48 results about "Radial basis function neural" patented technology

This application discloses a digital twin control method for facility poultry farming environments based on deep spatiotemporal data fusion, relating to the field of poultry farming technology. It proposes a spatiotemporal sequence prediction model for environmental parameters using a bidirectional long short-term memory network (VMD-Attention-BiLSTM) that integrates variational mode decomposition and attention mechanisms, effectively overcoming the system's large time lag. A digital twin decision engine based on nonlinear model predictive control is constructed, and the optimal control sequence for energy consumption and environmental quality is solved under multiple constraints by establishing a joint state equation for thermodynamics and gas diffusion. Specifically, for complex winter conditions, a ventilation-heating collaborative game strategy based on real-time heat loss compensation is designed. Adaptive sliding mode control with radial basis function neural network compensation is introduced into the underlying execution unit, significantly enhancing the system's anti-disturbance capability. This achieves precise environmental control under complex conditions, improving poultry welfare and energy efficiency.

The application discloses a path tracking control method for a man-machine co-driving type intelligent vehicle, and steps are as follows: a vehicle path tracking error model is established, and a hierarchical control architecture is constructed, the hierarchical control architecture comprising an upper controller and a lower controller; the upper controller is configured to: in a human driving mode, generating a first front wheel steering angle through a two-point preview driver model, and in an automatic driving mode, generating a second front wheel steering angle through an automatic driving controller fusing a radial basis function neural network, a heuristic reinforcement adjustment mechanism and a non-singular fast terminal sliding mode control; the lower controller is configured to: driving a finite state machine to switch the driving mode based on a driving risk assessment index, and performing weighted fusion on the first front wheel steering angle and the second front wheel steering angle in a driving authority transition stage to output a final front wheel steering angle control instruction; the application solves the contradiction between path tracking precision and chattering, switching safety and comfort, and is suitable for man-machine collaborative control of an intelligent vehicle under multiple working conditions.

This invention relates to a method, device, and electronic equipment for underwater vehicle formation control. The method includes: constructing a formation configuration maintenance framework based on rigid graph theory, and determining the relative positional relationships between vehicles through distance constraints; establishing an obstacle avoidance control model based on the artificial potential field method, calculating the potential field force based on the distance between the vehicle and the obstacle and the distance between vehicles, thereby achieving obstacle avoidance and collision prevention between vehicles; performing online identification and compensation of unmodeled dynamics and external disturbances in the vehicle dynamics model based on a radial basis function neural network, using a Gaussian function as the activation function; and vector-fusing the formation maintenance control force, potential field force, and compensation output of the radial basis function neural network to generate control commands for each vehicle, thereby achieving formation control. This invention can achieve high-precision formation maintenance, effective obstacle avoidance, and adaptive disturbance compensation capabilities in complex underwater environments.

This invention discloses a neural network-based active disturbance rejection (ADRR) composite control method and system for permanent magnet synchronous motors (PMSMs). The method includes: constructing a second-order PMSM mathematical model; constructing a second-order nonlinear ADRR controller; generating a first control quantity based on the second-order nonlinear ADRR controller according to position commands and initial position feedback, and inputting the first control quantity into a current loop controller to control the PMSM; adjusting and optimizing the second-order nonlinear ADRR controller using a radial basis function neural network based on current position feedback and position commands; generating a second control quantity based on the adjusted and optimized second-order nonlinear ADRR controller according to position commands and current position feedback; predicting multiple sets of switching state combinations based on a dual-vector finite control set model using the second-order PMSM mathematical model and the second control quantity, and selecting the optimal switching state combination from the multiple sets of switching state combinations; and controlling the PMSM according to the optimal switching state combination. This significantly improves the real-time performance and robustness of the system.

