Particle swarm optimization based nonlinear active noise control under saturation nonlinearity

In this paper, a method is proposed to overcome the saturation non-linearity linked to the microphones and loudspeakers of active noise control (ANC) system. The reference microphone gets saturated when the acoustic noise at the source increases beyond the dynamic limits of the microphone. When the controller tries to drive the loudspeaker system beyond its dynamic limits, the saturation nonlinearity is also introduced into the system. The secondary path which is generally estimated with a low level auxiliary noise by a linear transfer function does not model such saturation nonlinearity. Therefore, the filtered-x least mean square (FXLMS) algorithm fails to perform when the noise level is increased. For alleviating the saturation nonlinearity effect a nonlinear functional expansion based ANC algorithm is proposed where the particle swarm optimization (PSO) algorithm is suitably applied to tune the parameters of a filter bank based functional link artificial neural network (FLANN) structure, named as PSO based nonlinear structure (PSO-NLS) algorithm. The proposed algorithm does not require any computation of secondary path estimate filtering unlike other conventional gradient based algorithms and hence has got computational advantage. The computer simulation experiments show its superior performance compared to the FXLMS, filtered-s LMS and genetic algorithms under saturation present at both at secondary and reference paths. The paper also includes a sensitivity analysis to study the effect of different parameters on ANC performance.     

New recursive-least-squares algorithms for nonlinear active controlof sound and vibration using neural networks

In recent years, a few articles describing the use of neural networks for nonlinear active control of sound and vibration were published. Using a control structure with two multilayer feedforward neural networks (one as a nonlinear controller and one as a nonlinear plant model), steepest descent algorithms based on two distinct gradient approaches were introduced for the training of the controller network. The two gradient approaches were sometimes called the filtered-x approach and the adjoint approach. Some recursive-least-squares algorithms were also introduced, using the adjoint approach. In this paper, an heuristic procedure is introduced for the development of recursive-least-squares algorithms based on the filtered-x and the adjoint gradient approaches. This leads to the development of new recursive-least-squares algorithms for the training of the controller neural network in the two networks structure. These new algorithms produce a better convergence performance than previously published algorithms. Differences in the performance of algorithms using the filtered-x and the adjoint gradient approaches are discussed in the paper. The computational load of the algorithms discussed in the paper is evaluated for multichannel systems of nonlinear active control. Simulation results are presented to compare the convergence performance of the algorithms, showing the convergence gain provided by the new algorithms.     

Fast Affine Projection Algorithms for Filtered-x Multichannel Active Noise Control

In recent years, affine projection algorithms have been proposed for adaptive system applications as an efficient alternative to the slow convergence speed of least mean square (LMS)-type algorithms. Whereas much attention has been focused on the development of efficient versions of affine projection algorithms for echo cancellation applications, the similar adaptive problem presented by active noise control (ANC) systems has not been studied so deeply. This paper is focused on the necessity to reduce even more the computational complexity of affine projection algorithms for real-time ANC applications. We present some alternative efficient versions of existing affine projection algorithms that do not significantly degrade performance in practice. Furthermore, while in the ANC context the commonly used affine projection algorithm is based on the modified filtered-x structure, an efficient affine projection algorithm based on the (nonmodified) conventional filtered-x structure, as well as efficient methods to reduce its computational burden, are discussed throughout this paper. Although the modified filtered-x scheme exhibits better convergence speed than the conventional filtered-x structure and allows recovery of all the signals needed in the affine projection algorithm for ANC, the conventional filtered-x scheme provides a significant computational saving, avoiding the additional filtering needed by the modified filtered-x structure. In this paper, it is shown that the proposed efficient versions of affine projection algorithms based on the conventional filtered-x structure show good performance, comparable to the performance exhibited by the efficient approaches of modified filtered-x affine projection algorithms, and also achieve meaningful computational savings. Experimental results are presented to validate the use of the algorithms introduced in the paper for practical applications.      

一种非线性PID控制算法的仿真研究

研究PID控制系统优化问题,工业控制被控对象均具有非线性、时变和大时滞性,引起系统的品质性能差,传统的线性控制难以达到所要求精度。为了提高系统控制精度,利用PID控制器各增益参数与偏差信号间的非线性关系,提出一种非线性PID控制算法。首先将PID参数转化为优化问题,然后采用粒子群算法的全局、并行搜索能力对非线性控制参数进行求解,得到一组最优的PID控制参数。仿真结果表明,相对于传统线性PID控制,非线性PID控制器超调小,调节时间短,并提高了控制精度,有效解决了传统PID难以准确控制非线性对象的难题。     

Functional link artificial neural network applied to active noise control of a mixture of tonal and chaotic noise

