Neural Network Based Adaptive Tracking Control for a Class of Pure Feedback Nonlinear Systems With Input Saturation

In this paper, an adaptive neural networks(NNs)tracking controller is proposed for a class of single-input/singleoutput(SISO) non-affine pure-feedback non-linear systems with input saturation. In the proposed approach, the original input saturated nonlinear system is augmented by a low pass filter.Then, new system states are introduced to implement states transformation of the augmented model. The resulting new model in affine Brunovsky form permits direct and simpler controller design by avoiding back-stepping technique and its complexity growing as done in existing methods in the literature.In controller design of the proposed approach, a state observer,based on the strictly positive real(SPR) theory, is introduced and designed to estimate the new system states, and only two neural networks are used to approximate the uncertain nonlinearities and compensate for the saturation nonlinearity of actuator. The proposed approach can not only provide a simple and effective way for construction of the controller in adaptive neural networks control of non-affine systems with input saturation, but also guarantee the tracking performance and the boundedness of all the signals in the closed-loop system. The stability of the control system is investigated by using the Lyapunov theory. Simulation examples are presented to show the effectiveness of the proposed controller.

Nonlinear Systems Identification via an Input-Output Model Based on a Feedforward Neural Network

This paper develops a feedforward neural network based input output model for a general unknown nonlinear dynamic system identification when only the inputs and outputs are accessible observations. In the developed model, the size of the input space is directly related to the system order. By monitoring the identification error characteristic curve, we are able to determine the system order and subsequently an appropriate network structure for systems identification. Simulation results are promising and show that generic nonlinear systems can be identified, different cases of the same system can also be discriminated by our model.

Dimensionality Reduction with Input Training Neural Network and Its Application in Chemical Process Modelling

Many applications of principal component analysis (PCA) can be found in dimensionality reduction. But linear PCA method is not well suitable for nonlinear chemical processes. A new PCA method based on im-proved input training neural network (IT-NN) is proposed for the nonlinear system modelling in this paper. Mo-mentum factor and adaptive learning rate are introduced into learning algorithm to improve the training speed of IT-NN. Contrasting to the auto-associative neural network (ANN), IT-NN has less hidden layers and higher training speed. The effectiveness is illustrated through a comparison of IT-NN with linear PCA and ANN with experiments. Moreover, the IT-NN is combined with RBF neural network (RBF-NN) to model the yields of ethylene and propyl-ene in the naphtha pyrolysis system. From the illustrative example and practical application, IT-NN combined with RBF-NN is an effective method of nonlinear chemical process modelling.

Quantifying the thermal damping effect in underground vertical shafts using the nonlinear autoregressive with external input(NARX) algorithm

As air descends the intake shaft, its infrastructure, lining and the strata will emit heat during the night when the intake air is cool and, on the contrary, will absorb heat during the day when the temperature of the air becomes greater than that of the strata. This cyclic phenomenon, also known as the "thermal damping effect" will continue throughout the year reducing the effect of surface air temperature variation. The objective of this paper is to quantify the thermal damping effect in vertical underground airways. A nonlinear autoregressive time series with external input(NARX) algorithm was used as a novel method to predict the dry-bulb temperature(Td) at the bottom of intake shafts as a function of surface air temperature. Analyses demonstrated that the artificial neural network(ANN) model could accurately predict the temperature at the bottom of a shaft. Furthermore, an attempt was made to quantify typical "damping coefficient" for both production and ventilation shafts through simple linear regression models. Comparisons between the collected climatic data and the regression-based predictions show that a simple linear regression model provides an acceptable accuracy when predicting the Tdat the bottom of intake shafts.

Adaptive Backstepping Sliding Mode Control for Nonlinear Systems with Input Saturation

An adaptive backstepping sliding mode control is proposed for a class of uncertain nonlinear systems with input saturation.A command filtered approach is used to prevent input saturation from destroying the adaptive capabilities of neural networks (NNs).The control law and adaptive updating laws of NNs are derived in the sense of Lyapunov function,so the stability can be guaranteed even under the input saturation.The proposed control law is robust against the disturbance,and it can also eliminate the impact of input saturation.Simulation results indicate that the proposed controller has a good performance.

Optimal Neuro-Control Strategy for Nonlinear Systems With Asymmetric Input Constraints

In this paper,we present an optimal neuro-control scheme for continuous-time(CT)nonlinear systems with asymmetric input constraints.Initially,we introduce a discounted cost function for the CT nonlinear systems in order to handle the asymmetric input constraints.Then,we develop a Hamilton-Jacobi-Bellman equation(HJBE),which arises in the discounted cost optimal control problem.To obtain the optimal neurocontroller,we utilize a critic neural network(CNN)to solve the HJBE under the framework of reinforcement learning.The CNN's weight vector is tuned via the gradient descent approach.Based on the Lyapunov method,we prove that uniform ultimate boundedness of the CNN's weight vector and the closed-loop system is guaranteed.Finally,we verify the effectiveness of the present optimal neuro-control strategy through performing simulations of two examples.

