50 results about "Displacement gradient" patented technology

The invention discloses a bridge health prediction device based on a Beidou system and a monitoring analysis method, and relates to the field of bridge health monitoring. The method comprises the following steps: S1, deploying monitoring points at key structure nodes of a bridge, and collecting three-dimensional position information and multi-source data by using a Beidou system and a multi-type sensor; s2, constructing a time-varying displacement field and a continuous deformation tensor, fusing stress and strain to construct a structural fatigue factor, and introducing environmental disturbance to realize dynamic correction of the fatigue factor; s3, a health state function is constructed based on the stress response, the displacement gradient and the deformation tensor, a damage mapping function is constructed in combination with accumulated deformation and a historical threshold value, and a damage thermodynamic diagram is generated; and S4, finally fusing fatigue, health and damage indexes to evaluate bridge risks, and performing dynamic early warning. Through fusion of mechanical tensor features and environmental disturbance modeling, a multi-index linkage fatigue and damage assessment mechanism is constructed, and the precision, robustness and global perception ability of bridge health monitoring are improved.

The invention relates to the technical field of target detection, in particular to a photovoltaic panel detection method and system based on machine vision, and the method comprises the following steps: obtaining the displacement data of a photovoltaic panel cell, calculating the displacement gradient of a connection point, dividing a rigid region and a non-rigid region, generating a deformation classification result, and constructing a global deformation field. Generating a time sequence deformation trend matrix; calculating a displacement mean value of a bracket fixing point; adjusting a deformation compensation parameter; generating an adaptive compensation parameter, adjusting a welding point position, calculating an abnormal region Euclidean distance, and generating a defect matching identification result. According to the method, rigid and non-rigid areas are accurately divided by calculating the displacement gradient of a connection point, a global deformation field is constructed in combination with the displacement change of a feature point, trend continuity is ensured, a compensation parameter is adjusted by utilizing a fixed point displacement mean value, a deformation increment factor is calculated, the weight and stability are corrected, the position of a welding point is adjusted, and inverse deformation correction is optimized. The Euclidean distance is combined to match defects, so that the detection precision and stability are improved.

The invention provides a building construction site safety monitoring and early warning method and system based on the Internet of Things, and relates to the technical field of data processing.The method comprises the steps that a three-base-point dynamic topological structure is established in a target monitoring domain, a first datum point is arranged at the extension end of the axis vertical line of a slewing bearing of a tower crane, and a second datum point is arranged at the extension end of the axis vertical line of the slewing bearing of the tower crane; the second datum point is embedded in the radial section of a force bearing flange plate of the hydraulic jacking module, the third datum point is fixedly connected to the intersection joint of a deep foundation pit top beam and a supporting system, and a non-collinear vector set is constructed through three-point coordinate offset; and based on the non-collinear vector group, deformation compensation correction is performed on the dynamic calibration parameter set through a vector space topological transformation algorithm, a topological reconstruction data stream containing a displacement gradient tensor and strain energy density is generated, and a risk situation feature space is constructed based on the topological reconstruction data stream. According to the invention, the intelligent construction safety management is realized.

The invention provides a shaft end nut locking position identification and tightening control method, and belongs to the technical field of industrial intelligent manufacturing, and the method comprises the steps: installing a high-precision inductive displacement sensor to monitor the axial displacement of a nut in real time; constructing a tightening control system of an integrated torque measurement and displacement monitoring unit; collecting real-time displacement data and eliminating interference through a sliding window filtering algorithm; identifying a nut locking critical point based on a displacement gradient analysis method; calculating an optimal torque function according to the shaft end contact mechanical mathematical model; the locking torque is accurately adjusted by adopting a temperature compensation type torque control method; a multi-mode locking intelligent identification model is used for analyzing locking process parameters and dynamically adjusting a tightening control parameter set, and accurate identification of an initial contact position and intelligent optimization adjustment of tightening parameters in the shaft end nut locking process are achieved.

