293 results about "Numerical stability" patented technology

The invention discloses a sparse Bayesian direction of arrival estimation method based on subspace compression and dictionary optimization, and belongs to the field of array signal processing. According to the method, the data dimension is reduced through the subspace compression technology, and the noise immunity is improved; signal power and noise power are automatically estimated in combination with a sparse Bayesian model, and dependence on information source number information is avoided; the calculation efficiency and the numerical stability are improved through a support set adaptive pruning strategy; and finally, a dictionary fine tuning mechanism is introduced, direction continuous optimization is realized on the basis of an original discrete grid, an off-grid error is eliminated, and direction-of-arrival estimation with sub-resolution precision is realized. The method is a novel method combining subspace compression, sparse Bayesian inference, adaptive pruning and angle optimization, can give consideration to estimation precision, calculation efficiency and application robustness, and is especially suitable for direction estimation under the complex actual conditions of low signal-to-noise ratio, few snapshots, unknown signal source number and the like.

The invention discloses a landslide image instance segmentation method based on a dual adaptation mechanism, relates to the technical field of image processing, and ensures that a model can have a better effect and numerical stability on data from different sources through a complete process from multi-source data collection to data normalization processing. The designed model is based on a multi-scale state alignment mechanism, in the forward process of the model, exponential moving average fusion is carried out on feature information in the same level, transmission fusion is carried out on feature information of different levels, feature robustness is enhanced, and error accumulation caused by local deviation is reduced. A meta-context incremental learning mechanism is designed, and input data are dynamically converted into a series of key value vector sequences. In the reasoning process of the model, data distribution different from a training domain is dynamically recognized, a gradient descent process is implicitly executed according to the characteristics of data, and model parameters are finely adjusted, so that the characterization capability is greatly improved, and efficient and robust instance segmentation is realized.

The invention discloses an oil reservoir polymer flooding process oil-water distribution simulation method and device based on a cellular automaton, equipment, a medium and a product, and relates to the field of oil reservoir engineering and seepage mechanics. The method comprises the steps that rectangular grid division is conducted on an oil reservoir space, three-dimensional coordinates of cellular nodes are determined, and a cellular automaton model is constructed; according to the polymer oil displacement scheme parameters, oil reservoir state parameters at the end moment of water displacement are initialized, and boundary conditions of the mole-type cellular automaton model are established; making an evolution rule for each neighborhood cellular node in the Moire cellular automaton model; according to a numerical value stability rule, adjusting parameters of each neighborhood cellular node in the Mohr-type cellular automaton model; according to the method, a relative permeability nonlinear regulation and control rule is formulated for the Mohr-type cellular automaton model to simulate oil-water distribution in the oil reservoir polymer flooding process of the Mohr-type cellular automaton model, the starting pressure gradient is accurately simulated, and the low-permeability oil reservoir evolution process is accurately described.

On the basis of tight coupling of the SINS and the USBL, a dual-hydrophone differential model is provided, constant errors in USBL measurement are eliminated, and the influence of slant range noise on navigation precision is relieved; a measurement equation of a dual-hydrophone differential model is established, so that the precision and reliability of the integrated navigation system are improved; a robust iterative Kalman filtering algorithm based on approximate least-square estimation is introduced, measurement dimensions are comprehensively evaluated, measurement noise covariance is dynamically adjusted, and the numerical stability and robustness of the algorithm are improved. According to the method, whether the measured data at the current moment is an abnormal value or not is judged according to previous data, the robust weight is reasonably distributed to the measured data, the adverse effects of subsequent steps such as state estimation are reduced through a method of reducing the weight of the abnormal value, and normalization operation is carried out on the measured residual error according to the adverse effects, so that the residual error distribution is more regular, and the accuracy is improved. And the updating result of the measurement noise covariance is further optimized, so that the precision of the whole Kalman filtering algorithm is improved.

