1191 results about "Discretization" patented technology

The invention provides a new energy output prediction method and system based on Monte Carlo Dropout, and the method comprises the steps: carrying out the probabilistic prediction of new energy output through a Monte Carlo Dropout technology, generating a dynamic prediction result containing a confidence interval, and quantifying the uncertainty of meteorological sudden change and equipment state; secondly, constructing a multi-stage random dynamic programming model, discretizing a prediction interval into multi-scene input, designing a non-linear objective function based on the discharge depth, and synchronously optimizing the electricity purchase cost and the energy storage aging cost; and finally, realizing rolling optimization of the system in combination with a model prediction control framework, and dynamically adjusting an energy storage aging cost weight by updating prediction data and a scheduling instruction on line and embedding an energy storage health state real-time feedback mechanism. The photovoltaic and wind power consumption rate can be remarkably improved, the full life cycle cost of an energy storage system is reduced, and meanwhile, the robustness of a scheduling strategy in extreme weather is ensured.

The invention relates to the field of thin-wall structure displacement field monitoring, and discloses a coupled functional unilateral inverse finite element cylindrical thin-wall structure displacement field monitoring method, which comprises the following steps of: performing discretization by adopting a four-node inverse shell unit, and arranging a sensor at a geometric center position of each unit; based on the basic mathematical model of the inverse finite element, establishing an overall pseudo-stiffness matrix and an overall pseudo-force matrix; and calculating to obtain a displacement field of the structure by solving a linear equation set formed by the overall pseudo-stiffness matrix, the overall pseudo-force matrix and the global displacement vector. According to the method, the structural displacement field is calculated by using the strain information measured at one side, and the global deformation is accurately and reversely deduced under the condition that only the local strain data is obtained by combining the advantages of finite element analysis and inverse problem solution. By optimizing local strain data solution, high cost and low efficiency of global strain measurement in a traditional method are avoided, and the method has remarkable practical advantages.

The invention relates to a quick river flood forecasting method based on a physical information neural network, and belongs to the field of river flood forecasting. The method comprises the following steps: on the basis of a traditional physical information neural network (PINN), introducing a boundary condition parameter as an input variable, and enabling the PINN to learn a flood wave propagation rule under different boundary conditions. Furthermore, on this basis, a physical information neural network flood fast forecasting framework (RFF-PINN) integrated with a hydrodynamic method is provided, a numerical solution based on grid discretization is reconstructed into a continuous function in a time-space domain through a piecewise polynomial interpolation method, residual error loss between network output and a hydrodynamic model simulation value is constructed, and therefore, a flood fast forecasting result is obtained. The network parameters are optimized in cooperation with the PDE loss, and the problem that the network parameter optimization effect is reduced due to the fact that the PDE loss of the complex flow state area is difficult to converge is solved. The method has the beneficial effect that the water depth change process of each section of the river channel under any boundary condition can be accurately and quickly predicted.

The invention relates to the technical field of clean environment control, in particular to a dynamic pipe network and model predictive control clean room energy-saving pressure control method and system. Obtaining current pipe network impedance according to the pipe network impedance curve; calculating the current valve impedance according to the valve opening; calculating the total pressure drop of the current pipe network; constructing a fan dynamic model based on a fan similarity law to calculate the air volume at the next moment; a clean room pressure difference dynamic model is constructed, and the clean room pressure difference at the next moment is calculated; constructing a state space model and a target function of a model prediction controller; and when the pressure difference of the clean room is unstable, predicting the change trend of the pressure difference of the clean room in the next time period, and reversely solving the optimal fan frequency and valve opening of the discretization state space model by using the model prediction controller so as to achieve the lowest pipe network impedance and obtain the optimal fan frequency. The control precision and robustness of the pressure difference of the clean room are remarkably improved, and meanwhile energy consumption is reduced.

