3603 results about "Surface roughness" patented technology

The invention provides an equipment supervision data real-time analysis method and system, and the method comprises the steps: calculating an energy value of a frequency band corresponding to a sub-band signal, determining the energy characteristics of the surface roughness and dust deposition distribution uniformity of a blade in each sub-band in a square sum form, and generating energy distribution vectors which are arranged according to a frequency band sequence; performing time-frequency analysis on the vibration signal by using the acquired real-time rotating speed information of the gas turbine, identifying a frequency component corresponding to the eccentric distance value of the rotor and the vibration frequency offset, and determining the amplitude of the frequency component; according to the classification result and the real-time rotating speed information, the comprehensive influence index of the material density change rate and the ash deposition layer thickness difference is calculated, the fault early warning level is determined through a weighted average method, and an early warning signal is generated; and according to the change trend of the early warning signal and the blade edge wear degree, a random forest algorithm is adopted to predict a development curve of the vibration frequency offset in the future target time, and real-time equipment supervision analysis data are obtained.

The invention relates to the field of abrasion morphology determination and abrasion loss calculation, and discloses an abrasion morphology and contact stress analysis and prediction method and system based on a rough contact surface, a medium and equipment, and the method comprises the steps: constructing a dynamic contact model, and obtaining an accurate movement rule and a stress analysis result of a moving mechanical component; constructing a three-dimensional rough surface digital twinborn model conforming to an actual contact surface; establishing a contact stress field multi-physical field analysis model to obtain elastic deformation and a friction heat effect; based on the contact stress distribution and the local wear rate which are obtained in real time, carrying out dynamic reconstruction on the wear morphology, and establishing a nonlinear iteration relation between the wear loss and the contact stress distribution; determining an accurate boundary condition based on an accurate movement rule and a stress analysis result of the moving mechanical component, and constructing a failure criterion based on contact stress gradient mutation and surface roughness evolution; key parameters are calibrated through experiments, and a mapping relation database of wear rate-stress state-morphological characteristics is established so as to carry out friction wear analysis.

The invention discloses a cleaning control method and system based on an electronic cleaning agent, and the method comprises the steps: obtaining the distribution characteristics of pollutants and a material sensitivity judgment result through obtaining the micron-sized particle distribution data, chemical residue concentration data and surface roughness parameters of the pollutants on the surface of an electronic workpiece; according to the pollutant distribution characteristics and the material sensitivity judgment result, matching a candidate formula from a preset cleaning agent formula database, calculating pollutant removal rate predicted values of different concentration and time combinations through a linear regression algorithm, and determining optimized cleaning agent formula parameters and cleaning strength control parameters; and in the cleaning process, the residual quantity of pollutants and the surface damage degree are collected in real time, if the residual quantity of the pollutants exceeds a preset threshold value T1 or the surface damage degree exceeds a preset threshold value T2, cleaning agent injection pressure and cleaning time parameters are dynamically adjusted, and updated cleaning execution parameters are generated. Accurate and efficient cleaning of the surface of the electronic workpiece is achieved.

The invention discloses a numerical control machining equipment state monitoring system based on big data analysis, and relates to the technical field of intelligent manufacturing. The method is used for solving the problems of correlation analysis and real-time feedback between tool wear and machining quality in the machining process. A time sequence correlation feature vector is extracted through a vibration signal, an acoustic emission signal and main shaft axial micro-displacement data, a machining quality degradation index is calculated in combination with a nonlinear regression model, and the machining quality is monitored in real time. And on the basis of adaptive feature weight adjustment of the types of the processing materials, feature fusion during processing of different materials is enhanced, and the precision of quality evaluation is improved. And establishing a nonlinear mapping relation between the tool wear and the surface roughness by utilizing a gradient lifting tree model, and realizing accurate prediction of the tool wear and the machining quality. And finally, through a closed-loop dynamic adjustment mechanism, the cutting speed and the feeding amount are automatically adjusted according to the machining quality degradation index and the fault positioning result, and the stability of the machining process and the product quality are ensured.

