17results about How to "Simple model" patented technology

The invention provides a DETR-based 3D target detection and tracking method. The method comprises the steps of obtaining a point cloud data stream of continuous frames from an LIDER camera; original point cloud data of continuous frames are converted into regular voxel grids, and high-dimensional video features of the regular voxel grids are extracted; a group of 4D Tubelet Query is learned, and a time self-attention module and a space cross attention module are utilized to perform modeling on dynamic space-time characteristics of video clips so as to enhance the characterization capability of the dynamic space-time characteristics; and finally, performing bounding box prediction on each Tubelet Query by the model, and measuring the difference between a true value and a predicted value by using set-to-set loss to realize end-to-end learning of a multi-target tracking task. According to the method, the Tubelet is designed, the information in the space-time dimension is deeply mined and integrated, and a multi-target tracking task is simplified into a detection task, so that the unification of the detection task and the tracking task is realized, and the serious dependence of traditional multi-target tracking on a frame-level detector is reduced.

The invention relates to a ship main engine oil consumption modeling method and system based on noon report data, and the modeling method comprises a noon report data collection and screening step, a noon report data classification step, a ship type classification step, a partial load condition fitting modeling step and a unified fitting and draught correction modeling step. Modeling operand and modeling complexity are greatly reduced, meanwhile, multi-level classification of weather (wind level and wave level) and ship types under different dimensions is carried out, a plurality of ship main engine oil models are constructed by adopting a partial load condition fitting method, a unified fitting method and a draught correction method respectively, ship slip rate is adopted as an intermediate variable, and a ship main engine oil model is constructed. According to the method, the relation between the navigational speed and the ship slip rate and the relation between the oil consumption and the ship slip rate are fitted respectively, the fitting flexibility and the formula complexity are improved, the fitting effect is better, the multiple built ship main engine oil models are accurate and efficient, cost reduction, efficiency improvement and safe and reliable performance improvement of enterprises are facilitated, and environmental protection and compliance management can be promoted.

The invention relates to the technical field of over-temperature protection of a vehicle-mounted motor controller of a new energy automobile, in particular to an IGBT junction temperature estimation method, which comprises the following steps of: establishing an experimental platform, and placing the experimental platform in a constant-temperature box; and obtaining data such as IGBT equivalent internal resistance and a temperature coupling table according to an experiment. Calculating the total loss of the IGBT; and building an equivalent thermal resistance model, and calculating the IGBT theoretical junction temperature of the equivalent thermal resistance model according to the thermal resistance value and the total loss of the IGBT. According to the method, the current IGBT junction temperature can be accurately and effectively estimated in the operation process of the controller, the method has the advantages that the internal structure does not need to be damaged, the estimated junction temperature precision is high, modeling is simple, and other sensors or temperature acquisition circuits do not need to be added, engineers can perform protection in advance before the IGBT fails due to the too high junction temperature, the IGBT is effectively protected, and the reliability of the controller is improved. And the service life of the IGBT is greatly prolonged.

This invention provides a spacecraft escort control method based on zero-sum differential games, belonging to the field of aerospace technology. The control method includes the following steps: obtaining spacecraft escort problem parameters for both the pursuing and defending satellites; establishing a spacecraft escort model based on zero-sum differential games and inputting initial parameters into the model; transforming the spacecraft escort model into a two-point boundary value problem model; using a genetic algorithm to solve the two-point boundary value problem model to obtain the state trajectory within the game time, thus completing the spacecraft escort control. This spacecraft escort control method can effectively solve the spacecraft escort problem under continuous thrust, and is an effective extension of existing continuous thrust escort problem models and methods. The model employs a differential game method described by a zero-sum cost function, thus simplifying the solution of the difficult-to-solve weighted performance index escort problem and improving the solution efficiency of the spacecraft escort game problem.

The invention discloses a modal shape and frequency identification method for a motor stator system with an inner gear ring, and the method comprises the steps: building a shell unit and beam unit mixed analysis model of the motor stator system provided with the inner gear ring, and obtaining a total stiffness matrix and a total mass matrix under a local rectangular coordinate system through a matrix displacement method, obtaining a structural stiffness matrix and a structural mass matrix under the overall cylindrical coordinate system by using coordinate transformation; and establishing a discrete free vibration model, inputting the structural stiffness matrix and the structural mass matrix into the model, and outputting modal shapes of different orders and corresponding inherent frequencies of the motor stator system, thereby completing identification. The method comprehensively considers the influence of the stator teeth and the tooth part of the inner gear ring, helps to rapidly identify the modal shape and inherent frequency of the stator system, provides guidance for subsequent vibration acquisition and optimization, and has the advantages of simple modeling, rapid identification and high precision.