This invention discloses a method for constructing a bridge digital twin health monitoring system, belonging to the field of civil engineering and smart infrastructure technology. The method includes: constructing a three-dimensional geometric model of the bridge; establishing an initial finite element model based on the bridge's three-dimensional geometric model, and using a genetic algorithm to optimize and correct its key parameters to obtain a physical simulation model; acquiring structural response sample data based on the physical simulation model, and constructing an RBF surrogate model using the response surface methodology or radial basis function neural network method to obtain a digital twin bridge model; integrating the digital twin bridge model into a three-dimensional visualization platform to realize visual perception, interactive operation, and remote management of the bridge structural status, thereby constructing a bridge digital twin health monitoring system.

This invention provides a method for identifying transient processes in traction networks based on phase space reconstruction and neural networks. The method includes: acquiring transient current signals; calculating the maximum Lyapunov exponent of the transient current signals; if the exponent is greater than zero, proceeding to the next step; for transient current signals that meet the phase space analysis conditions, adaptively determining the optimal delay time using the average mutual information method and adaptively determining the optimal embedding dimension using the pseudo-nearest neighbor method; reconstructing the phase space of the transient current signals to obtain a reconstructed phase space vector; extracting dynamic attribute features reflecting the dynamic characteristics of the transient process and constructing a joint feature vector; inputting the joint feature vector into a pre-established and trained radial basis function neural network classification model, and outputting the category identification result of the transient process through the model. This invention can solve the problems of poor identification ability, difficulty in fully revealing the deep dynamic characteristics of transient signals, and limited adaptability to complex scenarios in existing technologies.

This invention belongs to the technical field of compressor aerodynamic stability analysis. It proposes a deterministic learning-based method and system for analyzing the aerodynamic stability of axial compressors. Using pulsating pressure data during the aerodynamic instability development process of an axial compressor, a deterministic learning identification algorithm based on radial basis function neural networks is used to model the nonlinear dynamics of the compressor's internal flow field, thereby obtaining dynamic diagrams of normal and instability precursor states. Dynamic indices are extracted from these diagrams to construct dynamic feature vectors describing the compressor's aerodynamic stability. Based on the dynamic feature vectors corresponding to different operating states, the dynamic classification boundary between normal and instability precursor states is determined. The distance between the dynamic feature vector of the compressor data under test and the dynamic classification boundary is calculated to determine the compressor's operating state. This invention enables early warning of aerodynamic instability and has good interpretability and engineering application value.

This invention relates to the field of intelligent control technology for nonlinear systems, and discloses a controller design method based on radial basis function (RBF) network compensation. The method includes: estimating the uncertainties of an aircraft's nonlinear system based on a radial basis function neural network; constructing an aircraft attitude controller based on the tracking errors of the attitude subsystem and the attitude angular rate subsystem, and the estimation results; and obtaining an adaptive update law for the weights of the radial basis function neural network based on the aircraft attitude controller. Therefore, the unmodeled characteristics of the system can be approximated through online learning using a radial basis function neural network, and unknown disturbances can be compensated for in the controller, greatly improving the controller's control accuracy and its ability to handle unknown disturbances.

This application discloses a multi-agent distributed adaptive dynamic dual-event triggered consensus control method and system, belonging to the field of multi-agent system control technology. It includes online approximation of unknown nonlinear terms using a radial basis function neural network and constructing a disturbance observer to estimate composite uncertainties, providing an achievable disturbance estimation signal for control compensation. Furthermore, it recursively designs virtual control and adaptive laws within a dynamic surface control framework, avoids derivative explosion in backstep design using a first-order filter, and combines constraint potential functions to handle state constraints, ensuring the boundedness of the closed-loop signal. It constructs control update trigger times and communication reporting trigger times separately, and designs trigger functions based on the errors of the two branches to reduce the control update and communication overhead of the networked system.