The invention discloses a hybrid Lp regularization magnetotelluric two-dimensional inversion method based on adaptive weight, and the method comprises the steps: reading magnetotelluric observation data, carrying out the quadrilateral finite element grid discretization of a whole inversion region, building an inversion grid, and building an initial model and a prior model of the inversion conductivity of the inversion region; based on the initial model and the prior model of the inversion conductivity and the magnetotelluric observation data, constructing a mixed Lp regularization inversion objective function comprising an L1 regularization item, an L2 regularization item and a data fitting item; in an inversion iteration process, adaptively adjusting weight factors of the L2 regularization item and the L1 regularization item; and based on the inversion grid, performing magnetotelluric two-dimensional inversion according to the inversion objective function mixed with Lp regularization, the initial model and magnetotelluric observation data, and outputting optimal model parameters of Gaussian Newton inversion to obtain underground electrical structure information. According to the invention, the resolution and accuracy of the inversion result are improved.

The invention relates to the technical field of polishing, and discloses an intelligent steel rail polishing control method which comprises the following steps: acquiring surface and internal defect data of a steel rail through laser and ultrasonic fusion detection, discretizing the data into multi-dimensional nodes, and constructing a dynamic graph model comprising a microscopic-macroscopic dual-cluster neural network to predict a defect evolution trend; based on fractal dimensions and information entropy, multi-source data are dynamically fused, abnormities are eliminated, chaotic optimization and wear prediction are adopted to generate global optimal polishing parameters, and finally parameter dynamic adjustment is achieved through closed-loop feedback of pulse frequency and control errors. Steel rail defect collaborative perception is achieved through a multi-scale dynamic graph model, fractal-entropy weight dynamic fusion is combined to enhance data reliability, global optimal parameters are generated through chaos optimization and abrasion prediction, accurate grinding is achieved through pulse-error closed-loop control, and the method is simple and convenient to implement. The problems of data fusion failure, defect response lag, parameter local optimization and insufficient control precision in the prior art are solved.

The invention relates to a quantum gate simulation and control parameter optimization method based on numerical calculation, and the method sequentially comprises the following three modules: a model Hamiltonian matrix representation module can calculate the matrix representation of a model Hamiltonian according to a specific problem; a time evolution and fidelity calculation module of the system discretely solves a Schrodinger equation by using a Hamiltonian matrix, gives a corresponding time evolution matrix, and calculates gate operation fidelity by using the obtained time evolution matrix; and the control parameter optimization module calls the first two modules, optimizes control parameters and quantum gate operation time by adopting a local optimization algorithm or a global optimization algorithm, and gives optimal control parameters and gate operation time. The problem that the gate fidelity of a current superconducting quantum computing system cannot meet the requirement for reliable operation of a practical quantum algorithm is solved, the detailed process of quantum gate operation can be simulated through a numerical calculation method, and the quantum gate operation fidelity is improved by optimizing control parameters.

The invention relates to the field of image processing, and discloses a lightweight container correction method and system based on machine vision, and the method comprises the steps: building a detection model through enhancing the feature response of a corner fitting region, detecting a corner fitting, calculating a disparity map of a multispectral image, reconstructing the three-dimensional coordinates of the corner fitting, and carrying out the pose conversion; identifying time-varying parameters on line to construct a state transition matrix, calculating predicted vibration displacement, projecting the vibration displacement to a world coordinate system, and calculating innovative sequence variance to dynamically adjust noise parameters; an inner and outer ring control framework is constructed, and an outer ring visual controller generates target force by calculating pose errors in a world coordinate system; an inner loop force control compensator combines predicted vibration offset and force sensor feedback to correct a mechanical arm control instruction, control gain is scheduled through a task stage, and vibration frequency self-adaptive adjustment parameters are analyzed; and performing force control compensation based on the target force, and discretizing the continuous impedance model to obtain a recursive control law to resist external disturbance.