The invention relates to the technical field of material testing, and discloses a coating performance simulation method based on high and low temperature circulation and prestress loading. The method comprises the following steps: constructing a coating matrix composite geometric model containing a microporous structure by acquiring actually measured working condition data of a pipeline or a part, and quantifying surface roughness and interface curvature characteristics in combination with reverse engineering; a dynamic temperature-pressure coupling equation is generated based on the thermal expansion effect of a high-pressure medium, a micropore optical load is converted into equivalent heat flux density through a ray tracing algorithm, transient boundary conditions of a temperature field and a stress field are associated, a crack initiation path is simulated through an extended finite element method, and a life attenuation curve is predicted in combination with a Paris formula. And material parameters are dynamically corrected based on experimental data, and a multi-parameter collaborative optimization model is generated. According to the method, the problems of interface stress evaluation distortion and life prediction deviation under the combined load are solved, and multi-physics field co-simulation support is provided for coating compression resistance optimization.

The invention relates to the technical field of industrial machine vision detection and image processing, in particular to an automobile part production mold surface smoothness detection system based on image enhancement, which comprises a data acquisition module used for acquiring an original grayscale image of the surface of an automobile part mold; performing low-pass filtering processing on the original grayscale image to eliminate imaging thermal noise; the manifold reconstruction module is used for constructing a structure tensor field; reversely deducing a pseudo-curvature field of the mold surface; the adaptive enhancement module is used for generating a corrected image; constructing a texture orthotropic diffusion model; generating a texture reconstruction reference image; the surface metering module is used for calculating the difference between the corrected image and the texture reconstruction reference image and generating a defect saliency image; calculating the surface roughness value of the mold surface; according to the method, the problem that design textures and abnormal scratches are difficult to distinguish in the prior art is effectively solved, and the technical bottleneck that micro defects are easily missed in a complex geometric structure in traditional visual detection is overcome.

The invention relates to the technical field of material removal in laser processing, in particular to a laser positioning stripping system and method for special-shaped neodymium iron boron waste, and the method comprises the following steps: S1, carrying out multispectral scanning on the surface of the special-shaped neodymium iron boron waste; s2, dynamically adjusting energy parameters of laser lift-off, and constructing an energy thickness matching model; s3, generating a fractal scanning path of the adaptive special-shaped surface based on the geometric contour data of the three-dimensional characteristic spectrum; s4, in-situ surface roughness detection is carried out in the laser lift-off process, a secondary finishing mode is triggered according to a detection result, and the safe lift-off allowance is dynamically corrected based on lift-off depth data; and S5, in-situ separation of the oxide layer fragments and the matrix particles is achieved through the hot air flow field in combination with the gas-solid separation dynamics principle, and matrix particle recovery is assisted and strengthened through the magnetic field. The laser energy, the scanning path and the thermal control strategy can be dynamically adjusted according to real-time data, so that material loss and waste are effectively avoided, and the processing efficiency and the recovery rate are improved.

The invention provides an intelligent control method for a floodlight illumination system, and the method comprises the steps: obtaining an environment surface roughness coefficient, spectral reflection distribution and spatial geometric change data, determining the wall surface material absorptivity and shielding region coordinates, and obtaining environment reflection characteristics and shielding distribution data; analyzing shielding boundary sharpness and shielding angle distribution according to the spatial geometric change and the coordinates of the shielding area, calculating a luminous flux distribution curve of a local dark area, and judging whether the influence range of the dark area exceeds a preset threshold range or not; calculating the illumination coverage integrity according to the updated light field distribution and polarization reflection characteristics, determining the boundary sharpness of a light beam, and obtaining a light field uniformity parameter; and according to the light field uniformity parameter and the shielding thermal radiation effect, calculating and optimizing light field parameters.

The invention provides a surface roughness measuring method and device based on a multispectral laser speckle image, and belongs to the field of surface quality visual inspection. Comprising the following steps: respectively acquiring speckle images formed by irradiating the surface of a workpiece with lasers of N wavelengths at different angles, and constructing a multispectral laser speckle data set; the speckle images of N wavelengths at a single angle are spliced according to channels and subjected to feature fusion, and then a multispectral laser speckle fusion image is generated; the fused images of the multiple angles form an image sequence, and single images in the sequence are subjected to multi-scale feature extraction and fusion to generate single-angle features; the multi-angle features generate global illumination information through a multi-angle feature aggregation module, and finally a roughness value is calculated through a multi-layer perceptron. Multi-spectral laser speckle coding surface features are adopted, multi-angle information is fused through the multi-angle feature aggregation neural network, and the relation between the light source wavelength, the incident angle and the surface three-dimensional shape and roughness is comprehensively considered, so that high-precision measurement is achieved.

