A dual filtering scheme for nonlinear active noise control

Active noise control problems are often affected by nonlinear effects such as distortion and saturation of measurement and actuation devices, which call for suitable nonlinear models and algorithms. The active noise control problem can be interpreted as an indirect model identification problem, due to the secondary path dynamics that follow the control filter block. This complicates the weight update mechanism in the nonlinear case, in that the error gradient depends on the secondary path gradient through nonlinear recursions. A simpler and computationally less demanding approach is here proposed that employs the updating scheme of the standard filtered‐x least mean squares (LMS) or filtered‐u LMS algorithm. As in those schemes, the calculation of the error gradient requires a signal filtering through an auxiliary system, here obtained through a secondary adaptation loop. The resulting dual filtering LMS algorithm performs the adaptation of the controller parameters in a direct identification mode and can therefore be easily coupled with adaptive model structure selection schemes to provide online tuning of the model structure, for improved model robustness. Copyright © 2013 John Wiley & Sons, Ltd.     

Nonlinear FXLMS algorithm for active noise control systems with saturation nonlinearity

In active noise control (ANC) applications, the saturation effect of the loudspeaker in the secondary path is considered as the most serious problem that could degrade performance of standard filtered‐x least mean square (FXLMS) control algorithm. When the loudspeaker exhibits nonlinearities, the linear modeling approach fails to identify the secondary path accurately. In the literature, the nonlinear FXLMS (NLFXLMS) algorithm has been proposed to update the ANC controller with a block‐oriented secondary path model. This model consists of nonlinear and linear filters whereby the nonlinear part which represents the saturation effect of the amplifier‐loudspeaker system is modeled by a scaled error function (SEF). The NLFXLMS algorithm requires an exact copy of the linear and nonlinear models of the secondary path. However, NLFXLMS cannot be implemented in real time because the modeling of the SEF cannot be realized. In this paper, a new method to model the secondary path using the Hammerstein model structure and tangential hyperbolic function (THF) is proposed. The THF can represent the SEF to a certain degree of accuracy. Furthermore, the modeling of the THF can be realized using least mean square (LMS) algorithm and utilized in the NLFXLMS control scheme. Simulation results show that the performance of the THF‐based NLFXLMS algorithm is comparable with the SEF‐based NLFXLMS. © 2012 Institute of Electrical Engineers of Japan. Published by John Wiley & Sons, Inc.     

二级倒立摆的递阶模糊神经网络控制

为了表明模糊神经网络控制器比较适合于控制快速、多变量、强非线性、绝对不稳定系统，可以克服用模糊神经网络控制多变量系统时的规则组合爆炸问题。本文提出用递阶模糊神经网络控制二级倒立摆。这种方法可以有效地减少多变量输入的模糊神经网络控制器的规则数，有利于利用专家的控制经验初始化网络参数，从而有利于下一步利用遗传算法对其进行优化。实验结果表明：与线性最优控制相比，本文方法的控制效果好、鲁棒性强。     

Compact self‐constructing recurrent fuzzy neural network with decision feedback for quadrature amplitude modulation signaling systems

This paper proposes a novel adaptive decision feedback equalizer (DFE) based on compact self‐constructing recurrent fuzzy neural network (CSRFNN) for quadrature amplitude modulation systems. Without the prior knowledge of channel characteristics, a novel training scheme containing both compact self‐constructing learning (CSL) and real‐time recurrent learning algorithms is derived for the CSRFNN. The proposed CSL algorithm adopts two evaluation criteria to intelligently decide the number of fuzzy rules that are necessary. The real‐time recurrent learning is performed simultaneously with the CSL at each time instant to adjust DFE parameters. The proposed DFE is compared with several neural network‐based DFEs on a nonlinear complex‐valued channel. The results show that the CSRFNN DFE is superior to classical neural network DFEs in terms of symbol‐error rate, convergence speed, and time cost. Copyright © 2011 John Wiley & Sons, Ltd.     

