frequency spectrum shows that there is a dominant modal frequency of 15Hz vibration, a compensation vector can be configured, the direction of which is opposite to the deviation direction, and the compensation amount is 3mm, and the time sequence and amplitude of the compensation vector are adjusted considering the influence of the vibration frequency. The pre-division of the space grid can decompose the complex displacement deviation problem into a plurality of small area problems, so that the analysis and processing are more accurate; the multi-scale correlation analysis can comprehensively consider the time and space factors, and improve the accuracy and effectiveness of the displacement compensation vector.

[0034] Further, based on the plurality of first pre-division space grids, the deviation feature set is correlated in multi-scale to configure a first displacement compensation vector, and the method comprises:

[0035] Based on the deviation feature set, long-term dependence features of the displacement deviation time sequence are extracted using a bidirectional LSTM network, wherein the number of hidden layer nodes of the bidirectional LSTM network is N; based on the plurality of first pre-division space grids, a three-dimensional convolution kernel is used to capture the spatial correlation mode, and the feature channel is expanded to M dimensions, M

[0036] Specifically, the bias feature set is a multi-dimensional data set containing time series data of displacement bias and vibration frequency spectrum features for subsequent analysis and compensation; the bidirectional LSTM network is a neural network capable of simultaneously processing forward and reverse information of time series data, suitable for extracting long-term dependence features; the number of hidden layer nodes refers to the number of nodes in the hidden layer of the neural network, which determines the complexity and processing capacity of the network; the three-dimensional convolution kernel is a convolution kernel used to capture the correlation pattern of three-dimensional spatial data, commonly used in image and spatial data processing; feature channel expansion refers to expanding feature data to a higher dimension to enhance feature expression capability; the gated attention mechanism refers to dynamically adjusting feature weights through a gating unit and an attention coefficient, used to highlight important features; the attention coefficient is a weight value that measures the importance of a feature, with a higher attention coefficient indicating that the feature is more important in the current task.

[0037] The execution steps are as follows: based on the bias feature set, the bidirectional LSTM network is used to extract long-term dependence features of the displacement bias time series, the bidirectional LSTM network can simultaneously process forward and reverse information of the time series, capturing more comprehensive features, thus more accurately predicting the trend of displacement bias; based on multiple first pre-divided spatial grids, a three-dimensional convolution kernel is used to capture spatial correlation patterns, the three-dimensional convolution kernel can analyze the correlation of spatial grids in three dimensions, such as displacement correlation in X, Y, Z directions, and the feature channel is expanded to M dimensions, preferably M < N, which can more efficiently express spatial features.

[0038] Based on the long-term dependence features of the displacement bias time series and the spatial correlation patterns, a gated attention mechanism is set up, which dynamically adjusts the weights of the first displacement compensation vector through an attention coefficient, highlighting the correction features. If the displacement bias features of a certain spatial grid show strong long-term dependence in the time series, the gated attention mechanism will assign a higher attention coefficient, enhancing the influence of the feature in the compensation vector. The powerful time series processing capability of the bidirectional LSTM network can capture the long-term trend of displacement bias, the spatial analysis capability of the three-dimensional convolution kernel can extract the correlation patterns of spatial grids, and the gated attention mechanism can dynamically adjust the feature weights, highlighting important correction features.

[0039] Further, the method comprises:

[0040] Based on multiple first pre-divided spatial grids, the dominant modal frequency in the vibration frequency spectrum is extracted; when the frequency offset of the dominant modal frequency from the natural frequency is less than a set frequency threshold, a dynamic vibration absorber is activated, and the mass block position of the dynamic vibration absorber is adjusted in real time by a servo motor.

[0041] Specifically, the vibration spectrum is a representation method to describe the vibration characteristics of an object. It usually converts time-domain signals into frequency-domain signals through Fourier transform, showing the amplitudes of different frequency components. In the vibration spectrum, the frequency component with the largest amplitude is the dominant modal frequency. The natural vibration frequency of a structure under no external force is the inherent frequency, which is determined by the physical properties of the structure (such as mass and stiffness). The frequency offset refers to the difference between the current vibration frequency and the inherent frequency. The set frequency threshold is a pre-set frequency difference. When the actual frequency offset is less than the threshold, a specific control action is triggered. The dynamic vibration absorber is a device that can absorb vibrations of a specific frequency by adjusting its parameters (such as mass, stiffness, and damping). The servo motor is an electric motor that can accurately control position, speed, and acceleration, used in situations that require high-precision adjustment.