The invention discloses an integrated simulation method for flood routing, and the method is characterized in that the method comprises the following steps: delineating the plane contour of a ground feature in a research region, and carrying out the recognition and classification of the ground feature; fine modeling of ground objects in the research area is carried out; performing space-time numerical discretization on a two-dimensional control equation describing water flow movement on a computational grid, and constructing a two-dimensional hydrodynamic solver; establishing a ground feature calculation module and a connection calculation module of the river channel and the urban / flood storage and detention area along the river; establishing a space-time connectivity control module of the river channel and the urban / detention and detention area along the river, and carrying out simulation of flood routing and regulation and control of the river channel and the urban / detention and detention area along the river step by step; results such as a water flow plane physical field, a section water level and a flow process calculated by the flood rehearsal cooperative work system are stored in a hard disk, and flood regulation and control analysis and evaluation are carried out. The method has the advantages of being good in stability, high in reliability, high in calculation efficiency, high in universality and the like.

The invention discloses an engine additive component service life prediction method based on modal polycondensation and expansion, and the method specifically comprises the steps: firstly, achieving the grid discretization of an initial defect structure based on a finite element method, and carrying out the local refinement of a defect region; performing modal polycondensation on the overall model to obtain a low-dimensional system model, and performing rapid calculation under the low-dimensional model to obtain modal response; establishing a conversion relation between a displacement vibration mode and a stress vibration mode, and obtaining full-field stress by utilizing modal response and based on modal extension; and solving a stress intensity factor-time history of the crack tip by adopting a stress extrapolation method, obtaining a delta K-R-n three-variable rain flow matrix by utilizing a rain flow counting method, and finally carrying out residual life evaluation through an iLAPS model. According to the method, system dynamics, fracture mechanics and modal order reduction technologies are fused, a new efficient and accurate fatigue life prediction path adapting to complex working conditions is provided for additive manufacturing components with defects, and the method has remarkable engineering application value and popularization potential.

A long-distance pipe-jacking tunnel construction method based on three-dimensional stability analysis comprises the following steps: constructing a three-dimensional numerical model based on geological exploration data; performing high-precision discretization processing to form an infinitesimal triangular mesh; assigning physical and mechanical parameters to the soil body in the model; deducing a pipe-jacking tunnel excavation face supporting force calculation formula by applying a limit analysis upper limit theorem; developing an integrated calculation program, and performing comparative analysis with measured data; the accuracy of a calculation program is verified through actual measurement data, the jacking force is dynamically optimized by integrating construction parameters, and the influence of additional stress of the upper structure on construction stability is evaluated; the three-dimensional stability analysis model is optimized and updated, and accurate control over the construction process is achieved; and according to the optimization model and a data analysis result, a construction scheme is adjusted in real time to adapt to stratum change and construction deviation. Compared with a traditional method, the stability analysis precision of pipe-jacking tunnel construction is improved, the construction scheme is optimized, and the method has remarkable engineering application value and popularization prospects.

The invention relates to the field of power system frequency stability control, and discloses a virtual synchronous machine energy storage frequency modulation method and system based on model prediction and adaptive control, and the method comprises the steps: constructing a prediction model of a virtual synchronous machine, wherein a virtual inertia coefficient and a virtual damping coefficient are included; designing a cost function, wherein the cost function is a weighted sum of squares of the frequency increment and the input power increment; solving the optimal control sequence by using quadratic programming, and further correcting the input power of the VSG in real time; and in combination with the dynamic characteristics of the energy storage device, the virtual inertia coefficient and the damping coefficient are adaptively adjusted according to the frequency deviation and the frequency change rate. The method focuses on the combination of model prediction control and adaptive control, is applied to the energy storage frequency modulation of the virtual synchronous machine, and aims to effectively solve the problem of frequency stability caused by the access of high-proportion renewable energy to a power system.

The invention relates to the technical field of structural design, and discloses a topology and size joint optimization design method and system for a hierarchical composite structure, and the method comprises the following steps: dividing a structural domain into a plurality of hierarchical sub-domains, defining the size constraint range of each sub-domain, and distributing a multi-phase material for each sub-domain; grid discretization processing is carried out on the structural domain, a topological design variable is given to each discrete unit, acoustic unit modeling is carried out on the sound field domain, sound pressure boundary conditions are defined, and coupling boundary conditions are applied to the interface of the sound field domain and the structural domain; solving an acoustic-structure coupling equation based on a finite element method, calculating an objective function and constraint conditions, analyzing the sensitivity of the objective function to topological variables by adopting an adjoint method, and iteratively updating the topological design variables through an optimization algorithm until a convergence criterion is met; and on the basis of a topological optimization result, constructing a neural network agent model of an input layer, a hidden layer and an output layer, and carrying out global optimization on sub-domain size parameters by adopting a genetic algorithm so as to minimize a target function.