The invention discloses a multi-source 5G message video cooperative monitoring method and system, particularly relates to the technical field of video monitoring, and is used for solving the problems of multi-camera calibration parameter misalignment and video fusion dislocation caused by thermal deformation asynchronism in the prior art. A temperature sensor is arranged on an industrial camera supporting structure, a thermal deformation three-dimensional displacement compensation vector is calculated, a preferential transmission object is marked according to the displacement gradient of an adjacent camera and a material heat conductivity coefficient threshold value, and the displacement vector is transmitted to an edge calculation node according to the priority through a 5G message. All displacement data are fused at a node end to dynamically correct space calibration parameters, a collaborative calibration matrix is generated, and finally, the matrix is utilized to perform real-time registration on multiple paths of video streams and perform workpiece defect detection, so that the geometric consistency of a multi-camera system in a thermal deformation environment is guaranteed, and the visual detection precision and stability are improved.

The invention relates to the technical field of material detection, in particular to a solid wood furniture material detection method and system.The method comprises the following steps that form data of a furniture bearing component under the shearing-bending loading condition is obtained, wood connection node imaging is collected, crack displacement is calculated according to a loading sequence, and a fiber bonding point load value is extracted; and calculating the furniture crack growth rate. According to the method, form data is collected through shear-bending loading, crack displacement and fiber bonding point load calculation is achieved, the crack propagation rate is more accurate, the pixel displacement gradient and the maximum shear stress gradient area are combined, the stress concentration point recognition precision is improved, the crack path is tracked based on the local shear stress gradient, and the interlayer energy consumption rate is calculated. Extracting a critical state, identifying a crack instability point, improving damage evolution prediction precision, and improving bending resistance and shearing resistance evaluation comprehensiveness and quality detection reliability in combination with multi-layer stress change and gradient calculation.

The invention relates to the technical field of intelligent scheduling, in particular to a steel structure weld fatigue cracking real-time monitoring system based on machine vision, which is characterized in that an image acquisition module acquires a high-stability image sequence in a strong vibration environment; the displacement field reconstruction module adopts a frequency domain-space domain combined decomposition technology to isolate environmental noise and material non-uniformity interference in a breakthrough manner, and an anti-noise deformation field and a stress distribution diagram are synchronously output through dual-channel parallel processing; a damage response module accurately locks sub-pixel-level crack initiation coordinates based on strain energy accumulation rate mutation, and dynamically predicts a yield risk area in combination with a displacement gradient diffusion model; the early warning module intelligently triggers graded response according to the crack position and the space coupling state of the risk area, performs positioning maintenance in a low risk area and automatic strengthening in a high risk area, and outputs heat treatment scheme coordinates when no crack exists; the system solves the problems of false detection caused by vibration noise and lagging response of single-parameter monitoring, and active protection from damage initiation to failure blocking is realized.

This invention provides a method and system for monitoring surrounding rock deformation in underground coal mines based on three-dimensional point clouds, relating to the field of coal mine safety monitoring. The invention includes the following steps: installing a reference target in the monitoring area and collecting point cloud data; preprocessing the point cloud data, performing coarse registration using the reference target and fine registration using the iterative nearest point algorithm, and generating a triangular mesh model using the Poisson surface reconstruction algorithm; parameterizing the triangular mesh model of the monitoring area into a two-dimensional analysis mesh, calculating the three-dimensional displacement vector of each node, and establishing a displacement field matrix; calculating the displacement gradient tensor based on the displacement vector, deriving the Green-Lagrange strain tensor, and obtaining the full-field strain distribution; recording the strain value of each mesh node, and determining whether an instability warning is triggered based on the relationship between the strain value and the warning threshold; this method can achieve high-precision monitoring of surrounding rock deformation in underground coal mines.