The invention relates to the technical field of cable joint temperature monitoring, and discloses a cable joint temperature monitoring method and system. A target model is constructed based on equipment data of a cable intermediate joint, regions are divided according to geometric shapes and materials to form a topological structure, and an initial thermal circuit model is constructed in combination with temperature field distribution and a thermal resistance branch; and optimizing the model by using the real-time temperature data, and finally judging the running state of the joint through the thermal circuit model. Complex grid discretization is replaced by an equivalent thermal resistance branch through a thermal circuit model, and the calculation speed is remarkably increased; through topological structure division, the limitation of an analytical method on a regular shape is broken through, and a special-shaped structure can be accurately processed; and the thermal circuit model maps an actual boundary through a thermal resistance parameter, so that an error caused by boundary simplification in an analytical method is avoided. The technical problems that a numerical method and an analytical method adopted by a traditional cable connector need to consider grid division and numerical stability, so that the calculation speed is low, and complex geometrical shapes and boundary conditions cannot be accurately analyzed are solved.

The invention discloses a semiconductor device critical breakdown characteristic simulation method based on a hybrid finite volume method. The method comprises the following steps: firstly, carrying out geometric modeling on a device, and carrying out mesh generation; then, the quality of the grids is judged, an FBM method is adopted for the grids with the good quality, and a CVFEM method is adopted for the grids with the poor quality to discretize a drift-diffusion equation set; and then, solving the discrete drift-diffusion equation set system by adopting a full-coupling Newton iteration method, and realizing the optimization of the boost numerical simulation process by adopting an automatic boost strategy and an initial value extrapolation method. And finally, obtaining physical field information of a power semiconductor device (LDMOS) under different external bias voltages, calculating terminal current through a post-processing method, and obtaining an I-V characteristic curve of the device after the avalanche breakdown effect is considered. The method has good numerical calculation precision, numerical stability and grid adaptability, and the complete benchmarking with the solution precision of an international advanced semiconductor simulation tool Sentaur TCAD is realized.

The invention relates to an overlapped chromatographic peak positioning and splitting method and system based on a numerical stable function. The method comprises the following steps: de-noising chromatographic data and correcting a baseline; calculating a second derivative by using SG filtering, grouping continuous points according to a negative threshold value, and determining an initial position of each peak by using a second-order minimum value; establishing a peak function of bidirectional exponential correction by taking a peak position as an initial value, setting parameters and boundaries such as area, center, width, trailing and the like, constructing a residual error minimization target, and performing robust fitting by adopting a trust region reflection algorithm; and indexes such as suitability, output peak area, center, broadening, separation degree, theoretical plate number and the like are verified by residual errors. According to the method, valley point or prior component information is not needed, the method is insensitive to noise and baseline drift, serious overlapping and trailing / leading edge peaks of multiple components can be effectively processed, convergence is high, the method is insensitive to initial values, parameters can be expanded to different peak shapes, and the method is suitable for qualitative and quantitative analysis of automatic and high-flux chromatographic data.

The invention belongs to the technical field of crossing of artificial intelligence and computational mechanics, and particularly relates to an artificial intelligence-based MPM simulation method and system for a mold bag sand cofferdam, a three-dimensional simulation system of the mold bag sand cofferdam is constructed in combination with an MPM method, and the penetration state of sand and geotechnical cloth is detected in an accelerated manner by utilizing a penetration tracking algorithm and spatial Hash. An artificial intelligence algorithm is introduced, a reward function is designed with minimum penetration depth and penetration times as targets, penetration prediction of MPM particles is driven, the speed and stress of the MPM particles are further dynamically regulated and controlled on the basis of a predicted contact model and the artificial intelligence algorithm, penetration is prevented, numerical stability is ensured, and the prediction accuracy of the MPM particles is improved. The calculation precision is improved by adopting implicit integration and gradient descent methods; and bidirectional coupling of the sand and the geotechnical cloth is realized through a visualization module. The problems that in a traditional method, coupling simulation of sandy soil and a flexible structure is low in efficiency, easy to penetrate, unstable in numerical value and the like are solved, and the accuracy and reliability of simulation analysis of the mold bag sand cofferdam structure under the complex working condition are remarkably improved.