The application discloses a kind of cluster unmanned plane fixed time quick adaptive fault-tolerant cooperative control method, first, design a kind of elasticity distributed fixed time observer based on topology switching signal, for accurately estimating virtual leader state and weakening the influence of denial-of-service attack.Second, by combining power integrator technology with topology switching signal, for follower unmanned plane affected by unknown actuator fault, construct fixed time adaptive fault-tolerant cooperative control protocol, so that the cluster unmanned plane system can still realize trajectory tracking in fixed time under the condition of actuator fault and denial-of-service attack coexistence.Finally, by radial basis function neural network, online approximation of composite nonlinear disturbance is realized, more accurate cooperative control is realized.The application can be applied to cluster unmanned plane formation fault-tolerant cooperative control under actuator fault and denial-of-service attack.

This invention provides a control method and system for a five-degree-of-freedom marine crane in a non-inertial frame, relating to the field of crane automatic control technology. The method includes: acquiring the real-time state variables of the crane in a non-inertial frame; inputting the state variables into an adaptive radial basis function neural network to obtain a lumped disturbance estimate, the neural network being designed based on a five-degree-of-freedom dynamic model established in the ship's coordinate system using the Lagrange equations and the principle of virtual work, considering the ship's six-degree-of-freedom motion and installation position offset; calculating a tracking error function based on the state variables and the desired trajectory; inputting the disturbance estimate and the error function into a feedback controller to calculate and generate a control torque, which is then output to the actuator to drive the crane to track the desired trajectory and suppress load sway. This invention achieves direct modeling and control in a non-inertial frame, solving the coordinate transformation disturbance and singularity problems caused by inertial frame modeling, and improving trajectory tracking accuracy and disturbance rejection robustness.

This invention provides a fixed-time control method for a flexible double-link manipulator, applied in the field of robotics intelligence. The method includes: establishing a dynamic model of a flexible system with dead-zone input; deriving the initial control torque dependent on the dynamic model of the flexible system with dead-zone input using backstepping; designing a first radial basis function neural network update rate based on the unknown dynamic information in the initial control torque approximated by a radial basis function neural network; designing a second radial basis function neural network update rate based on the approximation of the unknown dead-zone function by the radial basis function neural network; and obtaining the fixed-time control torque of the novel flexible double-link manipulator based on the two radial basis function neural network update rates and the initial control torque. The fixed-time control strategy proposed in this invention ensures the rapid convergence of the flexible double-link manipulator, and can simultaneously improve trajectory tracking accuracy and vibration suppression performance even under operating conditions involving system constraints.

This invention provides a trajectory planning and intelligent control method for a three-stage swing-type overhead crane, relating to the technical field of intelligent overhead crane systems. The method includes: defining a planar coordinate system and determining the position coordinates of the crane's trolley, hook, frame, and load within the planar coordinate system; obtaining a dynamic model using the Lagrange equation based on the position coordinates; designing a controller based on the dynamic model; setting up a two-layer radial basis function neural network and determining the control force; and determining the overall control law based on the position coordinates, the controller, and the control force. According to this invention, intelligent anti-sway control of the three-stage swing system improves control quality, enhances the system's anti-interference capability, and improves the overall system safety level.

This invention discloses a method for predicting the rock fragmentation distribution in open-pit deep-hole blasting, comprising the following steps: acquiring historical blasting data; constructing a radial basis function neural network; inputting blasting design parameters and rock coefficients from the historical blasting data into a nonlinear feature cross layer to generate an enhanced feature vector; inputting the enhanced feature vector into a prior knowledge-guided hidden layer center selection layer to determine multiple center points of the radial basis functions, and assigning a corresponding radial basis function to each center point according to a preset rule; inputting the output result into a nonlinear output mapping layer to perform a nonlinear transformation on the output of the radial basis functions; inputting the nonlinearly transformed features into a multi-task output layer to simultaneously predict and output multiple different fragmentation feature values; and constructing a rock fragmentation distribution curve based on the multiple different fragmentation feature values ​​to complete the prediction of the blasting fragmentation distribution.