The invention relates to the technical field of mine slope deformation monitoring, in particular to a mine slope absolute displacement field monitoring and error self-correcting method and system. The method comprises the following steps: according to Kriging synergetics interpolation and adaptive weighted fusion, processing GB-SAR relative displacement, a projection coefficient matrix and a motion mode partition after atmospheric correction to obtain an absolute displacement field calibrated by a GNSS (Global Navigation Satellite System); processing the displacement period characteristic parameters and the area displacement gradient field by utilizing multi-characteristic fuzzy comprehensive evaluation to obtain a slope local stability comprehensive index field; and high-precision absolute displacement field monitoring and error self-correction are completed by combining a GNSS early warning level, an absolute displacement field calibrated by the GNSS and a local stability comprehensive index and utilizing graded early warning and displacement field traceability analysis. According to the invention, high-precision planar absolute displacement field monitoring, accurate displacement abnormal cause identification and error self-correction can be realized.

The present invention relates to the field of image data processing technology, and discloses a large-scale dangerous rock mass identification method based on digital image deformation processing, which aims to solve the problem of low efficiency and accuracy of the existing methods. The scheme mainly includes: obtaining two images of the target rock mass collected within two cycles after preprocessing, and generating two-dimensional grayscale matrices of the two images in the spatial domain respectively; performing two-dimensional discrete Fourier transform on the two-dimensional grayscale matrices of the two images respectively, calculating the cross-correlation function of the two images, calculating the first phase difference corresponding to the frequency domain according to the cross-correlation function, and determining the second phase difference corresponding to the spatial domain according to the first phase difference; obtaining the displacement field after phase demodulation according to the second phase difference, and calculating the displacement gradient and strain field according to the displacement field; determining the danger level of each area of ​​the target rock mass based on the pre-trained dangerous rock mass identification model. The present invention improves the efficiency and accuracy of dangerous rock mass identification, and is particularly suitable for large-scale dangerous rock mass identification.

The invention discloses a bridge crack sub-pixel identification method and system based on moire phase unwrapping, and relates to the field of structural health monitoring. The method comprises the steps that a bridge image is collected and uploaded to a system, and the image is processed to obtain a de-noised image; performing sub-pixel registration on the reference grating image of the denoised image to obtain a registered reference grating image; using the registered reference grating image to generate a moire image to calculate a global mass fraction, calculating a wrapped phase by the moire image greater than a threshold value, otherwise, requiring the unmanned aerial vehicle to perform re-sampling; calculating an unwrapping phase for the moire image greater than the threshold value based on the pixel weight, and deducing a displacement field of each pixel by using the unwrapping phase; and combining the displacement gradient amplitude and the level set function to identify a crack region and quantify crack information. Through automatic moire analysis and displacement field calculation, the bridge tiny crack is accurately detected, the size and depth of the crack are quantified, and reliable support is provided for bridge health monitoring and maintenance decision making.

The invention relates to the technical field of image data processing, discloses a large-scale dangerous rock mass identification method based on digital image deformation processing, and aims to solve the problems of low efficiency and accuracy of the existing method, and the scheme mainly comprises the following steps: obtaining two images of a target rock mass collected in two periods after preprocessing, two-dimensional gray matrixes of the two images in a spatial domain are generated respectively; performing two-dimensional discrete Fourier transform on the two-dimensional gray matrixes of the two images, calculating a cross-correlation function of the two images, calculating a first phase difference corresponding to a frequency domain according to the cross-correlation function, and determining a second phase difference corresponding to a spatial domain according to the first phase difference; obtaining a displacement field after phase demodulation according to the second phase difference, and calculating a displacement gradient and a strain field according to the displacement field; and determining the danger level of each region of the target rock mass based on a pre-trained dangerous rock mass identification model. The method improves the efficiency and accuracy of dangerous rock mass recognition, and is especially suitable for large-scale dangerous rock mass recognition.