The invention provides a multi-source data fusion correction method for a marine observation station, which belongs to the technical field of marine observation, and comprises the following steps of: establishing a hierarchical interpolation correction model comprising a weighted moving average algorithm and a space collaborative interpolation algorithm, and constructing a marine environment adaptive threshold adjustment mechanism based on a closed-loop feedback system; a Kalman filter is combined with a multi-factor coupling drift prediction model of a long-short-term memory network, a progressive degradation fusion strategy switched according to environmental conditions is designed, and a biological pollution gradual change detection correction model adopting a liquid neural network structure is constructed. A multi-source data quality evaluation system based on data continuity, numerical stability, consistency and historical reliability is established, high-precision self-adaptive fusion correction of the marine environment multi-source sensor data is achieved, and the technical problem that high-precision self-adaptive fusion correction of the marine environment multi-source sensor data is difficult to achieve is solved.

The invention discloses a thermal-fluid-solid medium seismic wave simulation method, device, program and medium, and relates to the technical field of oil and gas resource geophysical exploration, and the method comprises the following steps: constructing a first-order thermal-fluid-solid coupling wave equation in an equation coupling form; constructing a coupling wave equation staggered network finite difference discrete format; analyzing numerical stability conditions; constructing a complete matching layer absorption boundary; and performing forward modeling simulation of seismic waves in the three-layer heat-fluid-solid coupling model according to the model parameters and the seismic source parameters. The method breaks through the assumption that the classical seismic wave propagation theory is completely elastic under the isothermal or adiabatic deformation condition, considers the influence of temperature change on the propagation of seismic waves in fluid and solid media, and improves the description precision of the propagation process of the seismic waves in the heat-fluid-solid media. And the capability of inverting the underground temperature field based on seismic wave data in the later period can be effectively improved.

The invention provides an electromagnetic transient simulation method and system containing a synchronous generator, and is applied to the technical field of electromagnetic transient simulation of a power system. The electromagnetic transient simulation method comprising the synchronous generator comprises the following steps: acquiring a simulation equation of a circuit system to be simulated comprising the synchronous generator and a value of an electromagnetic transient state quantity of the circuit system at an initial moment; based on the value of the electromagnetic transient state quantity at the initial moment, adopting a Runge-Kutta algorithm to simulate the simulation equation to obtain a simulation value of the electromagnetic transient state quantity of the circuit system at the target moment; wherein the simulation equation is a port Hamiltonian form equation constructed based on the structure of the circuit system to be simulated. The port Hamiltonian-form simulation equation can keep numerical stability for a long time, calculation errors are reduced, an accurate and stable numerical solution can be obtained, and the requirements for long-time operation accuracy and stability in electromagnetic transient simulation numerical calculation are met.

The invention discloses a multi-view image Gaussian reconstruction method and system guided by LiDAR point cloud. The method comprises the following steps: step S1, realizing accurate registration of LiDAR point cloud and a multi-view image in the same earth-centered earth-fixed coordinate system through GNSS constrained aerial triangulation, and uniformly carrying out center-of-gravity normalization to ensure space consistency and numerical stability; s2, estimating surface normal and scale parameters based on the local geometric features of the LiDAR point cloud, constructing a Gaussian ellipsoid consistent with the local geometry, and realizing Gaussian ellipsoid initialization of geometric perception; and S3, generating a sub Gaussian ellipsoid according to the Gaussian gradient and the tangent plane direction, and improving the modeling density and geometric continuity of the detail region through local encryption and parameter inheritance. According to the method, the fusion of LiDAR point cloud high-precision geometry and multi-view image texture information is realized, the geometric precision, the structural integrity and the detail expression capability of three-dimensional reconstruction are effectively improved, and the method is particularly suitable for high-precision live-action three-dimensional modeling of unmanned aerial vehicle multi-view images.

The invention relates to a fractional order PID (Proportion Integration Differentiation) control method and system based on an FPGA (Field Programmable Gate Array). The method comprises the following steps of: 1, setting fractional order PID related parameters according to the requirements of a controlled object; 2, calculating a fractional order calculus weight coefficient according to the fractional order PID related parameters, and storing the fractional order calculus weight coefficient in the FPGA; 3, collecting a state feedback signal of the controlled object, and calculating an integral term operator and a differential term operator; step 4, calculating control output and performing amplitude limiting processing, and updating the state of the controlled object by using the control output after the amplitude limiting processing; and 5, repeating the steps 1-4 to realize fractional order PID closed-loop control. By optimizing a discrete convolution algorithm, adopting a parallel pipeline architecture and configurable parameter design and combining fixed-point number operation and security boundary constraint, the numerical stability and hardware security of the system under high-speed operation are ensured, the calculation speed is remarkably increased while the operation precision is ensured, and the method is suitable for large-scale popularization and application. And the method is suitable for control occasions with extremely high real-time requirements.