[0042] Execution steps: Based on multiple first pre-division spatial grids, analyze the vibration spectrum, extract the dominant modal frequency, specifically, in high-rise buildings, by analyzing the vibration spectrum of different spatial grids, it is found that the dominant modal frequency is 10 Hz, which indicates that the building vibrates most significantly at this frequency. Calculate the frequency offset between the dominant modal frequency and the inherent frequency of the building. If the inherent frequency is 10.2 Hz and the dominant modal frequency is 10 Hz, the frequency offset is 0.2 Hz. When the frequency offset is less than the set frequency threshold (such as 0.3 Hz), activate the dynamic vibration absorber. The dynamic vibration absorber changes its resonance frequency by adjusting the position of its mass block, so that it matches the external excitation frequency, thereby absorbing vibration energy.

[0043] The position of the mass block of the dynamic vibration absorber is adjusted in real time by the servo motor. The servo motor accurately adjusts the position of the mass block according to the control signal to ensure that the resonance frequency of the dynamic vibration absorber matches the current vibration frequency. The vibration spectrum analysis can accurately identify the dominant modal frequency, and the frequency offset detection ensures that measures are taken in time when the inherent frequency is approached. The combination of dynamic vibration absorber and servo motor can adjust the parameters of the vibration absorber in real time, effectively suppressing vibration.

[0044] Further, the position of the mass block of the dynamic vibration absorber is adjusted in real time by the servo motor, and the method comprises:

[0045] Through PID control, determine the target displacement of the mass block, wherein the integral term weight coefficient is negatively correlated with the absolute value of the frequency offset; drive the servo motor to move the mass block to the coordinate position corresponding to the target displacement, and combine the energy decay rate to optimize the amplitude and phase parameters of the target displacement in reverse, until the resonance frequency of the dynamic vibration absorber matches the external excitation frequency.

[0046] Specifically, PID control (Proportional-Integral-Derivative Control) is a common feedback control algorithm that adjusts the output of a system through proportional, integral, and derivative terms; the integral term weight coefficient is the weight coefficient of the integral term in PID control, used to adjust the influence of the integral term on the control output; the absolute value of the frequency offset is the absolute value of the difference between the current vibration frequency and the target frequency; a servo motor is an electric motor that can accurately control position, speed, and acceleration; the target displacement is the displacement that the dynamic absorber mass needs to move to; the energy decay rate is the speed at which vibration energy decays over time, and is commonly used to evaluate the vibration suppression effect; the amplitude and phase parameters are parameters that describe the amplitude and phase of a vibration signal, used to optimize the performance of a dynamic absorber.

[0047] The execution step: through the PID control algorithm, the target displacement of the mass is determined according to the absolute value of the frequency offset, wherein the integral term weight coefficient is negatively correlated with the absolute value of the frequency offset, that is, the smaller the frequency offset, the lower the weight coefficient of the integral term, specifically, if the frequency offset is 0.2 Hz, the integral term weight coefficient is set to 0.5; if the frequency offset decreases to 0.1 Hz, the integral term weight coefficient decreases to 0.3, and this negative correlation ensures that the control action is more gentle when approaching the target frequency, avoiding excessive adjustment.

[0048] According to the target displacement determined by the PID control, the servo motor is driven to move the mass to the corresponding target position, and the servo motor can accurately adjust the position of the mass with high precision (such as 0.01 mm), ensuring that the resonance frequency of the dynamic absorber matches the external excitation frequency; in combination with the energy decay rate, the amplitude and phase parameters of the target displacement are optimized in reverse, and the energy decay rate reflects the speed of reduction of vibration energy, and by optimizing the amplitude and phase parameters, the performance of the dynamic absorber is further improved, and by optimizing the amplitude and phase parameters, the energy decay rate can be improved, thereby suppressing vibration more quickly. The PID control can dynamically adjust the target displacement of the mass, ensuring that the resonance frequency of the dynamic absorber matches the external excitation frequency. The high-precision adjustment capability of the servo motor ensures the accurate movement of the mass, and the optimization of the energy decay rate further improves the vibration suppression effect.