The invention provides a surface quality monitoring method and system based on image-signal multi-modal data, and relates to the technical field of machining process detection, and the method comprises the steps: obtaining a workpiece surface image and a main shaft vibration acceleration signal in a milling process, inputting the workpiece surface image and the spindle vibration acceleration signal into a surface roughness classification model to obtain a machining state recognition result; wherein in the surface roughness classification model, shallow texture features of a workpiece surface image are extracted through an image processing module, and frequency domain features of a spindle vibration acceleration signal are extracted through wavelet transform and a frequency attention mechanism; and splicing projection features of the shallow texture features and the frequency domain features into a combined feature vector, carrying out dynamic weight distribution and fusion on the combined feature vector based on a self-adaptive fusion strategy of a gated attention mechanism to obtain weighted fusion features, inputting the weighted fusion features into a grade classifier, and outputting the grade of the surface roughness.

The invention is suitable for the technical field of performance detection, and provides a method for detecting the ablation resistance of a burning-resistant nozzle and an arc contact, and the method comprises the steps: collecting a temperature field matrix, an arc image sequence, an electrical parameter time sequence and a surface displacement curve of an ablation region, and generating a multi-modal data set; based on the data set, extracting spatial features of the ablation core region through a cavity convolution layer and a channel attention mechanism; inputting a preset bidirectional LSTM network, fusing thermal stress data of the COMSOL simulation model to drive gating weight update, and extracting time sequence evolution characteristics; generating a space-time fusion feature vector containing the temperature field gradient, the arc form change rate and the surface roughness evolution feature; inputting the space-time fusion feature vector, the thermal stress distribution data and the material phase change threshold parameter into a constructed ablation prediction model for training; the ablation prediction model after training is completed is optimized; and a performance prediction result is generated in real time by using the optimized ablation prediction model, so that the detection accuracy is effectively improved.

Disclosed in the present invention are a PET release film base film having low surface roughness and a preparation method therefor. The release film base film comprises three layers which are stacked: a core layer and two surface layers located on both sides of the core layer. The release film base film is obtained via three-layer co-extrusion by means of taking an auxiliary extrusion resin as a raw material of the two surface layers and a primary extrusion resin as a raw material of the core layer; the auxiliary extrusion resin is obtained by mixing an anti-blocking agent PET masterbatch and a PET base material in a mass ratio of 1:5-1:15; and the primary extrusion resin is a pure PET base material. The release film base film of the present invention has low surface roughness, low haze and excellent overall performance. Since large-particle benzoic acid-modified nano aluminum oxide has the refractive index very close to that of PET base films, the present invention can effectively avoid adverse visual impacts caused by a difference in the refractive index at interfaces between nanoparticles having large particle diameters and resin matrixes, thus effectively reducing the haze.

The invention provides a collaborative tool servo diamond turning method and system based on track pre-filtering, and the method comprises the steps: S1, integrating a fast tool shaft to a machine tool shaft in the cutting depth direction, and constructing a master-slave collaborative tool servo system; wherein the machine tool shaft serves as a main servo shaft of the master-slave cooperative tool servo system, and a low-frequency tool path obtained after reference tool path pre-filtering is executed; the fast cutter shaft is used as a slave servo shaft of the master-slave collaborative cutter servo system, and the deviation between the actual displacement of the master servo shaft and the reference cutter path is compensated in real time; s2, implementing a collaborative tool servo machining process based on a master-slave collaborative tool servo system; and S3, after machining is completed, surface contour data are obtained, and quantitative evaluation is conducted on a machining result through surface shape errors and surface roughness indexes.