The invention relates to the field of aero-engine complete machine full-three-dimensional numerical simulation, and discloses an aero-engine complete machine full-three-dimensional simulation initial field presetting method and system, and the method comprises the steps: obtaining initial flow field boundaries of different parts from the zero-dimensional level of a complete machine by solving inlet section parameters and outlet section parameters of each part; then obtaining an initial field of each blade row in a turbine part, which has the most critical influence on the convergence of an engine, through a quasi-three-dimensional layer S2, and realizing initial field decomposition from the part to the part; and finally, by adopting a radial basis function interpolation method, mapping the zero-dimensional data and the S2 quasi-three-dimensional parameters to each grid unit of the complete machine full-three-dimensional model of the aero-engine, and then smoothly splicing the initialized flow fields of all the components at the interface to form a continuous complete machine three-dimensional simulation initial field under the examination working condition. The convergence of a whole machine simulation example, the simulation convergence speed and the numerical stability are greatly improved, the simulation trial and error iteration is reduced, the pretreatment time consumption is reduced, and the simulation speed and efficiency are improved.

The invention discloses a hydropower station reservoir runoff intelligent forecasting method based on a large language model, and belongs to the technical field of hydrological forecasting, and the method comprises the following steps: S1, data collection and preprocessing; s2, constructing a BERT-Hydrology deep learning model for the runoff forecast of the reservoir of the hydropower station; s3, designing a weighted mixed loss function, and introducing an optimizer, learning rate scheduling and an early stop mechanism to improve the convergence speed and numerical stability of model training; and S4, establishing an evaluation system, performing rolling forecasting, and outputting an evaluation index and a significance test result. According to the intelligent hydropower station reservoir runoff forecasting method based on the large language model, the accuracy and the stability of runoff forecasting are remarkably improved by constructing the runoff forecasting model, excavating the hidden rule in the hydrological sequence and establishing the multi-step forecasting mapping, and the forecasting accuracy and the forecasting stability are improved. And technical support is provided for intelligent scheduling of a hydropower station reservoir and efficient management of water resources.

The invention discloses a pulverized coal combustion soot generation numerical calculation method based on an orthogonal moment method, and belongs to the technical field of soot simulation. According to the method, through coupling of an improved split-based extended orthogonal moment method (S-EQMOM) and a small flame model and in combination with a detailed chemical reaction mechanism, the problems that multi-modal distribution of soot cannot be accurately described and the numerical stability is insufficient in a traditional method are solved. The method comprises the specific steps that a gas-particle two-phase control equation is established, and a population balance equation is solved through an S-EQMOM split number density function; discretizing the processing model and coupling a pressure-velocity field; physical and chemical behaviors such as pyrolysis, nucleation and coagulation in the soot generation process are analyzed. According to the method, by decomposing the complex dynamic process, numerical error accumulation is reduced, calculation stability is remarkably improved, the soot volume fraction, particle size distribution and other attributes can be accurately predicted, and theoretical support is provided for design of the pulverized coal burner.

The invention discloses an industrial robot motion pose error calibration method based on constraint solution. The industrial robot motion pose error calibration method comprises the steps that S1, joint angles, tail end coordinates and temperature data of multiple measurement points are collected; s2, correcting the kinematic model by using the thermal elongation and constructing a parameter identification model; s3, constructing a mixed integer nonlinear constraint pool according to mechanical limitation; s4, introducing a semi-definite programming slack variable to convert the non-convex constraint into a linear matrix inequality; s5, combining a primal-dual interior point method and an alternating direction multiplier method to iteratively solve the optimal kinematics parameter correction; s6, introducing a square root information filtering improvement mechanism to correct a state vector square root factor, and calculating an updated nominal kinematics parameter; and S7, based on the condition number of the error Jacobian matrix, judging observability and repeatedly calibrating. According to the method, the problem of precision reduction caused by thermal errors is solved, and the identification precision, the numerical stability and the self-adaptive calibration capability of the kinematics parameters of the robot are improved.