[0049] Further, a second displacement compensation vector associated with the target vertical point and the building core node is configured, and the method comprises:

[0050] The reference plane is pre-divided into a plurality of second pre-division space grids through the target vertical point and the building core node, and a second displacement compensation vector is configured based on the plurality of second pre-division space grids and the deviation feature set in multiple scales.

[0051] Specifically, the target vertical point refers to a key point in the building that needs to be controlled for displacement deviation; the building core node refers to a key point of the core structure of the building, which is usually closely related to the stability and deformation of the structure; the reference plane refers to a reference plane of the building, which is used to locate and measure the positions of other points; the spatial grid pre-division refers to dividing the reference plane into multiple small spatial units for more accurate analysis and processing of data; the deviation feature set contains a multi-dimensional data set of time series data of displacement deviation and vibration frequency spectrum characteristics; the displacement compensation vector is used for the mathematical model of rectification, indicating the direction and size of displacement adjustment that needs to be applied.

[0052] The execution steps are: performing spatial grid pre-division on the reference plane based on the target vertical point and the building core node, obtaining multiple second pre-division spatial grids, each second pre-division spatial grid representing a specific spatial area; based on the divided second pre-division spatial grid, performing multi-scale correlation with the deviation feature set to configure a second displacement compensation vector. Specifically, by analyzing the displacement deviation time series and vibration frequency spectrum characteristics in the deviation feature set, combined with the position information of the spatial grid, the displacement compensation demand of each grid unit is calculated, and the time sequence and amplitude of the compensation vector are adjusted considering the influence of vibration frequency. Spatial grid pre-division can decompose complex displacement deviation problems into multiple small area problems, making analysis and processing more accurate. Multi-scale correlation analysis can consider time and space factors comprehensively, improving the accuracy and effectiveness of the displacement compensation vector.

[0053] Further, based on the multiple second pre-division spatial grids, the multi-scale correlation with the deviation feature set is performed to configure a second displacement compensation vector, and the method comprises:

[0054] Constructing a topological relationship graph of the building core node, wherein each node in the topological relationship graph represents a building core structure point, and each edge in the topological relationship graph represents the connection stiffness of the structure; performing feature extraction on the topological relationship graph to generate a structure topological embedding pointer; performing cross-modal fusion of the structure topological embedding pointer and the deviation feature set to assign correlation weights corresponding to the multiple second pre-division spatial grids; based on the multi-scale features after cross-modal fusion, generating a second displacement compensation vector; the compensation direction of the second displacement compensation vector is inversely matched with the structure deformation trend of the building core node.

[0055] Specifically, the building core node refers to a key point of a building core structure, which is usually closely related to the stability and deformation of the structure; the topological relationship graph is used to represent the connection relationship and stiffness between the building core nodes; the structural connection stiffness is used to describe the rigidity degree of the connection between the building core nodes, which is usually represented by a numerical value; the correlation weight represents the importance of different spatial grids in the fused features; the compensation direction is the direction of the displacement compensation vector, which is used for correction; the structural deformation trend refers to the deformation direction and degree of the building core node under stress or vibration.

[0056] The execution steps are as follows: constructing a topological relationship graph of building core nodes, wherein the nodes represent building core structure points, and the edges represent structural connection stiffness. Specifically, in a high-rise building, key structure points such as core tubes and shear walls are taken as nodes, and the connection stiffness between the nodes is taken as the weight of the edges to construct the topological relationship graph of the building core nodes, which intuitively represents the connection relationship and stiffness distribution of the building core structure; performing feature extraction on the topological relationship graph to generate a structural topological embedding pointer. Further, each node in the topological relationship graph represents a building core structure point, and each edge in the topological relationship graph represents a structural connection stiffness, which is converted into an embedding pointer through a graph neural network (GNN).

