Adaptive Neural Network Control for a DC Motor System with Dead-Zone

In this paper, an adaptive control strategy for tracking of a direct-current (DC) motor system with a dead-zone is developed. The main contribution of the developed scheme is that we successfully integrate an asymmetric barrier Lyapunov function approach to relax the requirements on the initial conditions. The unknown functions in the DC system are approximated by using the radial basis function neural networks (RBFNN). It is shown that the DC motor can follow a selected trajectory and all the signals are guaranteed to be bounded. Simulation results are provided to confirm the effectiveness of the proposed control.     

Direct adaptive neural control of chaos in the permanent magnet synchronous motor

In this paper, a direct adaptive neural speed tracking control is addressed for the chaotic permanent magnet synchronous motor (PMSM) drive systems via backstepping. Neural networks are directly used to approximate unknown and desired control signals and a novel direct adaptive tracking controller is constructed via backstepping. The proposed adaptive neural controllers guarantee that the tracking error converges to a small neighborhood of the origin. Compared with the conventional backstepping method, the designed neural controller??s structure is very simple. Simulation results show that the proposed control scheme can suppress the chaos of PMSM and guarantees the perfect tracking performance even with the existence of unknown parameters.     

一种风力机气动计算的全自由涡尾迹模型

采用全自由方式建立风力机尾流场的涡尾迹模型,引入“虚拟周期”的概念,并发展一种自适应松弛因子方法,从而改善了自由尾迹迭代的稳定性,提高了迭代收敛速度。利用建立的自由涡尾迹模型,计算了风力机叶片的尾流场结构、气动性能及叶片载荷,并与实验结果进行了对比分析。结果表明,尖速比越大,自适应松弛因子方法对缩小模型计算时间越有效;全自由涡尾迹模型能准确给出风力机尾流场的结构,包括尾迹的扩张以及叶尖涡和叶根涡的产生、发展和耗散的过程,风轮扭矩与实验数据吻合;叶片载荷分布的计算结果在低风速下与实验值基本一致,但是在大风速下差别较大,说明需要一个准确的失速模型。     

A fuzzy wavelet neural network-based approach to hypersonic flight vehicle direct nonaffine hybrid control

A direct nonaffine hybrid control methodology is proposed for a generic hypersonic flight models based on fuzzy wavelet neural networks (FWNNs). The addressed strategy extends the previous indirect nonaffine control approaches stemming from simplified models of affine formulations. To cope with nonaffine effects on control design, analytically invertible models are constructed and then novel hybrid controllers are developed directly using nonaffine models. Furthermore, by employing FWNNs to devise adaptive terms, inversion errors are canceled via fuzzy neural approximations. In addition, robust terms are designed to achieve larger stable region in comparison with earlier work using Lyapunov synthesis. Finally, numerical simulation results from a hypersonic flight vehicle model are given to clarify the efficiency of the proposed direct nonaffine control scheme in the presence of parametric uncertainties.     

Nonlinear adaptive neural network control for a model-scaled unmanned helicopter

In this paper, a nonlinear adaptive neural network control is proposed for trajectory tracking of a model-scaled helicopter. The purpose of this research is to reduce the ultimate bounds of tracking errors resulted from small coupling forces (or small parasitic body forces) and aerodynamic uncertainties. The proposed control is designed under backstepping framework, with neural network compensators being added. Updating laws of neural networks are designed through projection algorithm, so that adaptive parameters are bounded. Derivatives of virtual controls are obtained through command filters. It is proved that, by using neural network compensators, tracking errors of the closed-loop system can be restricted within very small ultimate bounds. Superiority of the proposed nonlinear adaptive neural network control over a backstepping control is demonstrated by simulation results.     

Adaptive neural tracking control for a class of perturbed pure-feedback nonlinear systems

This paper focuses on the problem of the adaptive neural control for a class of a perturbed pure-feedback nonlinear system. Based on radial basis function (RBF) neural networks’ universal approximation capability, an adaptive neural controller is developed via the backstepping technique. The proposed controller guarantees that all the signals in the closed-loop system are bounded and the tracking error eventually converges to a small neighborhood around the origin. The main advantage of this note lies in that a control strategy is presented for a class of pure-feedback nonlinear systems with external disturbances being bounded by functions of all state variables. A numerical example is provided to illustrate the effectiveness of the suggested approach.     