The invention relates to the technical field of solid oxide fuel cells, and discloses a construction method of a solid oxide fuel cell multi-physical field bidirectional strong coupling simulation model and a computer program product. The method comprises the following steps: respectively constructing an electrochemical model, a positive / negative electrode fluid flow model and a heat transfer model of the solid oxide fuel cell based on an established geometric structure model; solving equations of the physical fields in all the models are combined to form a full coupling matrix, and iterative solving is carried out according to a set initial value; and judging whether the constructed model is converged or not according to an iterative solution result. According to the model constructed by the method, the mutual influence among electrochemistry, flow and heat transfer processes is fully considered, and synchronous solving is performed through a unified matrix system, so that real-time interaction and collaborative updating among multiple physical fields are realized, and the convergence efficiency and numerical stability of a nonlinear coupling problem are remarkably improved; through inspection, the deviation between a simulation result and actually measured data is less than 8%, and the method has high engineering practical value.

The invention relates to the technical field of electromagnetic transient simulation, in particular to an electromagnetic transient simulation decoupling method, system and device based on a single-impedance branch and a medium, and the method comprises the steps: dividing a system to be simulated into two subsystems which are connected through the single-impedance branch, and carrying out the equivalent reconstruction to form a reconstructed branch model; establishing a discrete state space equation based on the reconstructed branch model; according to the main step length, the number of sub-steps meeting the stability condition is determined based on a small-step synthesis algorithm, and the main step length is equally divided into differential step lengths; in each main step length interval, performing iterative calculation on the state-space equation by using a differential step length to obtain a branch current and a capacitor voltage at the next moment; and injecting the calculated branch current serving as a known current source into an original network equation, solving in parallel to obtain the port voltage of the subsystem, and completing decoupling calculation and promoting simulation. According to the method, effective decoupling of the universal single-impedance branch can be realized, and the applicability, the numerical stability and the calculation efficiency of parallel simulation are improved.

The invention discloses a three-winding transformer power frequency impedance online monitoring and decoupling identification method and system. The method comprises the steps of collecting and preprocessing multi-dimensional operation data of the three-winding transformer, and constructing a feature set to invert real-time topology, load working conditions and energy flow distribution; establishing a power frequency impedance identification model, and introducing an identification stability factor for decoupling calculation of high, medium and low voltage side impedance parameters; and fusing the time sequence variable quantity and historical data, constructing a health degree evaluation index based on the dynamic impedance change rate, and realizing fault early warning and abnormity positioning. Aiming at the problem of error amplification of a parameter identification equation, concepts such as an identification stability factor and a dynamic impedance change rate are introduced, related factors influencing the numerical stability of the equation are deeply analyzed, a scheme for adjusting to-be-solved parameters according to energy flow is provided, adaptive matching is realized in combination with big data, and an online monitoring method is constructed. And multi-stage early warning of the winding state is realized.

The invention discloses a pump turbine flow-induced vibration multi-scale coupling calculation method, and belongs to the technical field of pump turbine fluid-solid coupling dynamics. According to the method, the gap flow effect is included in a multi-scale unsteady fluid-solid coupling analysis framework of the pump turbine, the gap drainage basin is included in three-dimensional modeling of the pump turbine, the three-dimensional model is subjected to grid division by adopting a multi-scale grid technology, and a small transition region is arranged between a movable guide vane and a rotating wheel, so that the gap flow effect is obtained. The transverse direction is consistent with a main runner grid, and the longitudinal direction corresponds to a gap grid, so that data coordination of large-scale and small-scale regions is realized. A fluid-solid coupling solving method of a bidirectional iteration strategy is adopted, a flow field and a structure field are solved respectively, a system coupling module carries out load transmission and coordinated updating in a residual weighting and node mapping mode, the physical consistency and numerical stability of coupling simulation are effectively improved, and the system performance is improved. The method has remarkable advantages in the aspects of improving simulation precision, enhancing calculation stability, improving design efficiency and the like.