The invention provides a stamping part surface roughness detection method and system based on image recognition, and the method comprises the steps: building a surface contour database according to a metal square lattice standard model stamped by a texturing die, obtaining the actual contour data of the surface of a stamping part, comparing the difference between the actual contour and the metal square lattice standard model, and determining the surface roughness of the stamping part. The method comprises the following steps: obtaining a microscopic deformation feature distribution image, optimizing the microscopic deformation feature distribution image, calculating probability distribution of each pixel point belonging to real surface roughness or microscopic deformation, obtaining a continuous microscopic deformation area mask, calculating a deviation between an actual contour and an ideal surface as a corrected roughness value, and aiming at the corrected roughness value, obtaining a microscopic deformation feature distribution image; according to the characteristics of the stamping part, the surface is divided into a plurality of characteristic areas, when new stamping part surface data is input, the surface roughness grade is judged through the surface quality evaluation model, and the stamping part surface roughness detection method is beneficial to improving the stamping part surface roughness detection precision.

The invention relates to the technical field of precision machining, and particularly discloses a complex curved surface precision machining method based on tool path optimization, which comprises the following steps: S1, acquiring a geometric model of a curved surface to be machined and tool kinetic parameters; s2, the optimal machining step length, the cutter roll angle and the feeding speed are calculated in real time according to the geometric features of the curved surface and the dynamic parameters of the cutter; s3, generating a self-adaptive tool path based on the dynamic optimization parameters in the step S2; s4, synchronously monitoring the cutting load in the machining process, and dynamically correcting cutter path parameters according to load feedback; and S5, outputting the optimized tool path instruction to the numerical control machine tool to execute machining. By calculating the Gaussian curvature and the average curvature of the curved surface in real time, dynamically adjusting the machining step length and the feeding speed, and automatically reducing the step length and reducing the feeding speed in a curvature sudden change area, the residual height of a steep area is controlled within a precise standard range, and the surface roughness is reduced by more than 30% compared with that of a traditional method; and the consistency of machining precision of each area of the complex curved surface is ensured.

The invention provides a robot milling surface roughness prediction method and system based on parallel convolution and coordinate attention mechanism feature fusion. According to the method, firstly, multi-source heterogeneous high-frequency signals in the milling process of a robot are synchronously collected, empty slice segments are eliminated and environmental noise is filtered out by utilizing a self-adaptive threshold value and a robust statistical method, and a standardized time sequence sample set is constructed; secondly, establishing a multi-source heterogeneous feature fusion roughness prediction model, and independently extracting deep features of cutting force and vibration signals by adopting a double-branch parallel convolution architecture; a coordinate attention mechanism is introduced, heterogeneous modal features are mapped into a virtual two-dimensional topological structure, the dynamic relative contribution degree of cutting force and vibration signals changing along with the machining state is captured through pooling aggregation along the modal dimension, and an attention weight map is generated to conduct dynamic weighting on the features. And finally, outputting a surface roughness predicted value through a regression network. According to the method, the problem that an existing fusion method neglects the dynamic dependency relationship between physical quantities is effectively solved, the anti-interference capability of the robot in the weak rigidity machining process is enhanced, and the precision and robustness of surface roughness prediction are remarkably improved. The method can be widely applied to robot complex component machining quality analysis and intelligent optimization, and has high efficiency and reliability.

The invention belongs to the field of metal materials, and particularly relates to a high-strength hot-rolled pickled steel plate for an automobile chassis and a production method thereof. The steel plate is mainly suitable for manufacturing automobile chassis structural parts. The steel comprises the following chemical components: C, Si, Mn, Al, V, Ni, Cu, Cr, Pt and Y, P is limited to be less than or equal to 0.015%, S is limited to be less than or equal to 0.005%, N is limited to be less than or equal to 0.006%, and the balance is Fe and inevitable impurities. The yield strength of the steel plate ranges from 498 MPa to 552 MPa, the tensile strength ranges from 608 MPa to 643 MPa, the transverse elongation A ranges from 33% to 38%, the steel plate is qualified after transverse cold bending by 180 degrees D = a, the surface quality is good, oxide scale stripes and chromatic aberration defects are avoided, the surface roughness Ra ranges from 1.10 micrometers to 1.35 micrometers, and the steel plate has excellent corrosion resistance.