The invention discloses a multi-source information fusion distribution transformer state evaluation method and system, and the method comprises the steps: collecting multi-source data, and obtaining a state quantity data set through preprocessing; constructing an electromagnetic thermal coupling model based on the state quantity data set, and outputting space physical field distribution; constructing a causal chain through displacement gradient mutation detection based on space physics field distribution, and outputting coordinates of a fault intrinsic source point; extracting a state response parameter by positioning a fault point based on the coordinates of the fault intrinsic source point, and obtaining an equipment health index through a critical physical parameter; and obtaining a risk score of the distribution transformer according to the equipment health index, and generating a state assessment report through a multi-dimensional risk assessment algorithm. According to the method, the electromagnetic thermal full-coupling model is constructed, and the constitutive equation is fused by adopting the cross-field energy analysis method, so that the hidden defect detection rate and the service life prediction accuracy of the distribution transformer are improved, the service cycle of key power equipment is prolonged, and the operation and maintenance cost is reduced.

The invention provides an atomic scale deformation field two-gradient decoupling method based on geometric phase analysis. The method comprises the following steps: (1) obtaining original data after crystal deformation through a high-resolution transmission electron microscope image; (2) performing Fourier transform on the original data to obtain reciprocal space information, selecting diffraction spots in the reciprocal space information of different crystal orientations to perform inverse Fourier transform to obtain phase information of a local lattice, and obtaining an atomic displacement field in a crystal grain based on the phase information; (3) decomposing the atomic displacement field into a first-order affine displacement gradient tensor and a second-order non-affine displacement gradient tensor, and respectively decomposing the first-order affine displacement gradient tensor and the second-order non-affine displacement gradient tensor to obtain deformation components of respective basic localization events; and (4) defining the participation degree of the atomic scale basic localization event, and analyzing the atomic scale physical mechanism of microdefect evolution. The method is reasonable in conception, and can effectively solve the multi-scale problem of regulation and control of macroscopic performance by a current material microscopic evolution mechanism.

The application discloses a kind of coal rock nonlinear fracture propagation length quantitative determination method based on multiple parameters, comprising: preparation of coal rock sample containing notch;Spray black and white speckle on the surface of sample, install acoustic emission sensor, clip type extensometer and displacement sensor;Carry out three-point bending peak after cycle loading and unloading fracture experiment, record load, crack opening, force point displacement, acoustic emission waveform and DIC image;Through COD-P and LPD-P curve calculation mechanics flexibility, back calculation to solve stress constraint crack length;From acoustic emission positioning, obtain the spatial distribution histogram of breakage dissipation energy, combined with DIC horizontal displacement gradient threshold cloud picture, determine equivalent crack tip and length;Further calculate coal rock fracture process zone length, and intuitively show nonlinear fracture propagation path.The application provides a kind of crack dynamic change quantitative determination method based on mechanics flexibility-displacement gradient-energy dissipation distribution multiple physical parameters, reduce the influence of human subjective consciousness, and the measurement result is accurate and reliable.

The present application relates to the technical field of medical elastic imaging and material multi-scale mechanical measurement, and particularly discloses an elastic imaging method and system based on a scattering displacement gradient field, which comprises the following steps: generating a scattering displacement field by arranging an excitation device around a target material, measuring an out-of-plane displacement gradient field by using transverse shear interference imaging, selecting a single-frame instantaneous gradient field or a frequency-domain gradient field obtained by time-frequency domain Fourier transform as a target field; obtaining a global shear wave velocity distribution by adaptive sliding window interception, two-dimensional autocorrelation calculation and Rayleigh wave spatial autocorrelation theoretical model fitting, and then reconstructing viscoelastic modulus distribution by using an inversion algorithm. The present application supports single-frame scattering displacement gradient field inversion elastic imaging, can be used without high-frame-rate equipment, has strong anti-environmental vibration interference ability and high system stability, is suitable for scenes with poor isolation conditions, has low cost, high efficiency and high measurement precision, and can meet the requirements of medical clinical diagnosis and material multi-scale mechanical measurement.