The invention provides an LBM large-deformation fluid-solid interface stabilization method and system based on energy flux regulation, and belongs to the technical field of computational fluid mechanics and solid mechanics coupling numerical simulation. Identifying a fluid-solid boundary area, and constructing a buffer layer area; solving fluid in a buffer layer area, and embedding an energy flux regulation mechanism; applying a load to the structure through fluid, and calculating a structure response; fluid is fed back, and the dynamic boundary is updated; and iterating repeatedly, and stopping and outputting after a set termination condition is reached. Active regulation and control of the fluid-structure interface energy exchange process are achieved, the numerical stability and physical precision of fluid-structure interaction simulation under the large deformation condition are remarkably improved, and the method has the advantages of being simple in structure, high in expandability and low in calculation overhead; the method is suitable for complex coupling scenes such as flexible structural body propulsion, airfoil pitching motion and high Reynolds number structural response analysis, and has wide engineering application value.

The invention belongs to the technical field of electric vehicle charging load prediction, and discloses a method for predicting the load of an electric vehicle charging station on a highway, and the method can accurately sense the sudden change trend of charging demands during holidays and festivals through introducing holiday category codes and a model, and improves the prediction precision during the peak period of holidays and festivals. A multi-granularity frequency-time sequence fusion module is adopted, long-term periodicity and short-time disturbance characteristics are effectively integrated, and the modeling capacity of the model for a non-stationary load sequence is enhanced; the designed dynamic Tanh normalization module has an adaptive learning capability, and can improve the numerical stability and nonlinear expression capability of the model in a severe load fluctuation scene; the parallel TCN-MLP structure realizes collaborative modeling of nonlinear interaction and time dependency relationship between variables, and enhances the multivariable prediction performance. The method is accurate in prediction and good in stability, has expandability and engineering landing potential, and can provide efficient and reliable technical support for intelligent scheduling, power resource configuration and holiday and festival energy management.

The invention relates to the technical field of underwater oil and gas production, and discloses a simulation method and system for the long-distance hydraulic transmission characteristic of an underwater umbilical cable, and the method comprises the steps: decomposing a to-be-simulated underwater umbilical cable model into a plurality of organically related hierarchical structures from top to bottom; on the basis of hierarchical structure division of a to-be-simulated underwater umbilical cable model, an underwater umbilical cable simulation model is constructed through types inheriting, expanding and reusing from bottom to top and an assembly splicing mode, and the underwater umbilical cable simulation model can be used for accurately evaluating the influence of the long-distance hydraulic transmission characteristic of the underwater umbilical cable on the dynamic response characteristic of a control system. The method has the characteristics that the calculation precision is high, the numerical stability is strong, and the model input and output meet the simulation analysis requirements of the underwater control system, and can assist engineers to quickly evaluate the design parameters of the umbilical cable of the underwater control system, realize reasonable model selection of parts, and provide reliable technical support for the digital design of the underwater control system.

The invention discloses a stable numerical simulation method for a total ionization dose (TID) and transient dose rate (TDR) synergistic effect in a semiconductor device. According to the method, a control volume finite element tearing interconnection (CV-FETI) method is utilized, discontinuous boundary conditions are included in common numerical properties, a Newton method is adopted to overcome non-convergence caused by nonlinear characteristics of a drift diffusion model, a stable three-dimensional parallel numerical format is adopted, and a non-continuous boundary condition is obtained. Reliable and accurate simulation of radiation of the semiconductor device is realized. Compared with a Gummel iteration method, the method has good numerical stability and precision under the same voltage step. Through comparison with a commercial software calculation result, the effectiveness of the method is verified. The method provided by the invention has the advantages of good stability, precision and applicability, and can be used for high-efficiency and high-precision numerical simulation of the radiation effect of a semiconductor device, thereby improving the performance, robustness and reliability of an integrated circuit.