[0057] The structural topological embedding pointer and the deviation feature set are cross-modally fused, and the correlation weights corresponding to a plurality of second pre-divided spatial grids are assigned. Specifically, through an attention mechanism, the weights of each spatial grid are assigned according to the correlation between the topological features and the deviation features, and the grids with higher weights play a greater role in compensation; based on the multi-scale features after cross-modal fusion, a second displacement compensation vector is generated. The compensation direction is reversely matched with the structural deformation trend of the building core node to ensure that the compensation vector can effectively correct the structural deformation. The topological relationship graph can accurately represent the connection and stiffness distribution of the building core structure, the feature extraction and embedding pointer generation enable the topological information to be fused with the deviation feature set, and the cross-modal fusion and weight assignment ensure the pertinence and effectiveness of the compensation vector.

[0058] In summary, the beneficial effects of the embodiments of the present application are as follows:

[0059] The three-axis laser displacement sensor is deployed at the target vertical point to collect displacement deviation data and environmental vibration spectrum in the vertical direction and generate a deviation feature set. According to the engineering building structure where the reference plane corresponding to the target vertical point is located, key nodes of a pile foundation and core nodes of a building are marked, and a first displacement compensation vector associated with the target vertical point and the key nodes of the pile foundation and a second displacement compensation vector associated with the target vertical point and the core nodes of the building are configured in combination with the deviation feature set. The weight coefficients and action timing of the first displacement compensation vector and the second displacement compensation vector are corrected based on the load mass distribution parameters of the engineering building structure. The displacement deviation of the target vertical point is controlled for correction according to the first displacement compensation vector and the second displacement compensation vector in combination with the correction result to drive a hydraulic leveling actuator. The displacement deviation control method and system for a vertical point are provided, high-precision three-axis laser displacement sensors are deployed to collect displacement deviation data and environmental vibration spectrum in the vertical direction in real time, displacement compensation vectors are configured, and weight coefficients are dynamically adjusted to realize accurate control of displacement deviation and improve the safety and stability of the overall structure.

[0060] Embodiment Two

[0061] Based on the same inventive concept as the displacement deviation control method for a vertical point in the foregoing embodiments, as shown in Figure 2 the embodiment of the present application provides a displacement deviation control system for a vertical point, wherein the system comprises:

[0062] A data acquisition module M100 is configured to deploy a three-axis laser displacement sensor at a target vertical point to collect displacement deviation data and environmental vibration spectrum in the vertical direction and generate a deviation feature set.

[0063] A compensation vector configuration module M200 is configured to mark key nodes of a pile foundation and core nodes of a building according to the engineering building structure where the reference plane corresponding to the target vertical point is located, and configure a first displacement compensation vector associated with the target vertical point and the key nodes of the pile foundation and a second displacement compensation vector associated with the target vertical point and the core nodes of the building in combination with the deviation feature set.

[0064] A correction module M300 is configured to correct the weight coefficients and action timing of the first displacement compensation vector and the second displacement compensation vector based on the load mass distribution parameters of the engineering building structure.

[0065] A correction control module M400 is configured to control the displacement deviation of the target vertical point for correction according to the first displacement compensation vector and the second displacement compensation vector in combination with the correction result to drive a hydraulic leveling actuator.

[0066] Further, the data acquisition module M100 is further configured to perform the following method:

[0067] A dual-frequency laser interferometer is arranged in a four-corner symmetry of the reference plane; a first data synchronization constraint condition is set based on the dual-frequency laser interferometer; wherein, under the condition of a preset sampling rate interval, a time alignment error limited by the first data synchronization constraint condition meets a preset stability requirement.

[0068] Further, the data acquisition module M100 is also used to execute the following method:

[0069] A capacitive micro-displacement probe is arranged along a vertical axis direction based on the reference plane; a second data synchronization constraint condition is set based on the capacitive micro-displacement probe; wherein, under the condition of a preset resolution interval, a frequency domain aliasing suppression ratio limited by the second data synchronization constraint condition meets a preset accuracy requirement.

[0070] Further, the compensation vector configuration module M200 is used to execute the following method:

[0071] A plurality of first pre-partitioned space grids are obtained by pre-partitioning the reference plane in space through the target vertical point and the pile key node; a first displacement compensation vector is configured based on the plurality of first pre-partitioned space grids and the deviation feature set through multi-scale association.