The present invention discloses an electroencephalogram (EEG) signal classification method, device and storage medium, wherein the method comprises acquiring an EEG signal; extracting first-level features of the EEG signal; calculating first-level important weights for the first-level features; selecting second-level features according to the first-level important weights; extracting first-level principal components of the second-level features; calculating second-level important weights for the first-level principal components; selecting second-level principal components according to the second-level important weights; confirming an optimal combination according to the second-level features and the second-level principal components; and classifying the optimal combination. The method has the advantages of low computational complexity, simple model and high efficiency. In addition, considering the correlation and globality between features, the performance of feature filtering is improved, thereby improving the classification accuracy of EEG signal classification.

The present invention relates to a flexible satellite linear robust control method based on a dynamic observer, comprising: S1, constructing a flexible satellite attitude system with partially immeasurable states, external disturbances and parameter perturbations, and constructing a linear state space model containing convex polyhedron uncertainty and input constraints; S2, designing a linear dimension reduction dynamic observer and a linear state feedback robust controller, and obtaining a closed-loop system based on the linear dimension reduction dynamic observer and the linear state feedback robust controller and with partially immeasurable states, convex polyhedron uncertainty and external disturbances; S3, deriving the solvability conditions of the linear robust control problem of the closed-loop system based on the linear dimension reduction dynamic observer under the constraint condition of input constraints; S4, solving the solvability conditions, obtaining solutions of the linear dimension reduction dynamic observer and the state feedback robust controller for the flexible satellite attitude system, and controlling the flexible satellite through the solutions.

The application discloses an INS / GNSS rod arm error online estimation method based on dimension reduction filtering. The method splits the INS / GNSS rod arm error and other error variables, constructs multiple low-dimensional filters, reduces the coupling between the error variables, has clear mechanism, simplifies the model, reduces the requirements of the combined navigation algorithm on the chip main frequency and the storage space, improves the calculation speed and the filtering estimation precision, and can improve the positioning performance of the vehicle-mounted combined navigation system after the INS / GNSS rod arm error online estimation and compensation, and has high engineering application value.

This invention relates to the field of aero-engine technology and discloses a Simpack-based method for analyzing the misalignment vibration of aero-engine blades with bosses. While ensuring analytical accuracy, it improves computational efficiency by simplifying the modeling process and establishes a clear mapping relationship between parameters and responses, thus providing reliable technical support for vibration control of aero-engine bladed disk systems. The method includes the following steps: establishing finite element modal analysis models of all boss blades and disks, and performing free modal calculations; using Simpack modal condensation to establish blade-disk reduced flexible body models of all boss blades and disks, and then assembling them to form a complete blade-disk coupled dynamic analysis model of the aero-engine; subsequently, simulating the frictional vibration reduction effect between boss blades by establishing pre-compressed spring damping units or custom nonlinear spring units between the bosses of adjacent boss blades; and simulating the interference fit relationship between bosses by using the spring pre-compression amount.

The invention provides a monitoring system remote regulation signal locking and unlocking method based on user behavior analysis, and the unlocking method comprises the steps: carrying out the real-time analysis of a user operation behavior through employing a long and short-term memory network algorithm; processing is carried out through the processes of user behavior data collection, data preprocessing, LSM model training, fuzzy logic classification, signal locking and unlocking, model updating and the like, the operation authority of a remote regulation signal is dynamically judged, user operation behaviors are automatically recognized and judged, locking and unlocking of the remote regulation signal are controlled, unauthorized remote regulation signal operation is effectively avoided, and the operation efficiency is improved. The system is prevented from being influenced by hostile attacks or operations, and meanwhile, when abnormal conditions are detected, automatic unlocking can be carried out after secondary verification, so that the influence caused by misoperation is reduced; the system continuously learns user behaviors, optimizes the model and adapts to different operation habits during operation, and the automatic locking and unlocking mechanism reduces human intervention, improves the operation efficiency of the system and guarantees the operation safety of the transformer substation.

The invention discloses a vertical shield propulsion system simulation device, which belongs to the field of shield machine simulation equipment. The device includes a six-axis platform system and a three-degree-of-freedom platform. The six-axis platform system propulsion system is simplified into six direct-connected electric cylinders. The six direct-connected electric cylinders are used to adjust the posture of the six-axis platform to simulate the tunneling working condition of the shield machine, thereby ensuring the response speed of the control and simplifying the model of the propulsion system. The six-axis platform system is placed on the upper platform of the three-degree-of-freedom platform in a manner of a support shoe at the bottom, and the ring composed of the six support shoes is co-centered with the upper platform. In this way, the shield propulsion system simulation device is placed vertically to reduce the floor space of the equipment, and the simulation of the pitch, yaw and other working conditions of the shield machine is realized by changing the elongation of the reciprocating electric cylinder of the three-degree-of-freedom platform. The device occupies a small area, can reduce costs, has a safe system and good stability, and can also more accurately simulate the shield propulsion system.