The invention relates to the technical field of optic nerve detection, in particular to a plastic particle quality detection method and system based on optic nerves, and the method comprises the following steps: capturing a plastic particle image through an industrial camera, carrying out the balance processing of the image contrast, extracting the brightness and color features of the image, and removing the noise interference; and adjusting the image brightness and optimizing the particle edge to obtain a particle optimization image. According to the invention, the accuracy and automation level of plastic particle quality detection are greatly enhanced by using the visual neural network and the image processing technology, and the edge and shape features of the particles can be accurately extracted from a complex background by automatically adjusting the image contrast and brightness and applying an advanced edge detection algorithm. According to the method, the visual quality of the image is optimized, tiny flaws such as cracks and bubbles of the particles can be effectively recognized and analyzed, and more detailed data support can be provided compared with a traditional method by accurately calculating the surface roughness and texture uniformity of the particles.

The invention discloses a carbon fiber cloth reinforced damaged concrete structure detection and analysis method, and relates to the technical field of concrete structure health monitoring, and the method comprises the steps: extracting a high-risk point set from a three-dimensional damage distribution map, carrying out the weighted K-means clustering to construct a Voronoi subspace, calculating the failure probability, and dividing an emergency repair block set; generating a processing queue by using the repairing blocks according to priorities, planning a spiral path of the plasma spray gun by adopting a TSP algorithm, and generating a surface roughness distribution diagram in combination with laser confocal and white light interference detection; according to the roughness partition matching epoxy resin-carbon fiber dynamic matching scheme, a pressure thermodynamic diagram is generated through finite element simulation and electrode time-sharing electrification. Through fusion of the damage evolution graph and the digital twin model, the plasma etching parameters and the conductivity of the carbon fiber cloth are dynamically adjusted, so that the tensile strength of the repaired area is improved, and full-life-cycle health monitoring is realized.

The invention relates to the technical field of machining, in particular to an intelligent control system and method for sapphire substrate polishing, and the method comprises the following steps: S1, surface topography data acquisition: collecting surface roughness data and a three-dimensional topography distribution diagram of a sapphire substrate in real time; s2, generating characteristic parameters: extracting the surface waviness, the local recess depth, the bulge density, the lattice distortion index and the residual stress distribution gradient as the characteristic parameters; s3, dynamic parameter adjustment: generating a dynamic adjustment instruction of the polishing pressure, the rotating speed of the polishing disk and the supply rate of the polishing solution; s4, distribution optimization of the polishing solution: realizing uniform coverage of the polishing solution on the surface of the substrate; s5, stress real-time monitoring: generating a stress abnormity early warning signal; s6, outputting a closed-loop control instruction: feeding back a stress abnormity early warning signal to the adaptive control model, and iteratively correcting the dynamic adjustment instruction; according to the method, the risks of stress concentration and fatigue damage can be reflected in real time, and the safety and high efficiency of the polishing process are ensured.

The invention belongs to the field of metal materials, and particularly relates to a 448 MPa grade hot-rolled pickled steel plate for an automobile and a production method of the 448 MPa grade hot-rolled pickled steel plate. The steel plate is mainly suitable for manufacturing automobile chassis structural parts. The steel comprises the following chemical components: C, Si, Mn, Al, Nb, Ta, Zr and Y, P is limited to be less than or equal to 0.015%, S is limited to be less than or equal to 0.005%, N is limited to be less than or equal to 0.003%, and the balance is Fe and inevitable impurities. The yield strength of the steel plate is larger than or equal to 308 MPa, the tensile strength is larger than or equal to 448 MPa, the transverse elongation A is larger than or equal to 42%, the hole expansion rate is larger than or equal to 125%, transverse cold bending is conducted by 180 degrees, D = a, the steel plate is qualified, the surface quality is good, oxide scale stripes and chromatic aberration defects are avoided, the surface roughness Ra is 1.00-1.40 micrometers, and the steel plate has good corrosion resistance.

The invention relates to an aluminum oxide CMP polishing solution for rough polishing of silicon carbide wafers and a polishing method of the aluminum oxide CMP polishing solution, and the aluminum oxide CMP polishing solution comprises the following components in percentage by weight: 2-10wt% of nano aluminum oxide abrasive; 0.2 to 2 wt.% of a dispersant; 0.001 to 0.01 wt% of a pH regulator; the concentration of the oxidizing agent is 1-5wt.%. By adopting the compound oxidant, potassium permanganate is utilized to oxidize a reduction product manganese dioxide on the surface of the silicon carbide wafer to activate potassium persulfate, so that a large amount of sulfate free radicals are generated in the polishing solution, the oxidability of the sulfate free radicals is extremely high, the polishing solution can be promoted to generate chain oxidation reaction, and the surface of the silicon carbide wafer is further oxidized; and an oxide layer with low hardness is formed on the surface of the silicon carbide and is finally removed under the mechanical action of the aluminum oxide abrasive, so that the ultra-smooth and damage-free surface of the silicon carbide wafer is formed. The surface roughness of the obtained optimal silicon carbide wafer is 0.125 nm, and the material removal rate is 3.5 [mu] m / h. Compared with an existing aluminum oxide polishing solution, the polishing solution has the advantage that the surface quality and the polishing efficiency of the silicon carbide wafer are remarkably improved.

