不确定连续系统预设性能非线性PID控制

研究不确定连续系统仅输入输出数据可利用情况下的非线性PID控制问题.在系统参考输入信号和输出信号连续可微的条件下引入预设性能控制理论,并基于预设性能控制的性能函数和误差转换思想设计了简单的非线性函数以组合形成非线性PID控制器,进而从理论上证明其可行性.仿真实验表明,所设计的控制器继承了传统PID的优点,且参数调整更灵活,系统的鲁棒自适应性和抗扰性更好.

Identification of dynamic systems using support vector regression neural networks

A novel adaptive support vector regression neural network （SVR-NN） is proposed, which combines respectively merits of support vector machines and a neural network. First, a support vector regression approach is applied to determine the initial structure and initial weights of the SVR-NN so that the network architecture is easily determined and the hidden nodes can adaptively be constructed based on support vectors. Furthermore, an annealing robust learning algorithm is presented to adjust these hidden node parameters as well as the weights of the SVR-NN. To test the validity of the proposed method, it is demonstrated that the adaptive SVR-NN can be used effectively for the identification of nonlinear dynamic systems. Simulation results show that the identification schemes based on the SVR-NN give considerably better performance and show faster learning in comparison to the previous neural network method.

Application of Generalized Predictive Adaptive Control Algorithm in the Design of Missile Control System

To study the application of the generalized predictive adaptive control algorithm in missile control system, the algorithm is presented based on the recursive least square estimation, and a controller of the pitch channel of a missile is designed by using this algorithm. The simulations verify that the designed controller can meet the demands of the task well.

Robust adaptive controller design for a class of nonlinear systems with unknown high frequency gains

In this note, a robust adaptive control scheme is proposed for a class of nonlinear systems that have unknown multi-plicative terms. Unlike previous results, except for the unknown control directions, we do not require a priori bounds on the unknown parameters. We also allow the unknown parameters to be time-varying provided that they are bounded. Our proposed robust adaptive controller is designed to identify on-line the unknown control directions and is a switching type controller, in which the controller parameters are tuned in a switching manner via a switching logic. Global stability of the closed-loop systems have been proved.

Adaptive robust output force tracking control of pneumatic cylinder while maximizing/minimizing its stiffness

The system considered in this work consists of a cylinder which is controlled by a pair of three-way servo valves rather than a four-way one.Therefore,the cylinder output stiffness is independently controllable of the output force.A discontinuous projection based adaptive robust controller (ARC) was constructed to achieve high-accuracy output force trajectory tracking for the system.In ARC,on-line parameter adaptation method was adopted to reduce the extent of parametric uncertainties due to the variation of friction parameters,and sliding mode control method was utilized to attenuate the effects of parameter estimation errors,unmodelled dynamics and disturbance.Furthermore,output stiffness maximization/minimization was introduced to fulfill the requirement of many robotic applications.Extensive experimental results were presented to illustrate the effectiveness and the achievable performance of the proposed scheme.For tracking a 0.5 Hz sinusoidal trajectory,maximum tracking error is 4.1 N and average tracking error is 2.2 N.Meanwhile,the output stiffness can be made and maintained near its maximum/minimum.

Robust adaptive control for a class of uncertain non-affine nonlinear systems using neural state feedback compensation

A robust adaptive control is proposed for a class of uncertain nonlinear non-affine SISO systems. In order to approximate the unknown nonlinear function, an affine type neural network(ATNN) and neural state feedback compensation are used, and then to compensate the approximation error and external disturbance, a robust control term is employed. By Lyapunov stability analysis for the closed-loop system, it is proven that tracking errors asymptotically converge to zero. Moreover, an observer is designed to estimate the system states because all the states may not be available for measurements. Furthermore, the adaptation laws of neural networks and the robust controller are given based on the Lyapunov stability theory. Finally, two simulation examples are presented to demonstrate the effectiveness of the proposed control method. Finally, two simulation examples show that the proposed method exhibits strong robustness, fast response and small tracking error, even for the non-affine nonlinear system with external disturbance, which confirms the effectiveness of the proposed approach.

Efficient WENOCU4 scheme with three different adaptive switches

Although classical WENOCU schemes can achieve high-order accuracy by introducing a moderate constant parameter C to increase the contribution of optimal weights,they exhibit distinct numerical dissipation in smooth regions.This study presents an extension of our previous research which confirmed that adaptively adjusting parameter C can indeed overcome the inadequacy of the usage of a constant small value.Cmin is applied near a discontinuity while Cmax is used elsewhere and they are switched according to the variation of the local flow-field property.This study provides the reference values of the adaptive parameter C of WENOCU4 and systematically evaluates the comprehensive performance of three different switches(labeled as the binary,continuous,and hyperbolic tangent switches,respectively)based on an optimized efficient WENOCU4 scheme(labeled as EWENOCU4).Varieties of 1D scalar equations,empirical dispersion relation analysis,and multi-dimensional benchmark cases of Euler equations are analyzed.Generally,the dissipation and dispersion properties of these three switches are similar.Especially,employing the binary switch,EWENOCU4 achieves the best comprehensive properties.Specifically,the binary switch can efficiently filter more misidentifications in smooth regions than others do,particularly for the cases of 1 D scalar equations and Euler equations.Also,the computational efficiency of the binary switch is superior to that of the hyperbolic tangent switch.Moreover,the optimized scheme exhibits high-resolution spectral properties in the wavenumber space.Therefore,employing the binary switch is a more cost-effective improvement for schemes and is particularly suitable for the simulation of complex shock/turbulence interaction.This study provides useful guidance for the reference values of parameter C and the evaluation of adaptive switches.