31 results about "Nonlinear decoupling" patented technology

The invention relates to a quadruped robot movement track control method based on a PSO-PD (Particle Swarm Optimization-PD) neural network. The quadruped robot movement track control method comprises the following steps: (1) calculating a target falling point of a robot on a movement track; (2) inputting the target falling point and a feedback body gravity center position into an input layer of the PSO-PD neural network, performing linear conversion, inputting into a first hidden layer, performing ratio calculation and differential operation, performing linear conversion, inputting into a second hidden layer, and acquiring instruction displacement in an x direction and a y direction, inputting into a third hidden layer, calculating an instruction displacement distance, instructing a body direction, inputting into a fifth layer, performing foot tract control and steering control on a quadruped robot, inputting into a sixth layer, adjusting the posture of the quadruped robot in case of disturbance, inputting into a seventh layer, calculating the body direction and the body gravity center position of the quadruped robot, feeding back the body direction to the fifth layer, and feeding back the body gravity center position to the input layer. The quadruped robot movement track control method has relatively good non-linear decoupling control capability, is precise in control and good in stability, and has relatively good anti-interference properties.

The invention relates to an overall input decoupling device for a multi-rotor unmanned aerial vehicle. The device comprises a coefficient matrix Mco storage module, a structure matrix Mrel storage module and a leading rotor speed operation module, wherein a coefficient matrix Mco and a structure matrix Mrel are multiplied and inversed to obtain a module of an inverse matrix (Mco.Mrel)<-1>. By non-linear decoupling mapping, the input decoupling of the multi-rotor unmanned aerial vehicle is realized under the condition of unknown models, so that the multi-rotor unmanned aerial vehicle, namely amulti-input multi-output system is decoupled into a plurality of single-input single-output subsystems, and each degree of freedom is only influenced by one subsystem loop. Therefore, the degree of freedom (positions and postures) is independently controlled, the control complexity of the system is reduced, the system is easy to control, and the device does not depend on a mathematical model of an object, has a wide application range and is suitable for multi-rotor unmanned aerial vehicles with arbitrary structures.

The invention discloses a bearing-free asynchronous motor suspension system fault-tolerant controller and a construction method of the bearing-free asynchronous motor suspension system fault-tolerant controller. A support vector machine and an integrator corresponding to the support vector machine are used for constructing a normal inverse model and a fault inverse model of a composite controlled object respectively and form an inverse model base. The inverse models in the inverse model base are used as inverse models of the composite controlled object, corresponding vector coefficients and threshold values of the support vector machine are adjusted and determined to enable the inverse model base to have the inverse system function. The inverse model base and the composite controlled object are connected in series to obtain a fake linear system. The fake linear system is decoupled into two two-order single-input single-output displacement linear sub systems, and two additional controllers are introduced into the two two-order displacement linear sub systems respectively to construct a linear closed-loop controller. The linear closed-loop controller and the inverse model base collectively form a support vector machine inverse fault-tolerant controller. In the situation of normal operation and operation with faults, nonlinear decoupling control of a bearing-free asynchronous motor suspension system is achieved, and good fault tolerance and robustness are achieved.

The invention discloses a flight control law design new method considering both transient response and robust stability. The method comprises the following steps: (1) an aircraft strong non-linear model is built; (2) a force and moment surrogate model is built; (3) an equivalent linear model is built; and (4) in combination with a control inner loop and a control outer loop, a new flight control law is designed. The method of the invention has the beneficial effects that a composite nonlinear feedback control theory and a variable factor technique are used for compromise design and analysis on static and dynamic performance indexes of a closed-loop control system, the flight control system is ensured to have good tracking performance, strong robust stability and satisfactory nonlinear decoupling control ability, the flight control law application range is enhanced, the complicated control quality requirements of the modern aircraft are met, and the mutual restriction problem between dynamic response and robust stability in the flight control system design is solved.

The invention discloses a nonlinear decoupling method used for wheel force transducers. The method includes the following steps: 1, extending transducer input-output linear relation into a nonlinear form according to nonlinear characteristics of a wheel force transducer system and static calibration results; 2, deducing the matrix form of the input-output relation of each channel according to an established quadratic equation set; and 3, obtaining a decoupling matrix according to established matrix relation, and calculating real multi-dimensional wheel force/torque by using the matrix so thatmulti-dimensional signal decoupling can be realized. Through the method, the degree of coupling between measuring channels can be greatly reduced, and obtained force/torque signals can be maximumly approached to real ground-wheel force/torque signals.