The invention relates to the technical field of numerical control machining process optimization, in particular to an aluminum alloy milling parameter optimization method based on an improved Bayesian neural network, and the method comprises the following steps: S1, building an aluminum alloy thin-wall part milling finite element model based on a Johnson-Cook constitutive model; s2, carrying out a single-factor milling experiment, and verifying the reliability of the finite element model; s3, constructing a Bayesian neural network model fused with a multi-head attention mechanism, and establishing a mapping relation between process parameters and surface quality indexes; s4, establishing a multi-objective optimization model based on the MHA-BNN prediction model, and solving by adopting an NSGA-II algorithm; and S5, performing experimental verification and analysis on an optimization result. According to the method, a multi-head attention mechanism is introduced into a Bayesian neural network, and a milled surface roughness and residual stress probability prediction model is constructed; and a multi-objective optimization framework fusing physical constraint and data driving is established, collaborative optimization of the surface roughness, the residual stress and the processing efficiency is realized, and meanwhile, the problem that a traditional deep learning method is insufficient in prediction precision under a small sample condition is solved.

The invention discloses a nondestructive testing method for a welding seam of a steel structure. The method comprises the steps that firstly, morphology data of the surface of a building steel structure are obtained through a high-precision laser scanner and an electromagnetic induction technology; coating thickness distribution and weld surface roughness are analyzed based on the morphology data, and a surface information matrix is constructed; then automatically adjusting ultrasonic wave beam parameters of the ultrasonic phased array by using the adaptive parameter regulation and control model and executing detection to obtain a preliminary ultrasonic echo signal; the time-frequency domain difference between the preliminary echo and the reference signal is calculated through an echo signal compensation model, and secondary compensation is completed to obtain an optimized ultrasonic echo signal; fusing the optimized echo signal, the morphology data and the surface information matrix to construct a detection data set, and adopting a pre-trained steel structure detection model to realize defect detection; and finally, evaluating the confidence of a detection result by adopting a three-dimensional confidence evaluation model, performing grading, and performing local re-detection and parameter adjustment on a low-confidence region to form closed-loop optimization. According to the method, the surface state of the steel structure can be adaptively matched, the accuracy of echo signals and the defect detection precision are improved, and the detection reliability is guaranteed through three-dimensional confidence evaluation.

The invention discloses a flattened bamboo differentiation online spraying system and method based on multi-modal information. The system comprises a conveying module, an image collecting module, a surface roughness detecting module, a data fusion processing module and a spraying module. Wherein the image acquisition module consists of a color camera and a depth camera, and is used for identifying flattened bamboo wood types (bamboo green or tabasheer) and structural characteristics (bamboo joints or internodes); the surface roughness module collects Ra value information by using a contact probe; and the data fusion processing module performs conjoint analysis on the multi-modal data, and generates spraying control information containing path instructions and coating parameters according to the bamboo structure and the surface state. For an area with a complex structure or high adhesion risk, traditional multi-channel spraying is replaced by increasing the primer coating amount at a time, and accurate and efficient differential coating control is achieved. The method is accurate in recognition, efficient in spraying, accurate in control and suitable for intelligent bamboo furniture and decorative plate surface treatment scenes.