This application discloses a flight control method for trajectory tracking of a coaxial dual-rotor UAV and the UAV itself, relating to the field of UAV technology. The method includes: establishing a dynamic model of the coaxial dual-rotor UAV (CRA) based on the Euler-Poincaré equations; establishing a control objective for an adaptive robust trajectory tracking control method for the CRA based on the flight characteristics of CRA trajectory tracking; setting an attitude controller based on an adaptive radial basis function neural network (RBFNN) according to the control objective; and setting a position controller based on adaptive parameter estimation according to the motion characteristics of CRA trajectory tracking; and controlling the coaxial dual-rotor UAV using the attitude controller and the position controller. This application achieves precise and stable trajectory tracking control of the CRA by improving the accuracy of the mathematical model and the robustness of the CRA controller.

This invention discloses a method and system for stabilizing a robotic arm based on fractional-order sliding mode adaptive compensation. The method includes: constructing a unified dynamic model incorporating time-varying delays, nonlinear disturbances, and neural network approximation errors; defining the sliding mode function using Caputo fractional derivatives, establishing an improved trend law based on the improved Fal function and hyperbolic tangent function, and obtaining a control law based on the error-driven dynamic model and the improved approach law; constructing a dynamic adaptive compensator linked to time-delay gradient information to address the residual errors of the radial basis function and the unknown nonlinear terms in the neural network approximation, and dynamically adjusting the compensation gain based on the dynamic adaptive compensator; constructing a Lyapunov functional adapted to the fractional-order system, deriving the exponential stability criterion for the closed-loop system using Young's inequality and Halany's inequality, and determining the quantitative correlation between the fractional-order order and the convergence rate. This invention solves the problems of imbalance between robustness and smoothness and weak time-delay adaptability in existing technologies, significantly improving control accuracy and reliability.

This invention discloses a backstepping-based variable iterative learning control method for robotic arms, applicable to motor-driven robotic arm systems. This method converts the robotic arm system model into a third-order strict feedback form. By defining coordinate transformation error and introducing a command filter, an auxiliary system is constructed to compensate for the deviation between the filter and the virtual control law. A radial basis function neural network is used to approximate the unknown dynamics and disturbances of the system online. A backstepping controller is designed based on the compensation error, and a parameter learning law with variable iteration length is constructed to ensure system stability and convergence of the compensation error. By dynamically adjusting the iteration length during the control phase, this method can reduce computational resource consumption while maintaining high-precision tracking performance. When used for robotic arm control of repetitive tasks, it significantly improves the tracking accuracy and robustness of the system, demonstrating good engineering application value.

This invention relates to the construction and application of a collaborative prediction model for rock strength and toughness properties. Addressing the problems of size effect, single-parameter prediction limitations, and difficulties in cross-scale mapping in existing rock strength and toughness parameter testing technologies, the following key technical solutions are proposed: constructing a model that includes geometric parameters (L / W / S / B) and crack characteristics (…). a 0 / Δ a fic ), particle size parameters ( g av A dataset with 9-dimensional multi-source heterogeneous features, including Whale Optimization (WOA), is used to adaptively optimize the learning rate and hidden layer nodes of a Radial Basis Function Neural Network (RBFNN) to establish... f t - K IC A two-parameter synchronous prediction model. This model reveals the virtual crack propagation (Δ) through SHAP interpretability analysis. a fic ) and average particle size ( g av ) is the key controlling variable, for f t and K IC The predictive determination coefficient ( R ²) Reaching 0.997 and 0.996 respectively, it breaks through the limitation of small sample size effect and realizes cross-scale mapping from laboratory parameters to engineering scale, providing efficient and accurate technical support for the stability assessment of deep rock mass engineering.

This application belongs to the field of medical devices and protective equipment, specifically involving a design method and layered structure for an enhanced mouthguard based on musculoskeletal stability optimization. The method includes: collecting multimodal data such as the full dentition morphology, jawbone structure, mandibular movement trajectory, surface electromyography, and plantar pressure of the subject; constructing a correlation system between mandibular position and overall stability using a radial basis function neural network; setting anatomical boundaries, joint spaces, and muscle symmetry constraints; and using a genetic algorithm to iteratively optimize and lock in a personalized musculoskeletal stability position; based on this, performing layered digital design of the mouthguard's flexible base layer and posterior tooth retention components, and then 3D printing manufacturing. This application, through precise optimization of jaw position, achieves a leap from passive protection to actively enhancing body stability in sports mouthguards, significantly improving exercise efficiency and wearing comfort while ensuring physical protection.