基于RBF神经网络的无刷直流电动机速度无关控制法

为提高无刷直流电动机无位置传感器控制精度,提出了一种基于RBF神经网络的无刷直流电动机速度无关控制新策略。该策略主要包含两个部分:一方面,利用RBF神经网络的自适应、非线性控制等优良性能,结合电机运行状态,修正神经元之间的连接权值,从而可以克服由于无刷直流电动机系统的非线性和部分参数不确定性造成精度下降的负面影响。另一方面,神经网络的输出经过滤波处理后,采用速度无关位置函数法(函数法)输出电机换相信号。该方法在转子转速由近零到高速变化的过程中,都能够对转子位置进行检测并给出换相时刻。仿真和实验表明,该策略具有优良的控制性能。     

A novel fuzzy neural network based distance relaying scheme

This paper presents a new approach to distance relaying using fuzzy neural network (FNM). The FNN can be viewed either as a fuzzy system, a neural network or fuzzy neural network. The structure is seen as a neural network for training and a fuzzy viewpoint is utilized to gain insight into the system and to simplify the model. The number of rules is determined by the data itself and therefore a smaller number of rules is produced. The network is trained with the backpropagation algorithm. A pruning strategy is applied to eliminate the redundant rules and fuzzification neurons, consequently a compact structure is achieved. The classification and location tasks are accomplished by using different FNN's. Once the fault type is identified by the FNN classifier the selected fault locating FNN estimates the location of the fault accurately. Normalized peaks of fundamental voltage and current waveforms are considered as inputs to all the networks and an additional input derived from the DC component is fed to fault locating networks. The peaks and DC component are extracted from sampled signals by the EKF. Test results show that the new approach provides robust and accurate classification/location of faults for a variety of power system operating conditions even with resistance in the fault path     

Adaptive TS‐FNN control for a class of uncertain multi‐time‐delay systems: The exponentially stable sliding mode‐based approach

This paper presents an adaptive Takagi–Sugeno fuzzy neural network (TS‐FNN) control for a class of multiple time‐delay uncertain nonlinear systems. First, we develop a sliding surface guaranteed to achieve exponential stability while considering mismatched uncertainty and unknown delays. This exponential stability result based on a novel Lyapunov–Krasovskii method is an improvement when compared with traditional schemes where only asymptotic stability is achieved. The stability analysis is transformed into a linear matrix inequalities problem independent of time delays. Then, a sliding mode control‐based TS‐FNN control scheme is proposed to achieve asymptotic stability for the controlled system. Since the TS‐FNN combines TS fuzzy rules and a neural network structure, fewer numbers of fuzzy rules and tuning parameters are used compared with the traditional pure TS fuzzy approach. Moreover, all the fuzzy membership functions are tuned on‐line even in the presence of input uncertainty. Finally, simulation results show the control performance of the proposed scheme. Copyright © 2008 John Wiley & Sons, Ltd.     

Modeling of non‐linear systems by FWNNs and their intelligent control

By utilizing some of the important properties of wavelets like denoising, compression, multiresolution along with the concepts of fuzzy logic and neural network, two fuzzy wavelet neural networks (FWNNs) are proposed for approximating any arbitrary non‐linear function, hence, identifying a non‐linear system. We have fuzzified the output of DWT block, which receives the given inputs, in the proposed two methods: one using compression property and other using multiresolution property. We present a new type of fuzzy neuron model, each non‐linear synapse of which is characterized by a set of fuzzy implication rules with singleton weights in their consequents. It is shown that noise and disturbance in the reference signal are reduced with wavelets and also the variation of somatic gain, the parameter that controls the slope of the activation function in the neural network, leads to more accurate output. Identification results are found to be accurate and speed of their convergence is fast. Next, we simulate a control system for keeping output at a desired level by using the identified models. Two self‐learning controllers are designed in this simulation. One is a self‐learning fuzzy PI controller and other is a NN controller. Simulation results show that the NN controller is more adaptive and fast. Copyright © 2005 John Wiley & Sons, Ltd.     