The invention relates to the technical field of marine environment numerical simulation, in particular to a near-shore typhoon wave refined simulation method based on terrain adaptive grids and parameter dynamic correction, and the method comprises the steps: obtaining basic data and simulation parameters of a target sea area, constructing a computational domain and an initial grid, and constructing a target sea area; the initial grid is an unstructured grid fused with multiple physical constraint factors; performing optimization processing on the initial grid based on a numerical stability criterion to obtain a final calculation grid meeting a calculation convergence requirement; establishing a bidirectional coupling model, and applying the final computational grid to the bidirectional coupling model; carrying out localization iteration calibration on key physical parameters of the bidirectional coupling model to obtain an optimal parameter set; and carrying out typhoon wave process simulation calculation and constructing a typhoon wave feature database by utilizing the final calculation grid and the optimal parameter set, and realizing high-precision simulation and prediction of the typhoon wave process by constructing an adaptive optimization grid, a bidirectional coupling model and a parameter localization calibration mechanism.

The application discloses a lightweight target detection method for unmanned aerial vehicle (UAV) inspection of insulators, first proposes a rotating rectangular frame labeling method for aerial insulator images, and labels the insulator images; then proposes an improved GhostConv module based on a YOLOv5 model, greatly reduces network parameters of the model through ghost convolution, and enhances numerical stability of the model through a Hardswish function; proposes an improved C3Ghost module, reduces parameter quantity and floating point calculation quantity of the model through lightweight network design; introduces an SE attention mechanism between a feature extraction module and a feature fusion module, and suppresses attention degree of a feature channel that is unimportant to a current task through weight distribution; introduces a CSL to replace regression by classification to avoid loss sudden increase problems caused by boundary problems during rotation angle loss calculation; and proposes a high-precision rotating frame regression loss function R-EIoU Loss to accelerate the training process of the model and improve detection precision of the model.

The application provides a simulation method and system for a hot-rolling roller thermal-mechanical coupling process, and belongs to the technical field of multi-field simulation and analog, and comprises the following steps: a two-dimensional geometric model of the roller is established in a simulation software COMSOL, the two-dimensional geometric model is divided into multiple angle zones in the circumferential direction according to a working condition section; the two-dimensional geometric model is meshed; a variable node is added under a global definition module, an angle variable and a rotation variable are added in the variable node, the rotation of the roller in a hot continuous rolling process is simulated by setting the rotation variable, a physical field module is added, and a roller thermal-mechanical coupling finite element simulation model is obtained; the roller thermal-mechanical coupling finite element simulation model is solved, and the change curves of the temperature field distribution, the stress field distribution and the equivalent plastic strain are obtained. The application effectively simplifies the model construction process, improves the calculation efficiency and numerical stability of long-period simulation, and can be used in engineering scenes such as roller cooling system optimization, composite roller structure design and rapid evaluation of rolling process parameters.

The application discloses a power distribution network prediction auxiliary state estimation method and system based on an improved iteration process, and the method comprises the following steps: establishing a nonlinear discrete time state space model; generating a group of Sigma points through an unscented transformation, and substituting the Sigma points into a state transition equation to calculate a state prediction value and a state prediction error covariance matrix at a current time; substituting the Sigma points into a measurement equation to obtain a measurement prediction value, and calculating a Kalman gain; calculating a process noise covariance matrix at a next time; and correcting the state prediction value and the state prediction error covariance matrix at the current time to obtain an optimal state estimation value and an error covariance matrix thereof. The application avoids the risk of losing semi-positive definiteness of the covariance matrix in the iteration process from the algorithm mechanism, significantly enhances the numerical stability and robustness of the UKF algorithm, and ensures reliable application in high-dimensional complex power distribution network state estimation.

This invention relates to the field of fluid machinery and valve engineering, and discloses a computational fluid dynamics (CFD) steady-state simulation method based on the adjustable opening of a V-type ball valve. This method establishes a three-dimensional geometric model including the valve body, valve sleeve, and V-type ball valve core, defining the cone angle of the V-type ball valve core as a function of the valve opening to achieve a continuously adjustable description of the valve opening. By meshing the internal fluid computational domain of the valve and locally refining the mesh at the valve core leading edge, the valve core-seat gap, and the boundary layer region, and combining the set initial and boundary conditions, the continuity equation and momentum conservation equation are solved using computational fluid dynamics methods to obtain the steady-state flow characteristics of the V-type ball valve at different openings. This invention describes the valve opening change in a functional form, improving the numerical stability of steady-state simulation under small opening conditions, while also improving the accuracy of flow characteristic calculations under large opening conditions, making it suitable for multi-condition parametric analysis of V-type ball valves.