[0072] Further, the compensation vector configuration module M200 is also used to execute the following method:

[0073] Long-term dependence features of displacement deviation time series are extracted using a bidirectional LSTM network based on the deviation feature set, wherein the number of hidden layer nodes of the bidirectional LSTM network is N; a three-dimensional convolution kernel is used to capture a spatial correlation pattern based on the plurality of first pre-partitioned space grids, and the feature channel is expanded to M dimensions, M

[0074] Further, the compensation vector configuration module M200 is also used to execute the following method:

[0075] A dominant modal frequency in a vibration frequency spectrum is extracted based on the plurality of first pre-partitioned space grids; when a frequency offset of the dominant modal frequency from an inherent frequency is less than a set frequency threshold, a dynamic vibration absorber is activated, and a mass block position of the dynamic vibration absorber is adjusted in real time by a servo motor.

[0076] Further, the compensation vector configuration module M200 is also used to execute the following method:

[0077] The target displacement amount of the mass is determined through PID control, wherein the integral term weight coefficient is negatively correlated with the absolute value of the frequency offset; the servo motor drives the mass to move to the coordinate position corresponding to the target displacement amount, and the amplitude and phase parameters of the target displacement amount are inversely optimized in combination with the energy attenuation rate until the resonance frequency of the dynamic vibration absorber matches the external excitation frequency.

[0078] Further, the compensation vector configuration module M200 is configured to perform the following method:

[0079] The reference reference plane is pre-divided into a plurality of second pre-divided space grids through the target vertical point and the building core node, and a second displacement compensation vector is configured based on the plurality of second pre-divided space grids and the deviation feature set.

[0080] Further, the compensation vector configuration module M200 is further configured to perform the following method:

[0081] A topological relationship graph of the building core node is constructed, wherein each node in the topological relationship graph represents a building core structure point, and each edge in the topological relationship graph represents a structure connection stiffness; a structure topological embedding pointer is generated by performing feature extraction on the topological relationship graph; the structure topological embedding pointer is cross-modally fused with the deviation feature set to assign an associated weight corresponding to the plurality of second pre-divided space grids; a second displacement compensation vector is generated based on the multi-scale features after cross-modal fusion; and a compensation direction of the second displacement compensation vector is reversely matched with a structure deformation trend of the building core node.

[0082] In summary, any step can be stored in a computer memory without limitation as computer instructions or programs, and can be called and recognized by a computer processor without limitation, and no redundant limitation is made herein.

[0083] Further, the above technical solutions only represent preferred technical solutions of the technical solutions of the embodiments of the present application, and some variations of certain parts made by a person skilled in the art also represent the principles of the novel embodiments of the present application. Obviously, a person skilled in the art can make various modifications and variations to the present application without departing from the scope of the present application.

Claims

1. A method for controlling displacement deviation at a vertical point, characterized in that, The method includes: At the target vertical point, a triaxial laser displacement sensor is deployed to collect displacement deviation data in the vertical direction and environmental vibration spectrum, and generate a deviation feature set; Based on the engineering building structure where the reference plane corresponding to the target vertical point is located, mark the key nodes of the pile foundation and the core nodes of the building, and in combination with the deviation feature set, configure the first displacement compensation vector associated with the target vertical point and the key nodes of the pile foundation, and the second displacement compensation vector associated with the target vertical point and the core nodes of the building. Based on the load mass distribution parameters of the engineering building structure, the weight coefficients and application sequence of the first displacement compensation vector and the second displacement compensation vector are corrected. Based on the first displacement compensation vector and the second displacement compensation vector, and combined with the correction result, the hydraulic leveling actuator is driven to correct the displacement deviation of the target vertical point. The method for configuring a first displacement compensation vector associated with the target vertical point and the key node of the pile foundation includes: By using the target vertical point and key nodes of the pile foundation, the reference plane is pre-divided into spatial grids to obtain multiple first pre-divided spatial grids; Based on multiple first pre-divided spatial grids, multi-scale association is performed with the deviation feature set to configure a first displacement compensation vector; The method for configuring a second displacement compensation vector associated with the target vertical point and the building core node includes: Using the target vertical point and the core building node, the reference plane is pre-divided into spatial grids to obtain multiple second pre-divided spatial grids. Based on multiple second pre-divided spatial grids, a multi-scale association is performed with the deviation feature set to configure a second displacement compensation vector.