This application proposes a method, apparatus, device, and storage medium for rapid prediction of underwater vehicle flow noise, relating to the field of underwater radiated noise prediction technology. The method includes: acquiring far-field radiated noise data of multi-configuration underwater vehicles using numerical simulation, and preprocessing the data to obtain training and testing datasets; training a hybrid prediction model using the training and testing datasets to obtain a target hybrid prediction model, which includes a kernel ridge regression prediction sub-model and a radial basis function neural network prediction sub-model. The prediction result of the hybrid prediction model is obtained by weighted summation of the prediction results of the two prediction sub-models based on the coefficient of determination, and subtracting the average deviation of the training dataset; inputting the configuration data of the underwater vehicle to be predicted into the target hybrid prediction model to obtain the flow noise prediction result of the underwater vehicle to be predicted. This application can achieve rapid and accurate prediction of the flow noise characteristics of underwater vehicles through the above method.

The application discloses a kind of self-adapting global fractional order terminal sliding mode serpentine robot control methods.In order to solve the problem of high complexity in modeling process, it is difficult to solve, an error dynamics model is constructed, and the design difficulty of control system is further simplified. In response to the uncertain terms and unknown disturbances that may exist in the model, a radial basis function neural network observer is introduced, which effectively approximates the estimation of these uncertain factors. In order to ensure that the joint serpentine robot system can converge quickly, an adaptive constant rate control strategy is designed, and combined with the global fractional order terminal sliding surface, not only the high-precision control of the system is realized, but also the robustness of the system is significantly enhanced.

This invention discloses a synchronous tracking control system and method for a dual-motor steer-by-wire system, comprising: establishing a model of the dual-motor steer-by-wire system; calculating the target front wheel steering angle to maintain vehicle stability; adjusting the target steering angle signals of the two steering motors in real time based on torque feedback control principles and a virtual master shaft model; estimating the nonlinear state functions of the state equations of the two steering motors in real time based on an adaptive radial basis function neural network, and calculating the current control signals of the two steering motors based on this; and calculating the current compensation control signals of the two steering motors based on the mean-deviation coupling synchronous control and variable approach law sliding mode control principles. This invention can effectively cope with perturbations and uncertain disturbances in the model parameters of the steering motors, ensuring the synchronous tracking performance of the dual-motor steer-by-wire system.

This invention discloses an adaptive preset-time bounded tracking control method under state constraints, belonging to the field of complex nonlinear system control technology. First, it proposes an actual preset-time bounded stability criterion, providing a theoretical basis for the artificial preset of the system convergence time. Second, it designs an adaptive actual preset-time bounded filter, achieving preset-time convergence of the filtering error while addressing the computational complexity explosion problem in backstepping control. Third, it maps the constrained system state to unconstrained variables through a state transformation function, naturally integrating state constraints into the controller design. Finally, it combines backstepping control and radial basis function neural networks to construct an adaptive preset-time bounded tracking controller. In this invention, all signals converge to a bounded range within a preset time, the system state always satisfies the preset constraints, and the tracking error reaches bounded stability within a preset time. This method is suitable for controlling high-order strict feedback nonlinear systems with state constraints.

The application relates to the technical field of permanent magnet synchronous motor control, in particular to a permanent magnet synchronous motor adaptive integral sliding mode control method based on a command filter, a second-order derivative and a third-order derivative of a position tracking error and a first-order derivative and a second-order derivative of a d-axis current tracking error are used to construct a second-order and third-order integral sliding mode surface based on the command filter, the inhibition capability of lumped interference is effectively enhanced, and the robustness of the system is significantly improved; for state-related uncertainty and time-varying interference, a radial basis function neural network is introduced to realize online approximation, and an interference adaptive law is combined to realize real-time estimation of the interference, effective estimation and compensation control of the radial basis function neural network and the interference adaptive law based on a scaling factor on the model uncertainty and time-varying interference of the permanent magnet synchronous motor are realized, so that the reliability and anti-interference performance of the controller are greatly improved without an accurate system model, and the tracking error is effectively reduced.