The method comprises the following steps: collecting first motion data of a conveying belt based on master-slave double detection modules on both sides of a guide groove; determining displacement gradient and fluctuation intensity of the conveying belt according to magnetic grating data and Hall array data in the first motion data; determining second motion data of the conveying belt within a predicted time length according to the displacement gradient and the fluctuation intensity; determining guide wheel adjustment attributes related to the conveying belt based on a relationship between the second motion data and a preset offset threshold, and adjusting the offset of the conveying belt based on the guide wheel adjustment attributes. The technical scheme of the embodiment of the present disclosure adjusts the offset of the conveying belt based on the magnetic grating data collected by the master detection module and the Hall array data collected by the slave detection module, realizes early prediction of the offset trend, effectively avoids wear and tear and jamming and other faults caused by the offset of the conveying belt, and achieves the effect of improving the running stability of the equipment.

The application provides a simulation analysis method for compression deformation characteristics of a rubber pad, and relates to the field of rubber pad product testing. The application constructs a finite element grid model based on geometric parameters and initial material parameters of the rubber pad, identifies geometric singularity regions with displacement gradient mutation; applies a test loading load on the model, extracts displacement path curvature and performs gradient correction on material parameters, and constructs a local material constitutive relation; performs compression deformation simulation based on the local constitutive relation, analyzes strain energy accumulation sequence and determines an energy transmission control region; determines a dominant constraint mode in combination with the spatial corresponding relationship between the energy transmission control region and the geometric singularity region, and performs incremental correction on the simulation result, so as to realize accurate prediction of non-uniform compression deformation characteristics of the rubber pad.

The present application relates to the technical field of structural health monitoring, in particular to an air building factory structural health monitoring method and system, in the present application, the node stress difference and displacement gradient comparison is established based on the load change rate and vibration offset, the stress state of the structure is formed into a time sequence stress response matrix, the stress dependence rate is calculated and introduced into a dynamic Bayesian network to establish the node time sequence relationship, the confidence level updating process is gradually iterated and smoothed by correcting the cycle load difference and the reaction force difference, the confidence distribution change rate is screened to select the nodes that have fallen below the threshold, the curve difference and the shape similarity are compared, and the long short-term memory network is used to predict the curve trend, so that the degradation trend can be revealed in advance and unified as a clustering trajectory, avoiding scattered results caused by abnormal judgment, and the degradation boundary recognition has high sensitivity and interval determination ability through the extraction of the curve slope, inflection point and span, combined with the rate comparison and cycle attenuation length fitting.

The invention relates to the technical field of geotechnical engineering, and particularly discloses a full-surface strain analysis method and device based on a discrete element model, and the method comprises the steps: constructing the discrete element model of a target region, and calculating the displacement data of each particle in the target region through the discrete element model when the discrete element model deforms; dividing the target area into a plurality of grids, and calculating the displacement data of each grid datum point according to the displacement data of the particles in each grid; constructing a continuous displacement field according to the displacement data of each grid datum point, and calculating a displacement gradient tensor of each grid datum point; calculating a Green Lagrange strain tensor according to the displacement gradient tensor; calculating the maximum principal strain of each grid according to the Green Lagrange strain tensor; calculating a non-uniform coefficient of the target area according to the standard deviation and the average value of the maximum principal strain of each grid in the target area; and determining an instability risk judgment result of the target area according to the non-uniform coefficient. According to the method, a stable strain result can be obtained under a relatively large deformation situation.

This invention relates to the field of process control technology, specifically to a digital management and control method and system for the entire process of cement pole finished products. The method includes the following steps: collecting the pressure difference between the steam pipeline and the cavity and combining it with the centrifugal radius to generate a control parameter set; calculating the real-time rotational speed proportional error to obtain the offset and generate an operational correction instruction set; collecting the displacement gradient during the molding cycle to assess the uniformity of density; and comparing and analyzing the collected temperature and humidity data to correct valve parameters and generate a control instruction set. In this invention, the overall production process is reconstructed by dynamically adjusting the centrifugal speed and curing temperature and humidity. A tracking list containing key process characteristics is constructed, and equipment status consistency checks are performed. This completely breaks down the lag barriers of manual inspection and paper-based records, achieving multi-stage closed-loop collaboration and adaptive error compensation of the entire process's operating data. This significantly improves the overall stability of the finished product manufacturing quality and the level of traceability and control throughout the entire product process.