2. The displacement deviation control method for vertical points as described in claim 1, characterized in that, The method further includes: Dual-frequency laser interferometers are symmetrically arranged at the four corners of the reference plane; Based on the dual-frequency laser interferometer, a first data synchronization constraint condition is set; Among them, under the condition of the preset sampling rate range, the time alignment error under the first data synchronization constraint meets the preset stability requirement.

3. The displacement deviation control method for vertical points as described in claim 2, characterized in that, The method further includes: Based on the aforementioned reference plane, a capacitive micro-displacement probe is configured along the vertical axis. Based on the capacitive micro-displacement probe, a second data synchronization constraint condition is set; Among them, under the condition of a preset resolution range, the frequency domain aliasing suppression ratio limited by the second data synchronization constraint meets the preset accuracy requirement.

4. The displacement deviation control method for vertical points as described in claim 1, characterized in that, Based on multiple first pre-divided spatial grids, a first displacement compensation vector is configured by multi-scale association with the deviation feature set and configuring the first displacement compensation vector. The method includes: Based on the aforementioned deviation feature set, a bidirectional LSTM network is used to extract long-term dependency features of the displacement deviation time series, wherein the number of hidden layer nodes in the bidirectional LSTM network is N. Based on multiple first pre-divided spatial grids, a three-dimensional convolutional kernel is used to capture spatial correlation patterns, and the feature channels are expanded to M dimensions, where M < N; Based on the long-term dependency characteristics of the displacement deviation time series and combined with the spatial correlation pattern, a gating attention mechanism is set to use the attention coefficient to perform prominent correction control on the first displacement compensation vector.

5. The displacement deviation control method for vertical points as described in claim 4, characterized in that, The method includes: Based on multiple first pre-divided spatial grids, the dominant mode frequencies in the vibration spectrum are extracted; When the frequency offset between the dominant modal frequency and the natural frequency is less than a set frequency threshold, the dynamic vibration absorber is activated, and the position of the mass block of the dynamic vibration absorber is adjusted in real time by a servo motor.

6. The displacement deviation control method for vertical points as described in claim 5, characterized in that, The position of the mass block of the dynamic vibration absorber is adjusted in real time by a servo motor, and the method includes: The target displacement of the mass block is determined by PID control, where the integral term weight coefficient is negatively correlated with the absolute value of the frequency offset. The servo motor is driven to move the mass block to the coordinate position corresponding to the target displacement, and the amplitude and phase parameters of the target displacement are optimized in reverse by combining the energy attenuation rate until the resonant frequency of the dynamic vibration absorber matches the external excitation frequency.

7. The displacement deviation control method for vertical points as described in claim 1, characterized in that, Based on multiple second pre-divided spatial grids, a second displacement compensation vector is configured by multi-scale association with the deviation feature set and the second displacement compensation vector is configured. The method includes: Construct a topology graph of the core nodes of the building, wherein each node in the topology graph represents a core structural point of the building, and each edge in the topology graph represents the structural connection stiffness. Feature extraction is performed on the topological relationship graph to generate a structural topological embedding pointer; The structural topology embedding pointer is fused with the deviation feature set across modes, and association weights are assigned to multiple second pre-divided spatial grids. A second displacement compensation vector is generated based on the multi-scale features obtained after cross-modal fusion. The compensation direction of the second displacement compensation vector is matched in the opposite direction to the structural deformation trend of the building core node.

8. A displacement deviation control system for a vertical point, characterized in that, For implementing the vertical point displacement deviation control method according to any one of claims 1-7, the system comprises: The data acquisition module is used to deploy a triaxial laser displacement sensor at the target vertical point to collect displacement deviation data in the vertical direction and environmental vibration spectrum, and generate a deviation feature set. The compensation vector configuration module is used to mark the key nodes of the pile foundation and the core nodes of the building according to the engineering building structure where the benchmark reference plane corresponding to the target vertical point is located, and to configure the first displacement compensation vector associated with the target vertical point and the key nodes of the pile foundation, and the second displacement compensation vector associated with the target vertical point and the core nodes of the building, in combination with the deviation feature set. The correction module is used to correct the weight coefficients and application timing of the first displacement compensation vector and the second displacement compensation vector based on the load mass distribution parameters of the engineering building structure. The correction control module is used to drive the hydraulic leveling actuator to correct the displacement deviation of the target vertical point based on the first displacement compensation vector, the second displacement compensation vector, and the correction result.

Images