8 results about "Kinetic theory" patented technology

The application belongs to the technical field of vehicle structure design, and particularly relates to a high-speed tracked vehicle structure lightweight design method for task reliability, which is based on a dynamics theory model to establish a high-speed tracked vehicle whole-vehicle multi-level mass matrix and a stiffness matrix, generates a whole-vehicle lightweight mass matrix according to a vehicle body and tracked joint lightweight design scheme, calculates lightweight design parameters of each lightweight design scheme for task reliability, determines an optimal lightweight design scheme, calculates modal kinetic energy distribution of the optimal lightweight design scheme, and determines an optimizable stiffness parameter. The application has the advantages of improving modeling efficiency, increasing lightweight design scheme reliability, refining optimizable stiffness parameters, reducing research and development costs, and the like, and can provide an effective high task reliability scheme for high-speed tracked vehicle structure lightweight design.

The application provides a quad-rotor unmanned aerial vehicle anti-disturbance control method, and specifically comprises the following steps: S1, a nonlinear model predictive controller outputs expected angular velocity to an angular velocity model reference adaptive controller; S2, the nonlinear model predictive controller outputs expected total thrust to a thrust model reference adaptive controller; S3, a rigid body dynamics theory is used to obtain a quad-rotor dynamics model, a collision impact disturbance analysis is performed on the angular velocity model reference adaptive controller, body torque instructions are output according to the quad-rotor dynamics model, a load uncertainty analysis is performed on the thrust model reference adaptive controller, and total thrust instructions are output according to the quad-rotor dynamics model; S4, a control distribution module is used to map the body torque instructions and the total thrust instructions to thrust instructions of each rotor of the quad-rotor unmanned aerial vehicle; and S5, the calculated thrust instructions are transmitted to the quad-rotor for execution.

The invention relates to the field of robot pollutant detection, and provides a dust and snow removal robot pollutant detection method and system based on deep learning, and the method comprises the steps: collecting an original accumulated snow image sequence, a physical wheel speed, a motor torque current and a meteorological feature vector in parallel, and generating a multi-modal data set; generating a visual perception and dynamic theory velocity field by using a dense optical flow algorithm and perspective projection logic, generating a residual feature map through vector difference, and extracting texture features; driving a cross-modal attention mechanism based on an environmental thermodynamics context label, fusing residual features and texture features, and outputting accumulated snow vertical structure prediction through a full-connection multi-branch prediction network; and executing decision branch logic and a down pressure balance equation to calculate walking driving control parameters and execute an action instruction, and generating an execution control instruction set in combination with motor torque current. According to the method, time sequence meteorological prior and heterogeneous sensor data are fused, and an accumulated snow layered recognition and quantification mechanism based on cross-modal attention is constructed.

The invention discloses a ternary system aluminum alloy component design method based on phase diagram calculation. According to the method, an equivalent precipitation diagram of the ternary system aluminum alloy is calculated according to a thermodynamic theory in phase diagram calculation, alloy components with the lowest quenching sensitivity under the equal strength are designed for alloys on a series of equivalent precipitation curves according to a kinetic theory, namely a TTT curve, in phase diagram calculation, and finally the alloy components with the quenching sensitivity under the equal strength are designed. Through a relational graph of the theoretical mass fraction of the eta phase and the maximum nose tip time, the relation between the strength and the quenching sensitivity is established, and component design of optimal strength and quenching sensitivity balance of the ternary system aluminum alloy is achieved.

The invention discloses a pump turbine guide vane opening degree optimization method based on entropy production and a vortex dynamics theory, and belongs to the technical field of water turbines, and the method comprises the following steps: S1, geometric model construction and verification; comprising the steps of establishing a pump turbine three-dimensional geometric model and performing grid division; s2, vortex dynamics and entropy production analysis; s3, multi-objective optimization decision making is carried out, and the optimal guide vane opening degree is obtained; according to the method, the entropy generation theory and the vortex dynamics are fused, a new guide vane opening degree optimization scheme is provided, the bottleneck is broken through, and an innovative solution is provided for accurately diagnosing hydraulic loss caused by a vortex structure and optimizing the guide vane opening degree.

The invention discloses a flexible actuator optimal position tracking control method based on predictive control and a disturbance observer. The method is used for improving the control performance of a flexible actuator system. According to the method, firstly, a dynamic model of a flexible actuator SEA system is established based on the dynamics theory, then, generalized predictive control (GPC) is adopted as a basic control method, Taylor series expansion is used for position prediction, performance indexes are constructed in combination with rolling optimization, and finally, matching interference and mismatching interference in predictive control are subjected to optimal optimization. Two reduced order disturbance observers (RDOBs) are introduced and are a motor end disturbance observer and a load end disturbance observer respectively, and the magnitude and the direction of disturbance are estimated from output of a system and used for compensating disturbance of the motor end and the load end respectively. According to the method, the effectiveness of the proposed method is proved in several experiments performed by the exoskeleton robot driven by the flexible driver, and verification shows that when external disturbance is suddenly added and removed, the GPC + RDOB can quickly recover the system performance, and the anti-interference capability is superior to that of PD and GPC.

The application discloses a kind of chain multi-link flexible joint mechanical arm dynamics simulation method based on deep reinforcement learning, and it is a kind of model-driven method.First, based on deep reinforcement learning and chain multi-link flexible joint dynamics theory, then the rigid-flexible coupling dynamics model of chain multi-link mechanical arm considering joint flexibility and joint damping is established by floating coordinate system method, finally, trajectory tracking problem is studied based on the mathematical model in combination with deep learning.The chain flexible joint mechanical arm dynamics simulation method based on deep reinforcement learning has good trajectory tracking effect, and has important value for actual chain multi-link flexible joint trajectory tracking control.

This invention belongs to the field of blast furnace ironmaking technology, specifically disclosing a method for constructing a three-dimensional numerical model of the gas-solid reaction flow in the vortex zone of a hydrogen-rich carbon-circulating oxygen blast furnace. The method involves geometric modeling and mesh refinement, definition of an Eulerian multiphase flow system, and establishment of the gas-solid two-phase mass-momentum-energy conservation equations. The standard turbulence model, particle dynamics theory, and Gidaspow drag model are used to close the equations, which are then solved using the finite volume method. The accuracy of the model is verified by combining actual production data. This solves the problem that traditional models cannot be adapted to high-precision simulation of hydrogen-rich carbon-circulating oxygen blast furnaces. Results show that the simulation error in the depth of the vortex zone is low, meeting the accuracy requirements, and achieving refined simulation of the vortex zone morphology, temperature field, and composition field.