电液位置伺服系统的模糊神经网络控制

针对电液位置伺服系统中的非线性、参数时变性等复杂因素,设计了一种模糊神经网络控制方案。由于常规的模糊神经网络学习算法具有权值调整复杂,收敛速度慢的缺点,因此采用模糊逐级误差逼近方法来调整模糊神经网络的权值。该算法易于实现,网络权值在线学习速度较快,而且计算量小于其他的常规神经网络学习算法。将该方法应用于电液位置伺服控制系统中,在对象参数摄动情况下,进行了仿真研究。仿真结果表明,采用该方法所设计的控制器满足系统对快速性和稳态精确度的要求,系统的鲁棒性增强,验证了方法的有效性。     

在噪声抵消应用中自适应滤波算法性能的仿真比较

介绍了噪声抵消的原理和从强噪声背景中自适应滤波提取有用信号的方法。利用模糊逻辑和RBF神经网络的等价性将模糊逻辑和神经网络有机的结合来构成模糊神经网络,并对BP神经网络、RBF神经网络和模糊神经网络三种基本自适应算法进行了对比研究。计算机模拟仿真结果表明,这几种算法都能通过有效抑制各种干扰来提高强噪声背景中的信号检测特性。相比之下,模糊神经网络算法具有良好的收敛性能,除收敛速度快于BP神经网络算法和RBF神经网络算法以及稳定性强外,而且具有更高的起始收敛速率,更小的权噪声,更大的抑噪能力。     

混合动力系统电能质量扰动分析及治理

基于并联有源滤波器(shunt active power filter,SAPF)和动态电压调节器(dynamic voltage restorer,DVR),首先,搭建了包含光伏和风机的混合动力系统,用以模拟分布式电源接入配电网中产生的电能质量(power quality,PQ)扰动。其次,利用模糊逻辑、神经网络和自适应神经模糊推理系统控制算法对SAPF的动态性能进行优化,对电能质量扰动进行治理,使用人工智能技术进行管理,使光伏和风能系统均实现最大功率点跟踪(maximum power point tracking, MPPT)。最后,在搭建的仿真系统中进行验证,线性负载和非线性负载输出侧谐波畸变率分别降至0.20%和2.05%,满足配电系统对于电能质量的要求。     

模糊神经网络实现的PID参数自调整控制及应用研究

模糊控制学习功能差,神经网络推理功能差,把二者结合可以起到互补作用,保证推理和学习功能的实现。本文用多层ＢＰ神经网络实现了一类模糊推理过程,该网络通过学习记忆ＰＩＤ参数调整的基本规则,实现了ＰＩＤ控制器参数的在线调整。通过对考虑非线性因素的伺服系统控制对象的仿真,说明了该方法的有效性。     

Adaptive neural network command filtered backstepping control of pure‐feedback systems in presence of full state constraints

An adaptive neural network (NN) command filtered backstepping control is proposed for the pure‐feedback system subjected to time‐varying output/stated constraints. By introducing a one‐to‐one nonlinear mapping, the obstacle caused by full stated constraints is conquered. The adaptive control law is constructed by command filtered backstepping technology and radial basis function NNs, where only one learning parameter needs to be updated online. The stability analysis via nonlinear small‐gain theorem shows that all the signals in closed‐loop system are semiglobal uniformly ultimately bounded. The simulation examples demonstrate the effectiveness of the proposed control scheme.     

模糊神经网络在噪声消除中的应用

提出了一种基于改进模糊聚类算法的训练模糊神经网络的算法,该方法采用遗传算法改进传统的模糊聚类算法,并给出了一个衡量聚类有效性的函数以确定聚类算法中的聚类总数,从而确定模糊神经网络结构,仿结果表明神经网络可成功的应用于噪声消除。     

基于模糊神经网络的单相自适应重合闸的研究

将模糊神经网络应用于单相自适应重合闸中,以模糊理论和人工神经网络理论为基础构造了一个多输入模糊神经网络,用于识别瞬时性故障与永久性故障.该网络以取大取小运算部分代替了乘积求和运算,并采用了从样本中获取模糊规则的方法.利用Matlab进行了大量仿真实验,验证了该方法的可行性与准确性;在仿真的基础上,将多输入模糊神经网络与BP神经网络进行了比较,证明了多输入模糊神经网络在单相自适应重合闸中应用的优越性.     