Terminal Angular Constraint Integrated Guidance and Control for Flexible Hypersonic Vehicle with Dead-Zone Input Nonlinearity

This paper presents an integrated guidance and control model for a flexible hypersonic vehicle with terminal angular constraints.The integrated guidance and control model is bounded and the dead-zone input nonlinearity is considered in the system dynamics.The line of sight angle,line of sight angle rate,attack angle and pitch rate are involved in the integrated guidance and control system.The controller is designed with a backstepping method,in which a first order filter is employed to avoid the differential explosion.The full tuned radial basis function(RBF)neural network(NN)is used to approximate the system dynamics with robust item coping with the reconstruction errors,the exactitude model requirement is reduced in the controller design.In the last step of backstepping method design,the adaptive control with Nussbaum function is used for the unknown dynamics with a time-varying control gain function.The uniform ultimate boundedness stability of the control system is proved.The simulation results validate the effectiveness of the controller design.

输入饱和约束下自适应RBF神经网络非线性反馈船舶航向控制

针对输入饱和约束下外界扰动和模型不确定情况下的船舶航向跟踪控制问题,提出一种自适应径向基函数(radial basis function,RBF)神经网络非线性反馈航向跟踪控制方法。利用自适应RBF神经网络对外界扰动和模型不确定项进行估计,并利用最小学习参数法减少计算量;将一个具有误差增益反相关特征的非线性函数嵌入控制律中,设计一种非线性反馈控制方法;利用李雅普诺夫理论证明所有信号在考虑外界扰动和模型不确定的船舶航向跟踪控制系统中都是一致有界的。通过仿真和比较,验证了所设计控制方法的有效性。所做研究可为输入饱和约束下船舶航向跟踪控制提供参考,具有工程实际意义。

Comparative analysis of time series neural network methods for three-way catalyst modeling

Relative Oxygen Level of the Three-Way Catalyst is an important parameter that affects the conversion efficiency of pollutants. ROL is a time-varying hidden state variable that is difficult to directly observe in practice. Therefore, it is common to use a method of clearing oxygen storage to simplify control in vehicles. However, this method negates the positive effects of ROL on pollutant treatment. ROL can be indirectly observed through modeling methods. Chemical modeling methods involve extensive computational requirements that cannot meet the demands of practical control. In contrast, time-series neural networks offer computational speed advantages when dealing with similar problems. Therefore, the ROL observation models using both NARX and LSTM neural networks are developed and compared in this study. The results indicate that the NARX neural network exhibits higher precision with a smaller number of neurons and time steps. The LSTM neural network demonstrates greater stability when dealing with data error fluctuations. In practical applications, the ROL model can monitor the TWC operating status and assist in the development of intelligent pollutant aftertreatment control strategies.

基于PSO-NARX网络的司机驾驶行为分析方法

舒适性、准时性、节能性等是衡量高速铁路自动驾驶水平的重要指标,通过不断学习优秀司机的驾驶行为,可以优化列车自动驾驶性能,促进高速铁路自动驾驶技术的发展。基于现场列车运行数据,提出一种带有外部输入的非线性自回归(NARX)网络的列车司机驾驶行为分析方法。该方法构建了具有时序特征的NARX网络模型,并选取多项影响司机决策的参数作为输入,利用粒子群优化算法(PSO)确定网络的权重和阈值,对下一时刻列车运行情况进行预测。仿真结果表明:本文提出的PSO-NARX网络分析模型的预测效果优于前馈型神经网络(BP)、PSO-BP、NARX,相比于BP算法,迭代步数降低了373步,误差降低了8382%,相关系数达到了90117%。通过此预测,可以优化列车的自动驾驶设备性能指标,保障列车准时的同时,提高了乘客乘坐的舒适性。

负荷数据特征分析的用户集群需求响应潜力预测方法

随着电力市场改革的逐步推进,需求响应将在未来新型电力系统中发挥越来越重要作用。针对目前DR潜力计算过程繁琐、用户数据不足的问题,提出了一种基于用户历史负荷、气温和电价数据的用户集群DR潜力预测方法。首先,通过对用户的历史负荷曲线进行数据处理和信息提取,从月负荷规律性、日负荷波动性、峰谷一致性3个方面对各用户的用电行为进行特征值计算,形成评估用户类型的指标体系。继而,提出基于时序带有外部输入的非线性自回归神经网络的用户负荷和DR潜力预测方法。最后,以工业用户为例采用Meanshift算法实现用户集群划分,并对通用零部件制造行业的DR调节功率进行预测,经与实际调节功率数据进行对比分析,验证了本文所提方法的有效性。