The invention discloses a thermal strain imaging method, a thermal strain imaging system and a thermal strain imaging device, and belongs to the field of ultrasonic imaging. Extracting baseline displacement of the feature points; finding a typical frame of the detection object at the end of contraction and diastole as a global reference frame i0; finding out a typical frame in at the last stage of contraction and relaxation of the detection object in each group of heated data In; calculating a correlation coefficient between the typical frame in and the global reference frame i0, calculating a relative displacement between the typical frame in and the global reference frame i0, and generating a displacement gradient map Gn according to the correlation coefficient and the relative displacement; converting the displacement gradient map into a strain map to obtain a thermal strain image of the cross section of the detected object, and detecting the offset of tissues such as water-containing tissues and lipids in echo signals by utilizing the characteristic of echo displacement gradient generated by sound velocity change caused by temperature rise, so as to judge the change gradient distribution of sound velocity and judge the tissue components. The tissue and organ detection accuracy is improved.

In one example, a method includes determining a feature displacement gradient (DG) tensor characterizing a deformation feature. The method includes, in an iterative routine, calculating an objective function value associated with a trial DG tensor, the calculating based at least in part on pixel values within an automatically determined ZOC, where the ZOC is based at least in part on the trial DG tensor. In another example, a method includes determining one or more rotational transformation components corresponding to a test image, and determining a feature DG tensor based at least in part on the rotational transformation components. In another example, a system includes an electro-optical assembly, a detector assembly, and a controller programmed to determine a feature DG tensor. In another example, a non-transitory computer-readable medium stores instructions that, when executed by a computer, cause the computer to perform a method of determining a deformation characteristic.

The invention relates to the technical field of material nondestructive testing, and discloses a deformation non-uniformity detection method for an oversize high-strength steel material. A detection point array is constructed on the surface of the super-huge high-strength steel member, interference measurement is performed on all detection points in an unloaded state and a loaded state respectively, two groups of interference intensities are obtained, and significant deformation points are marked by differential combination with a jump threshold value; the method comprises the following steps: normalizing interference intensity, extracting a phase difference, carrying out global phase unwrapping, converting the phase difference into a component surface normal displacement data set according to an interference displacement relationship, calculating a displacement difference of adjacent points according to dot matrix geometry to obtain a local displacement gradient, calculating an average value and a standard deviation of the gradient to form a global non-uniformity index, and setting a self-adaptive threshold value; and the gradient of each point is compared to identify non-uniform deformation risk points and output a detection result, so that surface deformation information is uniformly collected, real deformation and noise are distinguished, interference is reduced, and a risk area is identified in a component range.

The present invention discloses a thermal strain imaging method, system and device, which belongs to the field of ultrasonic imaging. The method collects ultrasonic images of the detection object before and after heating; extracts the baseline displacement of the characteristic points; finds the typical frames of the detection object at the end of contraction and diastole as the global reference frame i0; finds the data of each group after heating i1. n Typical frames i of the detected object at the end of systole and diastole n ; Calculate typical frame i n The correlation coefficient between the global reference frame i0 and the typical frame i is calculated n The relative displacement with the global reference frame i0 generates the displacement gradient map G according to the correlation coefficient and relative displacement n ; Convert the displacement gradient map into a strain map to obtain thermal strain imaging of the cross section of the test object. Utilize the echo displacement gradient caused by the change in sound velocity due to temperature increase to detect the offset of water-containing tissues and lipids in the echo signal, thereby determining the gradient distribution of sound velocity changes, discriminating tissue composition, and improving the accuracy of tissue and organ detection.