基于自适应RBF神经网络的变压器噪声有源控制算法

针对现有变压器噪声有源控制算法存在的不足,提出了一种用于抑制噪声的新算法。该算法融合了自适应算法、粒子群算法、改进梯度下降算法及RBF神经网络算法。首先利用自适应算法确定降噪系统控制器中RBF神经网络隐含层节点个数和相应的参数;然后,根据切换策略自适应地选择粒子群算法或者改进梯度下降算法,用来优化节点数目和参数;最后,将优化得到的隐含层结构和参数反馈至系统控制器中,使系统的次级声源更好地抵消源声源。通过将所提的改进RBF神经网络法与未改进的RBF神经网络法和BP神经网络法进行比较,表明该算法可有效地提高降噪系统的自适应能力和抗干扰能力,且能够将噪声控制在较低的范围内,获得较理想的降噪效果。     

基于模糊神经网络的锥形滚柱式磁悬浮轴承控制的研究

介绍了锥形滚柱式磁悬浮轴承的实验系统并利用线性控制原理对其建立了数学模型。应用神经网络和模糊控制理论设计了非线性控制器。从线性最优控制器设计值中提取经验规则，并将模糊化后的变量引入神经网络结构，然后采用变尺度算法对网络参数进行优化，解模糊后得到反馈增益矩阵。仿真结果表明控制器对静态和暂态稳定均有良好效果。     

改进蝗虫优化算法 在模糊神经网络PID控制中的研究

传统模糊神经PID控制算法易出现网络参数调整不合适导致控制效果差的问题。本文提出一种改进蝗虫算法优化下的模糊神经网络PID控制算法。首先针对传统蝗虫算法粒子多样性不足的问题引入Levy随机飞行策略,其次引入非线性缩减因子和模拟退火算法来改善算法寻优能力以及跳出局部最优解的能力,然后将改进的蝗虫算法与模糊神经PID结合来优化神经网络超参数以及实现控制参数自整定,最后由仿真结果验证所提出的改进蝗虫算法优化模糊神经网络PID算法的优越性和可靠性。     

过热汽温模糊神经网络预测控制器的设计

针对锅炉过热汽温的特点，设计前馈—反馈串级复合型控制系统。主控制器采用基于神经网络预测模型的模糊神经控制，即该控制器首先是将神经网络与预测控制相结合，采用改进的递阶遗传算法对神经网络的权值和结构同时进行训练，实现了非线性、大时滞系统模型的精确预测；然后将模糊控制与神经网络相结合，实现模糊神经预测控制。副控制器采用二自由度PID控制器。仿真结果表明，该控制显著提高锅炉过热汽温这一非线性、大时滞系统的控制品质，且易于工程实现。     

基于自适应网络模糊推理系统的开关磁阻电机建模方法

提出一种开关磁阻电机(switched reluctance motor,SRM)数学建模的新方法：在已知开关磁阻电机静态电感曲线和矩角特性曲线的基础上,将自适应网络模糊推理系统(adaptive network based fuzzy inference system,ANFIS)用于SRM的整体建模中。该模糊推理系统由5层网络构成,将模糊推理与神经网络有机结合起来,利用它的自学习功能计算出模糊系统的隶属度函数以及相应的模糊规则,形成一个结构简单、紧凑的网络来实现电机绕组电流、转子位置角与电感和转矩的非线性映射关系,然后离线训练得到电感与转矩模型。把这种基于ANFIS的电感和矩角模型应用于SRM的系统建模中,以550 W、6/4极SRM为例,进行了仿真与实验比较,结果表明此建模方法能够较好的反映SRM的实际工作状况,从而为SRM系统的建模分析与设计提供一种新的有力的工具。