The invention discloses a servo system speed loop parameter self-tuning method based on digital-analog linkage. The method comprises the following steps: establishing an open-loop transfer function of a speed loop of a servo system; parameters of a speed ring are initialized, a same-frequency and same-amplitude speed instruction is sent to a motor to drive a load to operate, and the total rotation inertia of the motor is calculated; a mapping relation equation is determined, and parameters of a speed PI controller are obtained; discretizing the parameters of the speed PI controller by adopting a bilinear method to obtain a discrete proportional gain and a discrete integral gain; determining constraint conditions of the servo system and calculating an evaluation function value according to the ITAE performance evaluation function; obtaining optimal control parameters by adopting a coordinate rotation method; corresponding discrete proportional gain and discrete integral gain are calculated according to the optimal control parameters to serve as parameters of a speed PI controller to complete parameter setting, and load operation is achieved according to a speed instruction. According to the method, the setting process is more efficient, the calculation parameters are more accurate, the manual debugging time can be greatly shortened, and the universality is good.

The invention discloses a simulation method and system based on multi-field coupling and dynamic grid processing, and belongs to the field of computational simulation, and the method comprises the steps: building and solving a coupling physical field model: carrying out finite element spatial discretization on a solving domain, building and solving a coupling partial differential equation set which at least comprises a temperature field, a displacement field and a stress field, and carrying out the finite element spatial discretization; obtaining numerical solutions of the physical fields in the current state; carrying out progressive unit removal based on a physical criterion; and finally, grid data structure updating and solver synchronization are executed. According to the method, the metal material can be gradually removed when the physical quantity reaches the critical condition, and the grid change is dynamically and robustly processed in the process.

The invention discloses a rapid calculation method based on a non-coordination and reanalysis theory and a computer device. The invention relates to the technical field of structural design optimization, and solves the problem that it is difficult to realize efficient structural response analysis while ensuring the calculation precision in the prior art. The method comprises the following steps: selecting a structural member of a to-be-calculated structural response and discretizing the structural member; preprocessing the data, and determining a coarse and fine grid index set corresponding to interface fine grid nodes; solving a unit shape function, and embedding the shape function into a body stiffness matrix according to a degree-of-freedom position to serve as an interface constraint condition; calculating a variable element stiffness matrix, and completing the construction of a modified model stiffness matrix; carrying out iterative calculation to obtain a basis vector; and calculating and solving the structural response of the modified model. According to the method, the efficient response analysis of the non-coordinated variable structure is realized on the premise of ensuring the calculation precision.

The invention relates to a thermocouple and Kalman filtering-based global mk-level precision temperature control method. The method comprises the following steps of S1, arranging a thermocouple temperature measurement network; s2, establishing a mathematical model of a temperature measurement and control system; s3, establishing a discrete temperature measurement and control system model containing process noise and a thermocouple observation model containing measurement noise; s4, acquiring a system temperature priori estimation value at the current moment, and calculating a priori state estimation error covariance matrix; s5, obtaining a system temperature posterior state estimation value, updating the posterior state error covariance matrix, and obtaining a temperature measurement and control system correction model; s6, obtaining an error vector of a temperature measurement and control system; s7, obtaining a full-domain control variable based on feedforward-feedback, and acting the full-domain control variable on the temperature control object through an execution component; and S8, repeating the steps S4-S7 to form closed-loop control so as to realize the global mk-level temperature control precision of the temperature measurement and control system. The method has the beneficial effects of good comprehensiveness and strong coordination.