半潜式海上浮式风机气动阻尼特性研究

由于海上漂浮式风机具有较大的支撑平台运动,气动阻尼效应对海上漂浮式风机的运动响应带来了重要的影响, 日渐受到相关国内外学者的关注. 为了研究海上浮式风机的气动阻尼特性,本文推导了海上浮式风机气动阻尼力的数学模型,并借助模型实验和数值计算的方法,研究了半潜式海上浮式风机的气动阻尼特性及其作用规律. 结果表明,浮式风机的风轮旋转时的气动阻尼比风轮非旋转状态时更加明显;在作业工况下,气动阻尼对半潜式浮式风机平台的纵荡、纵摇、机舱的运动有明显的抑制作用,且主要体现为对半潜式浮式风机的平台运动固有频率响应的抑制作用,对波频范围的平台运动作用甚微. 其变化规律与风速大小、波浪载荷等有关,在风机的额定工况之前,气动阻尼通常与风速呈正相关关系,但是增长率有逐渐减小的趋势;在控制系统作用下,当入流风速接近或超过风机额定风速时,容易出现气动负阻尼现象,反而进一步强化浮式风机的运动响应,此时通过降低变桨距控制器的比例系数,即降低变桨距控制器的灵敏度,有助于增加海上浮式风机的气动阻尼效果,并且在一定程度上减缓负的气动阻尼的发生,改善海上浮式风机的运动响应.      

Adaptive fault compensation control for a class of nonlinear systems with unknown time-varying delayed faults

An adaptive approximation design for the fault compensation (FC) control is addressed for a class of nonlinear systems with unknown multiple time-delayed nonlinear faults. The magnitude and occurrence time of the multiple faults with unknown time-varying delays are unknown. The function approximation technique using neural networks is employed to adaptively approximate the unknown nonlinear effects and changes in model dynamics due to the time-delayed faults. We design an adaptive memoryless FC control system with a prescribed performance bound to compensate the faults and to guarantee the transient performance of the tracking error from unexpected changes of system dynamics. The adaptive laws for neural networks and the bound of residual approximation errors are derived using the Lyapunov stability theorem, which are used for proving that the tracking error is preserved within the prescribed performance bound regardless of unknown multiple time-delayed nonlinear faults. Simulation examples are presented for illustrating the effectiveness of the proposed control methodology     

风速不确定性对风力机气动力影响量化研究

风力机气动力学一直是国内外研究的热点课题之一.目前相关研究大都是基于确定性工况条件, 但因风力机常年工作在自然来流复杂环境,风速随机波动致使风电系统呈现不确定性, 对电网稳定性带来巨大挑战,因此进行不确定风速条件下风力机气动力学研究具有重要意义.为揭示不确定性对风力机流场影响机理并明确其对气动力的影响程度,本文提出一种风力机不确定空气动力学分析方法,基于修正叶素动量理论和非嵌入式概率配置点法,建立水平轴风力机不确定性空气动力学响应模型; 以NREL Phase VI S809风力机叶轮为研究对象, 基于该模型提取风力机输出随机响应信息,量化不确定风速对风力机风轮功率、推力、叶片挥舞弯矩和摆振弯矩的影响程度;通过分析流动诱导因子不确定性在叶片展长方向上的分布规律,揭示不确定因素在风力机本体上的传播机制,为风电系统设计及应用提供理论依据和重要参考. 结果表明,风速波动对风力机功率和气动力影响显著,高斯风速标准差由0.05倍增大至0.15倍均值,功率和推力最大波动幅度分别由13.44%和8.00%增大至35.11%和22.02%,叶片挥舞弯矩和摆振弯矩最大波动幅度分别由7.20%和12.84%增大至19.90%和33.49%.来流风速不确定性导致叶片根部位置气流明显波动,可以考虑在该部分采取流动控制措施降低叶片对风速不确定性的敏感程度.      

Validation of adaptive unstructured hexahedral mesh computations of flow around a wind turbine airfoil

In this paper we investigate local adaptive refinement of unstructured hexahedral meshes for computations of the flow around the DU91 wind turbine airfoil. This is a 25% thick airfoil, found at the mid‐span section of a wind turbine blade. Wind turbine applications typically involve unsteady flows due to changes in the angle of attack and to unsteady flow separation at high angles of attack. In order to obtain reasonably accurate results for all these conditions one should use a mesh which is refined in many regions, which is not computationally efficient. Our solution is to apply an automated mesh adaptation technique. In this paper we test an adaptive refinement strategy developed for unstructured hexahedral meshes for steady flow conditions. The automated mesh adaptation is based on local flow sensors for pressure, velocity, density or a combination of these flow variables. This way the mesh is refined only in those regions necessary for high accuracy, retaining computational efficiency. A validation study is performed for two cases: attached flow at an angle of 6° and separated flow at 12°. The results obtained using our adaptive mesh strategy are compared with experimental data and with results obtained with an equally sized non‐adapted mesh. From these computations it can be concluded that for a given computing time, adapted meshes result in solutions closer to the experimental data compared to non‐adapted meshes for attached flow. Finally, we show results for unsteady computations. Copyright © 2005 John Wiley & Sons, Ltd.     