This invention relates to the field of 3D modeling technology, specifically a construction method for a large cantilevered steel frame canopy based on BIM and 3D scanning. The method includes the following steps: collecting component point cloud coordinates and calculating deformation displacement gradients to construct a 3D data model; extracting reference surfaces and calculating projection distance differences to generate a surface correction model; decomposing node position residuals to calculate correction ratio coefficients and generate offset vector data; and constructing a connection plate constraint model based on the offset vectors to calculate micro-correction values ​​and generate a control coordinate set. In this invention, spatial topography is reconstructed by calculating point cloud displacement gradients; deviation characteristics are mined by extracting reference surfaces and calculating differences to generate deformation parameters; node spatial position residuals are decomposed to identify offset patterns and dynamically calculate compensation ratios; a constrained deformation field distribution matrix is ​​constructed by combining the geometric boundaries of the connection plates; and finally, precise control coordinates are generated to eliminate accumulated positioning errors and interference risks, thereby significantly improving assembly reliability.

In an example, a method includes determining a characteristic displacement gradient (DG) tensor that characterizes a deformation characteristic. The method includes, in an iterative routine, calculating an objective function value associated with a trial DG tensor based, at least in part, on pixel values within a zone of consideration (ZOC) that is determined automatically based at least in part on the trial DG tensor. In another example, a method includes determining one or more rotational transformation components corresponding to a test image and determining a characteristic DG tensor based at least in part on the rotational transformation component(s). In another example, a system includes an electron optics assembly, a detector assembly, and a controller programmed to determine a characteristic DG tensor. In another example, a non-transitory computer readable medium stores instructions which, when executed by a computer, cause the computer to perform a method of determining a deformation characteristic.

The application provides a high-speed global deformation measurement method and system, comprising the following steps: S1, the camera and the lens optical axis are perpendicular to the sample surface, the digital images of the sample in different states are recorded by using the camera, an auxiliary global continuous displacement field and a constraint condition are introduced during image processing, and the displacement and displacement gradient of the field in each subset area are consistent with the local DIC calculation result; S2, the related function is modified based on the augmented Lagrangian function and the constraint condition; S3, the solution of the global problem is found by using the alternating direction multiplier method (ADMM) to coordinate the local sub-problems, so that the problem is iteratively solved; and S4, the fields of the quantities of interest including strain and velocity are derived according to the global displacement field. The application uses the augmented Lagrangian function, the calculation efficiency is much higher than that of the global DIC algorithm based on FEM, and the global displacement field obtained by introducing the constraint condition through the augmented Lagrangian function has continuity.

The present application relates to the technical field of machine learning, in particular to a surrounding rock deformation prediction method based on machine learning and array displacement meter data. The method obtains displacement time series formed by array displacement meters at different depth measuring points, carries out time alignment and incremental processing on the displacement time series, and generates measuring point state sequences containing cumulative displacement, displacement increment, deformation rate, deformation acceleration and displacement gradient of adjacent measuring points; constructs a space-time deformation graph according to the depth sequence relationship of the measuring points and the lag correlation relationship between the measuring points, inputs the space-time deformation graph into a machine learning model, and generates a surrounding rock state hidden vector through graph feature coding and time sequence feature coding. The present application constrains the prediction process by time evolution relationship, depth proximity relationship and lag propagation relationship, reduces the trend disconnection and gradient distortion of the prediction results of multiple measuring points in the depth direction, and improves the consistency of the surrounding rock deformation prediction results and the actual deformation transmission process.

This invention provides a geological disaster monitoring method and system based on UAV vision and target collaboration, belonging to the field of disaster monitoring technology. The method includes: acquiring geological images at different times; performing terrain modeling based on the geological images and target locations to obtain a digital surface model; identifying vegetated and bare areas based on the geological images; constructing a Delaunay triangulation based on targets in the vegetated areas to calculate vegetation displacement; calculating bare land displacement based on the digital surface model; fusing vegetation and bare land displacements using thin-plate spline functions to obtain global displacement information; detecting cracks using an improved Canny-Zernike operator based on the digital surface model and identifying crack skeleton lines using a morphological thinning algorithm; and modeling displacement gradient tensors based on crack skeleton lines and global displacement information to construct a crack opening distribution map and obtain monitoring results. This invention creatively combines physical noise reduction, intelligent fusion, and precise analysis to achieve high-precision positioning in harsh environments.