The invention relates to the technical field of geophysical exploration, and particularly provides a three-dimensional finite element transient electromagnetic forward modeling calculation method based on an iterative solution method.The method comprises the steps that tetrahedron space discretization is carried out based on a vector finite element method, and an electric field is given to six edges of a space tetrahedron; constructing a back-pushing Euler finite element equation set, and assembling a linear system formed by a large sparse matrix based on the equation set; performing iterative solution on the linear system based on an FGMRES algorithm to obtain an electric field value corresponding to a spatial tetrahedron unit edge; optimizing a spatial tetrahedral mesh based on a quadratic nesting method, and optimizing a time step length based on a third-order backward Euler difference format; a grid pre-encryption method is adopted, the calculation error of iterative solution is reduced, a region decomposition method is used for achieving parallel calculation, and the calculation efficiency is greatly improved. According to the method, the requirements of high precision and high efficiency in three-dimensional transient electromagnetic forward modeling calculation can be met, and an advanced calculation tool is provided for resource exploration in a complex geological environment.

The invention provides a variable-speed variable-pitch wind generating set torque control method and system and a computer readable medium, and the method comprises the steps: designing a rolling optimization torque and variable pitch instruction based on model predictive control (MPC) and sliding mode surface constraint on the basis of constructing a discretization model comprising a pneumatic mechanism, a transmission chain and a variable pitch mechanism; the robustness is enhanced in combination with the self-adaptive sliding mode surface design, the control gain is dynamically adjusted through fuzzy logic, and the power tracking precision and the mechanical loss suppression are further considered; meanwhile, the delay is reduced with the assistance of wind speed prediction feed-forward compensation, the method has high robustness and adaptive capacity, wind speed fluctuation and parameter time variation can be effectively handled, equipment is protected while the power generation efficiency is guaranteed, and an efficient and reliable solution is provided for wind power generation control.

The invention provides a lead screw precision milling closed-loop optimization and real-time control process based on multi-axis linkage, which comprises the following steps of: acquiring geometric parameters of a workpiece, material attributes and kinetic parameters of a machine tool, designing a cutter and clamp scheme, and setting initial processing parameters based on a cutting database; constructing a lead screw spiral groove high-precision curved surface model by using an NURBS curve, and optimizing a path through curvature adaptive discretization and quintic spline interpolation; an idle stroke path is iteratively optimized by adopting a real number coding genetic algorithm, and the total time of the path and the reversing frequency are shortened; multi-axis synchronous control is achieved through inverse kinematics solution, and the cutter inclination angle is dynamically adjusted in real time based on the contact angle; and virtual simulation and finite element analysis are utilized to predict cutting force, thermal deformation and interference risks. According to the method, the continuous and smooth multi-axis synchronization instruction is generated, the positioning error of each axis is effectively reduced, meanwhile, the cutter inclination angle is dynamically corrected through force feedback of the contact angle, cutting force fluctuation is effectively restrained, and it is ensured that the tooth surface roughness of the spiral groove meets the requirement.

The invention discloses a simulation method for evaluating damage of a concrete gravity dam under the action of an explosion load. The simulation method comprises the following steps: establishing geometric models of the concrete gravity dam, water and explosives, performing grid division on the geometric models, and dispersing units into particles to obtain a discretization model; giving material parameters, boundary conditions and contact to the discretization model, and setting a detonation point position and a detonation mode; coupling effects between the explosive and water, between the water and the concrete gravity dam and between the explosive and the concrete gravity dam are set; key information is initialized; a fluid-structure interaction force is calculated by adopting an asynchronous solution IHPD-SPH coupling method; an SPH solver is adopted to update the speed and density information of the fluid; an IHPD solver is adopted to update internal force and acceleration information of the solid; checking whether the calculation duration is met or not, and if yes, ending calculation; if not, recalculating the fluid-solid coupling force according to the currently obtained particle state; and obtaining a dam damage result, and performing damage grade evaluation.