Many practical noises emanating from rotating machines with blades generate a mixture of tonal and the chaotic noise. The tonal component is related to the rotational speed of the machine and the chaotic component is related to the interaction of the blades with air. An active noise controller (ANC) with either linear algorithm like filtered-X least mean square (FXLMS) or nonlinear control algorithm like functional link artificial neural network (FLANN) or Volterra filtered-X LMS (VFXLMS) algorithm shows sub-optimal performance when the complete noise is used as reference signal to a single controller. However, if the tonal and the chaotic noise components are separated and separately sent to individual controller with tonal to a linear controller and chaotic to a nonlinear controller, the noise canceling performance is improved. This type of controller is termed as hybrid controller. In this paper, the separation of tonal and the chaotic signal is done by an adaptive waveform synthesis method and the antinoise of tonal component is produced by another waveform synthesizer. The adaptively separated chaotic signal is fed to a nonlinear controller using FLANN or Volterra filter to generate the antinoise of the chaotic part of the noise. Since chaotic noise is a nonlinear deterministic noise, the proposed hybrid algorithm with FLANN based controller shows better performance compared to the recently proposed linear hybrid controller. A number of computer simulation results with single and multitone frequencies and different types of chaotic noise such as logistic and Henon map are presented in the paper. The proposed FLANN based hybrid algorithm was shown to be performing the best among many previously proposed algorithms for all these noise cases including recorded noise signal.     

基于多模型的非线性系统自适应最小方差控制

对于一类典型的离散时间非线性系统, 提出了一种基于多模型的自适应最小方差控制方法. 通过在平衡点附近建立线性模型, 用径向基函数神经元网络来补偿建模误差和未建模动态, 形成了非线性系统的多模型表示. 采用了具有积分性质的切换指标函数作为切换法则和最小方差的控制方法构成了多模型自适应控制器. 仿真实验的结果表明了这种方法的有效性.     

A Study on Kinematics and Workspace Determination of a General 6-<Emphasis Type="Underline">P</Emphasis> US Robot

This paper proposes the novel adaptive neural network (ADNN) compliant force/position control algorithm applied to a highly nonlinear serial pneumatic artificial muscle (PAM) robot as to improve its compliant force/position output performance. Based on the new adaptive neural ADNN model which is dynamically identified to adapt well all nonlinear features of the 2-axes serial PAM robot, a new hybrid adaptive neural ADNN-PID controller was initiatively implemented for compliant force/position controlling the serial PAM robot system used as an elbow and wrist rehabilitation robot which is subjected to not only the internal coupled-effects interactions but also the external end-effecter contact force variations (from 10[N] up to critical value 30[N]). The experiment results have proved the feasibility of the new control approach compared with the optimal PID control approach. The novel proposed hybrid adaptive neural ADNN-PID compliant force/position controller successfully guides the upper limb of subject to follow the linear and circular trajectories under different variable end-effecter contact force levels.     

Synthesis of the sliding-mode neural network controller for unknown nonlinear discrete-time systems

This paper develops a sliding-mode neural network controller for a class of unknown nonlinear discrete-time systems using a recurrent neural network (RNN). The control scheme is based on a linearized expression of the nonlinear system using a linear neural network (LNN). The control law is proposed according to the discrete L yapunov theory. With a modified real-time recurrent learning algorithm, the RNN as an estimator is used to estimate the unknown part in the control law in on-line fashion. The stability of the control system is guaranteed owing to the on-line learning ability of the RNN algorithm. The proposed control scheme is applied to numerical problems and simulation results that it is very effective.     

Adaptive neural sliding mode compensator for a class of nonlinear systems with unmodeled uncertainties

This paper addresses the problem of adaptive neural sliding mode control for a class of multi-input multi-output nonlinear system. The control strategy is an inverse nonlinear controller combined with an adaptive neural network with sliding mode control using an on-line learning algorithm. The adaptive neural network with sliding mode control acts as a compensator for a conventional inverse controller in order to improve the control performance when the system is affected by variations in its entire structure (kinematics and dynamics). The controllers are obtained by using Lyapunov's stability theory. Experimental results of a case study show that the proposed method is effective in controlling dynamic systems with unexpected large uncertainties.     

基于动态递归神经网络的自适应PID控制

提出一种基于动态递归神经网络的自适应PID控制方案，该控制系统由神经网络辨识器和神经网络控制器组成。辨识器采用单隐层的动态递归神经网络，网络结构为2-4-1；辨识算法为动态BP算法；控制器采用两层线性结构的神经网络，输入为系统偏差及其一阶、二阶微分，因此具有增量型PID控制结构。应用该控制系统对一非线性时变系统进行仿真研究，仿真结果表明该控制方案不仅具有良好的跟踪特性，而且对系统参数变化具有较强的鲁棒性。