Adaptive sliding mode control of dynamic system using RBF neural network

This paper presents a robust adaptive sliding mode control strategy using radial basis function (RBF) neural network (NN) for a class of time varying system in the presence of model uncertainties and external disturbance. Adaptive RBF neural network controller that can learn the unknown upper bound of model uncertainties and external disturbances is incorporated into the adaptive sliding mode control system in the same Lyapunov framework. The proposed adaptive sliding mode controller can on line update the estimates of system dynamics. The asymptotical stability of the closed-loop system, the convergence of the neural network weight-updating process, and the boundedness of the neural network weight estimation errors can be strictly guaranteed. Numerical simulation for a MEMS triaxial angular velocity sensor is investigated to verify the effectiveness of the proposed adaptive RBF sliding mode control scheme.     

Supercritical and subcritical dynamic flow-induced instabilities of a small-scale wind turbine blade placed in uniform flow

There is a growing interest in extracting more power per turbine by increasing the rotor size in offshore wind turbines. As a result, the turbine blades will become longer and therefore more flexible, and a flexible blade is susceptible to flow-induced instabilities. In order to design and build stable large wind turbine blades, the onset of possible flow-induced instabilities should be considered in the design process. Currently, there is a lack of experimental work on flow-induced instabilities of wind turbine blades. In the present study, a series of experiments were conducted and flow-induced instabilities were observed in wind turbine blades. A small-scale flexible blade based on the NREL 5 MW reference wind turbine blade was built using three-dimensional printing technique. The blade was placed in the test section of a wind tunnel and was subjected to uniform oncoming flow, representing the case of a parked wind turbine blade. The blade׳s tip displacement was measured using a non-contacting displacement measurement device as the oncoming wind speed was increased. At a critical wind speed, the blade became unstable and experienced limit cycle oscillations. The amplitude of these oscillations increased with increasing wind speed. Both supercritical and subcritical dynamic instabilities were observed. The instabilities were observed at different angles of attack and for blades both with and without a geometric twist. It was found that the blade twist had a significant influence on the observed instability: a blade without a twist experienced a strong subcritical instability.     

大于切出风速的大型风力机运行控制策略研究

运用非定常叶素动量(BEM)理论计算气动载荷,叠加重力载荷和惯性载荷,建立并数值求解全机动力学模型。基于快速非支配排序遗传算法(NSGA),在切出风速以上,优化得到变速变桨和定速变桨两种控制规律曲线,实现大型风力机在25m/s~40m/s风速之间正常运行的目的。比较两种控制策略的输出功率、风轮推力和转矩,得出变速变桨控制策略更适合于25m/s~40m/s之间风力机运行控制的结论。计算稳态工况时8种叶根载荷的极限值,由各载荷的变化趋势可知,Fy在25m/s之后增大9%,其他载荷均安全。     

台风过境下大型单桩式海上风机结构动力特性研究

风机大型化是我国海上风电技术发展的重要方向. 东南沿海是我国海上风电发展的重要基地, 这一区域频繁发生的台风对海上风机的影响不可忽略. 台风风场与常规大风风场有不同的湍流特性, 同时台风期间较高的风速会引起巨大的台风浪. 本文考虑台风经过期间独特的风场及波浪场, 开展风浪联合作用对大型单桩海上风机影响的研究. 基于DTU 10 MW大型单桩风机, 运用一体化分析软件SIMA建立风浪联合作用下大型单桩风机的耦合数值模型, 研究台风经过不同阶段大型风力机的动力响应特性. 计算结果显示, 叶片变桨能有效降低台风经过时风机叶片所受风载荷, 变桨状态下单桩风机所受风载荷主要来源于塔筒. 在台风经过的不同阶段, 大型单桩海上风机结构表现出不同的动力特性. 台风全过程塔筒运动均受波浪激发一阶频率控制, 塔基上方结构动力载荷以惯性载荷为主, FOVS至FEWS阶段及BOVS阶段至BEWS阶段塔筒运动一阶频率处响应能量增长较小, 响应能量向低频及波频转移. 塔基下方泥面线处剪力响应受波频控制, 弯矩响应受一阶频率控制.      