The present invention provides a method for controlling a multivariable input EHA system based on model prediction control. The method comprises a step of defining state variables of the multivariable input EHA system, and establishing a state variable equation model of the multivariable input EHA system, a step of carrying out model prediction control on the state variable equation model, establishing a target function of the model prediction control, solving the optimal solution of the target function through rolling optimization, and obtaining virtual input of the multivariable input EHA system, and a step of carrying out control distribution on the real input amount of the multivariable input EHA system by using the virtual input of the multivariable input EHA system. The method provided by the invention can be used for the input control of the multivariable input EHA system, a nonlinear decoupling control problem in the multivariable input EHA system is effectively solved, thus a research foundation is provided for multivariable input EHA system control, and a development direction correspondingly is provided for the further improvement of working efficiency and frequency of the multivariable input EHA system.

The invention discloses a vinyl chloride rectification temperature control method based on a fuzzy neural network, and relates to the field of vinyl chloride rectification production control. The vinyl chloride rectification temperature control method comprises the steps that firstly, sensor signals are discretely collected through a signal collecting module, and are subjected to filtering processing; secondly, fuzzy neural network control is conducted according to signal collecting module output, fuzzification, fuzzy reasoning, fuzzy decision-making and defuzzification output are conducted insequence to obtain required control parameter values, and nonlinear decoupling control is directly conducted on an actuator; and finally, through a BP learning algorithm, the link weight, the gaussian function center value and the width value in a controller are learned and corrected, thus the temperature error of a rectifying tower is converged after output of the controller, and the stability of the system is improved on the basis of improving controlling precision.

The invention discloses a temperature compensation and inter-dimensional decoupling method for a pore pressure static sounding probe. The method comprises the steps of firstly establishing a temperature compensation model according to the measurement modes of a multifunctional pore pressure static sounding probe and a three-channel sensor, taking the output of the multifunctional pore pressure static sounding probe as sample data, obtaining the model parameters through regression, and carrying out the temperature compensation; secondly, according to the working principle of the multifunctionalpore pressure static sounding probe, analyzing the relationship between the load of the multifunctional pore pressure static sounding probe and the output of the multifunctional pore pressure staticsounding probe sensor, establishing a nonlinear regression model, taking the output of the multifunctional pore pressure static sounding probe as an initial sample, returning model parameters by adopting a least square method, and carrying out nonlinear decoupling. According to the present invention, the problems of temperature influence and dimensional coupling of the multifunctional pore pressure static sounding probe are solved, and the real load output of the multifunctional pore pressure static sounding probe is achieved.

The invention discloses a control system and method for a rectifier of a charging pile of an electric vehicle. The control system comprises a direct-current-side voltage sensor, an alternating-current-side voltage sensor, a current sensor, an active disturbance rejection control module with a reduced-order expansion state observer, a PI controller, a coordinate transformation module, a nonlinear decoupling module and an SVPWM module, wherein the alternating-current-side voltage sensor and the current sensor are connected with the input end of the coordinate conversion module, and the coordinate conversion module and the direct-current-side voltage sensor are jointly connected with the active disturbance rejection control module; the active disturbance rejection control module is connected with the PI controller; the PI controller is connected with the nonlinear decoupling module; the nonlinear decoupling module is connected with the SVPWM module; and the SVPWM module is connected with a rectifier bridge circuit in the rectifier. According to the invention, by using the active disturbance rejection control module with the reduced-order expansion state observer, the order of the observer is effectively reduced.

The invention relates to the technical field of battery charging,and concretely relates to a control system and algorithm of a PWM rectifier. The system comprises a main circuit module, a coordinate conversion module, a PWM rectifier control module and a coordinate inverse transformation module which are sequentially arranged to form a closed loop. The PWM rectifier control module comprises a voltage sliding mode control unit, a current inner loop PI controller and a nonlinear decoupling module which are arranged in sequence. The control method has the advantages that after sliding mode control is adopted, the system can enter a stable state more quickly, the overshoot of the output voltage is smaller, the response speed is higher, and the PWM rectifier can be effectively, quickly and accurately controlled.

A water sample COD concentration nonlinear decoupling control method based on differential geometric feedback control comprises the following steps: S1, establishing a chemical oxygen demand model qCOD = F (T, H, q1, q2, q3, q4, q5), and inputting a model constraint; S2, analyzing the characteristics of the chemical oxygen demand model qCOD = F (T, H, q1, q2, q3, q4, q5), and judging whether a system can use a differential geometry feedback control method according to whether the Jacobian matrix of the nonlinear system F is full rank; and if the Jacobian matrix is full rank, that is, the Jacobian matrix accords with a relative order, determining that a differential geometric feedback control method can be used; S3, if the nonlinear system F meets the condition of a relative order, calculating a feedback transformation matrix phi of differential geometry feedback control; S4, performing matrix transformation on the feedback transformation matrix phi; S5, processing measurement data in a new domain; and S6, inversely transforming a result obtained in the new domain into an actual measurement domain. According to the invention, a differential geometric feedback control analysis method is applied to input and output analysis of a measurement system aiming at nonlinear characteristics of a system description function F.