The invention provides a key flow routing optimization method based on deep reinforcement learning, and belongs to the field of flow engineering. Under the constraint of routing overhead, a part of the total flow is selected as a key flow and rerouting is carried out, so that the purposes of minimizing the maximum link utilization rate and reducing the time delay are achieved. And splicing the traffic matrix, the topological matrix and the feature matrix as input, and carrying out feature extraction with higher dimension through a graph neural network containing a K-hop peripheral sub-graph. For the discrete action space domain and maximum entropy problem, the optimal selection of the key flow is obtained by using a discrete soft behavior person-reviewer, and finally, the optimal routing scheme is obtained by solving through linear programming constraint. Simulation results show that compared with a traditional algorithm and other reinforcement learning algorithms, the provided algorithm has better time delay performance and ability to adapt to dynamic network topology.

The invention provides a bridge pier nonlinear earthquake random response analysis method based on a BWBN model, and belongs to the field of civil engineering structure random vibration analys.The method comprises the steps that discretization processing is conducted on a concrete bridge pier structure, and a finite element nonlinear analysis numerical model is constructed; constructing a BWBN model by combining linear and nonlinear differential equations, setting a hysteresis parameter value of the BWBN model, simulating by adopting an Euler-Bernoulli beam unit, and establishing a degraded beam finite element model; obtaining a unit hysteresis vector, constructing a system motion balance equation, solving an equivalent linearization coefficient, constructing a covariance matrix of random response of the concrete pier structure, performing random response analysis on the concrete pier structure, and exploring typical working conditions of random excitation. According to the method, the operation time is shortened, the calculation precision is improved, and meanwhile, the response characteristics of the reinforced concrete pier structure under nonlinear earthquake random excitation can be truly represented.

The invention relates to the technical field of image precision optimization, in particular to a surveying and mapping image mathematical precision evaluation method based on dense point cloud matching, which comprises the following steps: acquiring and preprocessing image data, synthesizing an image gradient modulus length and a maximum principal curvature, and introducing a surface normal vector partial derivative to obtain a surface normal vector partial derivative; calculating a curvature-driven depth discretization step length, and generating an initial point cloud data set; constructing a point cloud complex network structure diagram and calculating the topology durability of a point cloud topology structure to obtain an optimized point cloud data set; based on the optimized point cloud data set, in combination with the target image information, the adaptive curvature gradient weight and the error diffusion control value, constructing an adaptive matching cost function, and calculating a point cloud matching cost matrix; and calculating an optimal point cloud transformation matrix based on the point cloud matching cost matrix. According to the surveying and mapping image mathematical precision evaluation method based on dense point cloud matching, low-texture region point cloud matching error suppression and structure optimization of high-curvature region point cloud on a complex curved surface are realized.

The invention provides a grid quality optimization method, computer equipment, a program product and a storage medium. The method comprises the steps that an initial grid is acquired, the initial grid is converted into a built-in data structure, and a grid object is acquired, wherein the grid object comprises a node set, a tetrahedron set and topological adjacency information; traversing each grid node of the grid object, and adding a constraint condition for each grid node; regarding the discretized three-dimensional grid as a hyperelastic body, and calculating the energy of the current grid for each tetrahedron unit in the grid object through a hyperelastic model; performing iterative calculation on the energy minimum value of the current grid according to the constraint condition, and applying random disturbance after each N iterations until a convergence threshold value or the maximum number of iterations is reached; and the grid corresponding to the minimum energy value is output as the optimal grid. According to the scheme, the overall quality of the grid can be effectively improved, and the boundary retention capability and the restoration capability of the degraded unit are enhanced, so that the simulation quality is improved.

The invention discloses a numerical simulation method for a semiconductor device based on spatial discretization and uncoupling iteration. The method comprises the following steps: establishing a basic equation set of the semiconductor device; spatial discretization is carried out on the non-uniform hexahedral mesh; discretizing the current equation to obtain a drift diffusion model equation set in a three-dimensional discrete format; and sequentially solving each equation in the drift diffusion model equation set by adopting a Gummel uncoupling method. According to the semiconductor device numerical simulation method based on spatial discretization and uncoupled iteration provided by the invention, a semiconductor partial differential equation is subjected to three-dimensional spatial discretization by adopting an FDM method, and a large-scale nonlinear equation set is solved by adopting a Gummel uncoupled iteration method, so that physical characteristics and internal physical images of the device are obtained; the occupied computer memory is relatively low, the method is not sensitive to an initial value, and the simulation speed is greatly improved.