The invention discloses a method for checking high-current wiring in a chip, which belongs to the technical field of chip layout design, and comprises the following steps of: determining metal layers in a layout subjected to conventional checking, and obtaining the minimum width and the number of through holes required when each metal layer passes through different current magnitudes under the limit working temperature of the chip; the method comprises the following steps: writing a DRC rule according to the minimum width and the number of through holes corresponding to different current magnitudes of each metal layer, executing the DRC rule by using a Calibre DRC tool to identify a first metal layer region, displaying the first metal layer region through a Calibre RVE tool, generating width information and through hole information of each first metal layer region, and determining the width information and the through hole information of each first metal layer region according to the current magnitude of a test current needing to pass through. The modification information is provided for the first metal layer area, inspection is carried out based on the preset DRC rule, the model is simple, the computing resource requirement is low, the inspection time is shortened, the inspection efficiency is improved, meanwhile, manual inspection is reduced, and the accuracy is improved.

The invention discloses a bearing fault diagnosis method combining a multi-scale residual error and multi-task depth separable convolution, and the method comprises the following steps: carrying out the preprocessing of a bearing signal, and generating a two-dimensional coding graph; dividing the two-dimensional coding graph into a training set, a test set and a verification set; establishing a shared network, and extracting universal shared features; building a branch network, and extracting task features of the subtasks; the shared network and the branch network are combined, a multi-task classification model is constructed, and shared features are screened through an attention mechanism and then input to the branch network; inputting the training set into a multi-task classification model, and reasonably allocating computing resources by using a dynamic loss weight algorithm; and introducing online label smoothness into the softmax classifier to accelerate the convergence speed of the model, and outputting the classification result of each subtask. According to the invention, the fault features of each subtask can be effectively extracted, and efficient and accurate multi-task bearing fault diagnosis is realized.

The invention relates to the technical field of building construction calculation models, in particular to a composite lining joint surface calculation model which comprises an outer lining model and an inner lining model, the outer lining model adopts a beam-spring model for simulation, the inner lining model adopts a beam unit for simulation, and prestress is applied to the inner lining model. Prestress is applied to joints of the lining beam units in a concentrated force mode by adopting an equivalent loading method; according to the composite lining joint surface calculation model provided by the invention, the multi-directional combined springs are arranged between the composite lining joint surfaces, and the value basis or calculation method of the key parameters of each spring is given; by activating the corresponding springs, simulation of various different joint surface treatment modes such as direct bonding between the composite linings, laying of a waterproof layer and arrangement of joint bars can be achieved.

The invention discloses a method for fitting rotation of a three-dimensional vehicle speed pointer, and the method comprises the steps: S1, firstly, describing an elliptical trajectory by using a GIMP drawing tool to form an elliptical curve, and obtaining a starting point, an ending point and a control point; s2, randomly selecting M sampling points on the elliptic curve, and reading coordinates of the sampling points; s3, the distances between the M sampling points and the circle center are calculated respectively, so that the radiuses of the corresponding sampling points are obtained; s4, performing ratio calculation on the radius of the sampling point and the radius of the initial position to obtain a ratio set; s5, averagely dividing the angle from the starting point to the ending point into M parts, and enabling the M parts to be in one-to-one correspondence with the ratio sets; and S6, performing polynomial function fitting by taking the rotation angle a as an independent variable and the ratio set as a dependent variable to realize three-dimensional rotation of the pointer along the elliptical orbit. According to the invention, a special display effect can be realized through a data fitting technology, for example, in graphic processing and visualization, smooth curves and graphs can be generated through the fitting technology.

The present application belongs to the field of aerospace space orbit game and space safety, and relates to a kind of orbit pursuit and evasion interception game strategy design method based on optimization theory, specifically comprising the following processes: determining orbit game scene parameters;Based on the relative motion equation of spacecraft, a round mode pursuit and evasion interception orbit game model is established;The pursuit strategy of the pursuer is designed, the interception strategy of the interceptor is designed, and the escape strategy of the evader is designed;Based on the orbit game scene parameters and the round mode pursuit and evasion interception orbit game model, the game indexes of the three are optimized, and the optimal values of the game indexes of the three are obtained respectively. The application of the method can provide an effective solution for the pursuit and evasion interception cooperation problem of spacecraft using impulse thrust, thereby laying a theoretical foundation for the realization of space failed spacecraft takeover, space failed spacecraft removal and other engineering applications.

The present invention discloses an electromechanical transient modeling method and a topological structure of a modular multi-level matrix converter, and belongs to the field of simulation of power systems. Existing simulation modeling schemes mainly focus on the direction of electromagnetic transient modeling, and lack electromechanical transient simulation models that can be actually applied. The electromechanical transient modeling method of the modular multi-level matrix converter of the present invention constructs a simplified model of the converter transient, which is used for an equivalent modular multi-level matrix converter, and then a simplified model that can be used for an electromechanical transient simulation program can be obtained. The scheme is simple, practical, and feasible, and can effectively improve the engineering calculation accuracy and engineering calculation efficiency of the electromechanical transient modeling method, fill the gap in the electromechanical transient model of the modular multi-level matrix converter, can be used for the planning and safe and stable calculation of large power grids, and provides an important method for the research of low-frequency power transmission systems, while being easy to program and implement.