Direct adaptive neural control for strict-feedback stochastic nonlinear systems

This note considers the problem of direct adaptive neural control for a class of nonlinear single-input/single-output (SISO) strict-feedback stochastic systems. The variable separation technique is introduced to decompose the coefficient functions of the diffusion term. Radical basis function (RBF) neural networks are used to approximate unknown and desired control signals, then a novel direct adaptive neural controller is constructed via backstepping. The proposed adaptive neural controller guarantees that all the signals in the closed-loop system remain bounded in probability. A main advantage of the proposed controller is that it contains only one adaptive parameter needed to be updated online. Simulation results demonstrate the effectiveness of the proposed approach.     

Pinning synchronization of weighted complex networks with variable delays and adaptive coupling weights

In this paper, the synchronization for time-delayed complex networks with adaptive coupling weights is studied. A pinning strategy and a local adaptive scheme to determine coupling weights and feedback gains are proposed. It is noted that our control strategies only rely on some local information other than the global information of the whole network. Finally, the developed techniques are applied to two complex networks which are respectively synchronized to an unstable equilibrium point and a chaotic attractor.     

Adaptive dynamic surface neural network control for nonstrict-feedback uncertain nonlinear systems with constraints

This paper presents an adaptive dynamic surface neural network control for a class of nonstrict-feedback uncertain nonlinear systems subjected to input saturation, dead zone and output constraint. The problem of input saturation is solved by designing an anti-windup compensator, and the issue of output constraint is addressed by introducing tan-type Barrier Lyapunov function. Furthermore, based on adaptive backstepping technique, a series of novel stabilizing functions are derived. First-order sliding mode differentiator is introduced into backstepping design to obtain the first-order derivative of virtual control. The real control input is obtained using dead-zone inverse method. It is proved that the proposed control scheme can achieve finite time convergence of the output tracking error into a small neighbor of the origin and guarantee all the closed-loop signals are bounded. Simulation results demonstrate the effectiveness of the proposed control scheme.     

Discrete non-linear adaptive data driven control based upon simultaneous perturbation stochastic approximation

A novel adaptive data driven control strategy is proposed for general discrete non-linear systems. The controller is designed based upon the Simultaneous Perturbation Stochastic Approximation (SPSA) method, and is constructed through use of a Function Approximator (FA), which is fixed as a neural network here. In this novel control strategy, the parametric estimation is designed to be adaptive with convergence analysis, and the control ability has been greatly improved. The proposed control method is finally applied into the non-linear tracking problems, as well as near-optimal control problems for discrete-time non-linear systems. Simulation comparison tests were conducted on typical non-linear plants, through which, the convergence and feasibility of the proposed adaptive data driven control strategy are well demonstrated.     

LS-SVM在随机振动在线自适应逆控制中的应用

针对随机振动试验中波形再现实现中的非线性和不确定性的特点,提出了一种基于最小二乘支持向量机的自适应逆控制方法.首先根据试验样本数据利用模糊贝叶斯推断确定最小二乘支持向量机的参数,然后给出了基于最小二乘支持向量机的自适应逆控制器的设计方法,最后给出了随机振动在线自适应逆控制结构.实验结果表明该方法的有效性.     

Property and evolution of the running-in attractor in an actual dynamic system

In this paper, the issue of adaptive neural tracking control for uncertain switched multi-input multi-output (MIMO) nonstrict-feedback nonlinear systems with average dwell time is studied. The system under consideration includes unknown dead-zone inputs and output constraints. The uncertain nonlinear functions are identified via neural networks. Also, neural networks-based switched observer is constructed to approximate all unmeasurable states. By means of the information for dead-zone slopes and barrier Lyapunov function (BLF), the problems of dead-zone inputs and output constraints are tackled. Furthermore, dynamic surface control (DSC) scheme is employed to ensure that the computation burden is greatly reduced. Then, an observer-based adaptive neural control strategy is developed on the basis of backstepping technique and multiple Lyapunov functions approach. Under the designed controller, all the signals existing in switched closed-loop system are bounded, and system outputs can track the target trajectories within small bounded errors. Finally, the feasibility of the presented control algorithm is proved via simulation results.