The invention discloses a double-capacity water tank control method based on multi-target reward Q learning, and belongs to the technical field of machine learning and automatic control. The method comprises the following steps: firstly, establishing a discretization mathematical model of a double-capacity water tank liquid level control system, and designing a multi-target reward function; secondly, empirical data are stored only when the liquid level tracking error exceeds the limit or the control quantity suddenly changes, and sampling weights are distributed according to award absolute values; and finally, updating the Q table based on the time difference error, and performing closed-loop real-time control on the double-tank water tank according to the dynamically adjusted PID parameter. According to the method, a five-dimensional reward function is designed, and multi-index dynamic balance is realized through linear weighting; the empirical tuple is stored only when the liquid level tracking error exceeds the limit, and the quick response requirement of the dynamic time-varying scene of the double-capacity water tank is met; dimensionality reduction is performed on a continuous state space by adopting a non-uniform grading strategy, so that the calculation complexity is remarkably reduced; a perception-decision-execution integrated framework is constructed, and the dependence of a traditional control method on a mathematical model of the water tank is abandoned.

The invention provides a bridge concrete strength detection method based on a pigeon inspired algorithm and an inversion theory, which comprises the following steps: establishing a two-dimensional concrete slab model of a railway overpass, and carrying out discretization modeling by adopting a finite element method; on the premise that initial parameters and boundary conditions are known, numerical analysis is conducted on the model, and the temperature field distribution situation of the concrete structure is obtained; carrying out optimization inversion on the temperature field data through a pigeon inspired algorithm, and constructing an objective function in combination with the existing concrete temperature field data, so as to invert the temperature distribution in the concrete; and evaluating the strength of the concrete based on the internal temperature data of the concrete, the cement hydration heat release characteristic and the transmissivity curve. The method has the beneficial effects that nondestructive testing of the internal temperature of the concrete is realized, so that the slump is accurately judged, the strength is accurately evaluated, and the problems that in the prior art, only the strength can be roughly judged, and damage to different degrees of a concrete structure cannot be effectively recognized are solved.

The invention discloses an application method of a deep learning model embedded based on a physical mechanism in urban flood ponding point identification. The method comprises the following steps: collecting and preprocessing basic data of a research area; taking the preprocessed basic data as input, and performing deformation processing, discretization and linearization processing by taking a Saint-Venant equation set as a hydrodynamic equation set; constructing a differentiable physical module based on the discretized deformation hydrodynamic equation set; on the basis of a differentiable physical module, a deep learning model is constructed, and the deep learning model comprises a backbone network, the differentiable physical module, an attention guidance fusion module, an iteration correction mechanism and a mixed loss function; pre-training the deep learning model, combining fine tuning training and constraint strengthening training, and adding adversarial training in the training process to simulate an extreme scene; and urban flood ponding point identification is carried out based on the trained deep learning model. According to the invention, high-precision real-time prediction of urban flood simulation is realized.

The invention relates to the technical field of detection and identification, and discloses a mobile ship dynamic tracking and locking method and system based on target detection and identification, and the method comprises the steps: constructing a manifold embedded network, and mapping the multi-modal features of a ship to a Riemannian manifold space; constructing a continuous evolution modeler of the Shenchang differential equation model, and predicting a continuous evolution trajectory of the features under time, view angle and scale changes; a multi-scale bridging network is constructed, and bidirectional conversion between long and short distance feature representations is realized; constructing a differential geometric attitude encoder, and representing a ship attitude in an SO (3) Lie group space; a feature memory and reconstruction mechanism is realized, and the problem of short-time disappearance target recovery is solved; the components are integrated to construct a unified tracking and locking system. According to the method, the problems of discretization processing limitation, feature representation fragmentation, cross-condition consistency deficiency and the like in the prior art are solved, remarkable technical effects are achieved in the aspects of cross-condition recognition capability, continuous feature evolution capability, short-time disappearance target recovery capability and the like, and the method is suitable for the scenes of Yangtze River shipping monitoring, maritime affair safety supervision and the like.