The invention relates to the technical field of ultrasonic detection, in particular to a building steel structure strength ultrasonic phased array nondestructive testing method which comprises the following steps: acquiring morphology data of the surface of a building steel structure; analyzing coating thickness distribution and weld surface roughness of the surface of the building steel structure, and constructing a surface information matrix; automatically adjusting ultrasonic wave beam parameters of the ultrasonic phased array by using the self-adaptive parameter regulation and control model, and performing ultrasonic phased array detection to obtain a preliminary ultrasonic echo signal; adopting an echo signal compensation model to perform secondary compensation on the preliminary ultrasonic echo signal through the difference between the reference signal and the preliminary ultrasonic echo signal to obtain an optimized ultrasonic echo signal; fusing the optimized ultrasonic echo signal, the morphology data and the surface information matrix, and performing defect detection on the building steel structure by using the building steel structure detection model; evaluating a detection result, and dividing a confidence level; and carrying out local re-detection and parameter adjustment on the low-confidence region to realize closed-loop optimization.

The invention provides a construction method and system of a rock-soil shear strength parameter prediction model, and relates to the technical field of model construction.The construction method comprises the steps that CT scanning data of a real rock-soil sample is obtained and preprocessed, and an irregular three-dimensional particle set for geometric modeling is obtained; performing particle surface roughness modeling analysis on the three-dimensional particle body set to obtain a surface roughness coefficient set of the three-dimensional particle body set; performing particle size cumulative distribution function segmentation processing and fractal dimension control on the surface roughness coefficient set to obtain a structure grading sample set; constructing a THMC multi-field coupling model based on the structure grading sample set to obtain a working condition particulate matter physical response parameter set; and performing a simulation experiment on the working condition particle physical response parameter set, and performing deep neural network modeling processing based on an experiment result to obtain a strength response model capable of predicting the cohesive force and the internal friction angle of the rock-soil body under any working condition input. According to the invention, the prediction efficiency and accuracy are significantly improved.

The invention provides a concrete strength prediction method based on deep learning, and the method comprises the steps: obtaining a coarse aggregate surface image through employing a three-dimensional scanning method, recognizing the surface roughness of the coarse aggregate, and extracting the pore distribution characteristics of the surface of the coarse aggregate through combining with an image segmentation algorithm; carrying out feature extraction on the interface transition area by adopting wavelet transform to obtain local features of the interface transition area, and carrying out weighted calculation on the local features in combination with an adaptive weighted fusion algorithm to obtain enhanced local feature representation; fusing the marked microstructure parameters with the macroparameters of the whole concrete, constructing a concrete strength prediction model, and outputting a predicted value of the whole strength of the concrete; and predicting the weights of the regions corresponding to different intensities through an adaptive weighted fusion algorithm to obtain local features contributed to the target intensity.

The invention discloses a printed circuit board welding spot three-dimensional shape high-precision analysis method and system, and the method comprises the steps: constructing a high-precision three-dimensional model, and obtaining three-dimensional point cloud data; acquiring feature data of a welding spot region, and judging whether pseudo soldering occurs or not through fusion analysis; the surface roughness and reflectivity loss data of a welding spot area are obtained, the welding spot solidification line density is calculated, the cold welding direction disorder degree is obtained, and the cold welding condition is judged; the minimum distance between adjacent welding spots is obtained, and the bridging condition is detected in combination with conductive path curvature verification and material diffusion characteristics; acquiring the surface texture gradient and reflectivity attenuation data of the welding spot, and performing welding spot oxidation treatment; and performing model training through a convolutional neural network, constructing a welding spot analysis model, and obtaining a welding spot analysis result. The method has the advantages that through high-precision three-dimensional scanning and convolutional neural network deep learning, pseudo soldering, cold welding, bridging and oxidation defects of the welding spots of the printed circuit board are accurately detected and analyzed, and the production efficiency and the product reliability are remarkably improved.

The invention provides an underwater target detection method and system based on deep learning. The method comprises the following steps: acquiring an underwater image sequence; depth information is generated based on the underwater image sequence, and boundary contours of different underwater objects are extracted through a deep learning algorithm in combination with the depth information; constructing a reflection mode of underwater light according to the surface roughness of various underwater objects, and then performing semantic division according to the reflection mode to obtain mask regions corresponding to different semantic levels of each image frame; for each image frame, determining texture direction features, determining main reflection directions of underwater light in different mask areas, and further determining semantic coordination degree of each mask area according to all the texture direction features and all the main reflection directions; and generating confidence semantic tags of different underwater objects in the underwater image sequence based on all semantic coordination degrees. By adopting the scheme provided by the invention, the anti-distortion detection can be performed on the underwater target under the interference of the underwater environment.