This invention relates to the field of solar power generation technology, and particularly to a control method for a dish concentrator based on an I3-RPS tracking mechanism. The control method includes: establishing a solar direction vector calculation model; establishing a multi-coordinate system to describe the motion of the dish concentrator, and deriving the homogeneous transformation matrix between the coordinate systems; obtaining the pose of the moving platform based on the solar direction vector, and establishing a mapping relationship between the length, velocity, and acceleration of the telescopic rod and the pose of the moving platform through inverse kinematics; establishing an inverse dynamics model based on the principle of virtual work; designing a hierarchical composite control strategy; at the top level, using a parallel synchronous architecture to coordinate the motion of the I3-RPS tracking mechanism; at the bottom level, designing an RBF-FSMC controller for the I3-RPS tracking mechanism, replacing the switching function with a saturation function, dynamically adjusting the saturation function gain and boundary layer thickness through fuzzy rules, and introducing a radial basis function neural network to approximate system modeling errors and external disturbances online. This control method exhibits high tracking accuracy and strong anti-interference capability.

This application relates to the field of robot control technology, and discloses a control method and system for an exoskeleton lower limb rehabilitation robot. The method includes: generating a gait trajectory based on patient information and collecting joint interaction forces and posture angles; statistically processing the joint interaction forces to obtain fatigue indices, and quantifying the posture angles to obtain imbalance indices; using a two-layer radial basis function neural network to output fatigue compensation parameters and posture correction parameters respectively, adaptively determining the fusion weights to generate final impedance parameters; and calculating and executing the joint driving torque. This application solves the problem that existing fixed impedance control cannot simultaneously cope with both dynamic changes in patient fatigue accumulation and posture imbalance.

This invention discloses an adaptive fault-tolerant control method and system for a multi-flexible robot system with preset indices, belonging to the field of fault-tolerant control technology for control systems. The method includes: establishing a dynamic model of the multi-flexible robot and a virtual leader model containing hub angle position, elastic vibration, and total displacement; constructing an intermittent actuator fault model including the number of faults, intervals, and unknown times; approximating system uncertainties using a radial basis function neural network; defining the tracking error and reconstructing the error dynamic equation, introducing a strictly decreasing function to construct a preset index set of boundary error state variables; converting constrained errors into unconstrained variables using a strictly increasing function; designing an adaptive fault-tolerant controller based on the unconstrained variables, and using a projection operator to update the estimated parameters online. This invention achieves consistent angular position tracking and suppression of elastic vibration, with the error converging to a predefined residual set, requiring no prior fault information, thus improving the system's robustness and engineering practicality.

This invention discloses a method for identifying anomalies in tailings dam seepage monitoring data based on a grey model and neural network. First, monitoring data is acquired, preprocessed using the arithmetic square root, and generated by a single accumulation. Then, a GM(1,1) grey model is established, and a dynamic adaptive threshold interval is constructed to coarsely screen anomaly data. Data is updated using a metabolic process; data exceeding the threshold is considered anomaly. Next, the grey correlation degree between the anomaly data and related measuring point data is calculated, and the input variable for the radial basis function neural network is selected accordingly. The network is trained to obtain the neural network's seepage prediction value, and the relative error is used to determine whether the anomaly is caused by environmental factors. Unidentified anomalies are retested and verified, and instruments are checked. If the retest is normal, the measurement error data is removed or corrected; if the retest is abnormal, it is determined to be a structural change and an early warning is issued. This invention combines a grey model and a neural network to achieve efficient and accurate identification of anomaly data and automatic identification of environmental factor responses.