Adaptive Tracking Control for Output-Constrained Switched MIMO Pure-Feedback Nonlinear Systems with Input Saturation

In this paper,an adaptive neural tracking control scheme for a class of uncertain switched multi-input multi-output(MIMO)pure-feedback nonlinear systems is proposed.The considered MIMO pure-feedback nonlinear system contains input and output constraints,completely unknown nonlinear functions and time-varying external disturbances.The unknown nonlinear functions representing system uncertainties are identified via radial basis function neural networks(RBFNNs).Then,the Nussbaum function is utilized to deal with the nonlinearity issue caused by the input saturation.To prevent system outputs from violating prescribed constraints,the barrier Lyapunov functions(BLFs)are introduced.Also,a switched disturbance observer is designed to make the time-varying external disturbances estimable.Based on the backstepping recursive design technique and the Lyapunov stability theory,the developed control method is verified applicable to ensure the boundedness of all signals in the closed-loop system and make the system output track given reference signals well.Finally,a numerical simulation is given to demonstrate the effectiveness of the proposed control method.

一类带有输入时滞的非线性关联大系统的分散自适应控制

为解决带有完全未知关联项和输入时滞的非线性关联大系统的跟踪控制问题,本文提出一种自适应分散控制策略,利用神经网络逼近理论处理未知项,同时引入合适的辅助系统消除输入时滞。为了消除未知关联项对系统的影响,定义新的未知常数和非线性光滑函数来界定未知关联项。根据Lyapunov稳定性理论,本文所提出的控制方法能够保证闭环系统一致最终有界,并且跟踪误差能够收敛到原点的一个小邻域内。最后通过仿真算例进一步检验所设计方法的可行性。

具有状态约束和输入非线性的PMSLM自适应神经网络控制

针对永磁同步直线电机系统中存在的模型不确定性、状态约束以及输入非线性(如非线性电磁驱动力/输入受限)问题,提出了一种基于神经网络的自适应控制器.具体来说,为了降低噪声敏感性,进一步提高跟踪精度,采用仅依赖于参考轨迹的期望信号来替代测量信号.然后,设计神经网络在线逼近未知模型和非线性函数,并通过构造连续控制的方法来处理逼近误差.此外,构造障碍李雅普诺夫函数确保系统在运行过程中状态始终满足约束条件;并且通过严格的理论分析证明了跟踪性能满足要求.最后,通过仿真实验验证了所提控制器的有效性和鲁棒性.

热弹性效应分析与机床进给系统热动态特性建模

通过一维杆的一维传热的分组显式数值求解,分析热弹性效应的存在及规律,得出随着时间的增长,温升—热变形之间的关系会逐渐趋近稳态,但不可能获得绝对的稳态;在传热过程中,随着距离增加,温度衰减很快,离热源越远的点的热弹性效环应越窄。提出用非线性时序模型与前向神经网络相结合的模型(Nonlinear auto-regressive moving average neural network with exogenousinputs,NARMAX-NN)来辨识热弹性效应。用NARMAX-NN模型对高速进给系统试验台的热动态特性进行建模,获得良好的效果。此方法比多变量回归模型、反馈神经网络模型及广义最小二乘输出误差模型有更好的精度和鲁棒性,能精确地对复杂结构、多热源的时变非线性热误差特性进行建模和预测。

基于改进小波神经网络的光伏发电系统非线性模型辨识

将光伏发电系统看成基于气象参数的非线性黑箱模型,用非线性自回归外推模型对不同天气条件下的光伏发电系统进行辨识。采用了对系统维数不敏感的基于方差分析展开的改进小波神经网络对系统进行非线性自回归外推模型辨识,辨识数据和验证数据均取自实际光伏发电系统。实例研究结果表明:与Sigmoid网络函数法、树分割法及基本小波神经网络法相比,基于改进小波神经网络的非线性自回归外推模型能更好地反应各种不同天气条件下光伏发电系统的动态行为;天气波动的剧烈程度对辨识效果影响较大。

ANN非线性时间序列预测模型输入延时τ的确定

用神经网络 (ANN)建立非线性时间序列预测模型时 ,ANN输入数据延时间隔τ的选取是必须考虑的一个方面 .目前关于延时间隔τ选取的流行做法是 :将τ确定为相空间重构时的最佳延时τs.本文提出了与此不同的观点 ,即神经网络输入数据延时间隔τ的选取与τs 无直接关系 .综合考虑其他一些因素 ,认为ANN输入数据延时间隔τ取为 1是最为合理的 .

有输入未建模动态的导弹鲁棒控制器设计

在导弹系统俯仰通道中存在输入未建模动态情况下,提出了一种基于RBF神经网络和反演控制技术的非线性鲁棒控制器的设计方法。首先应用两个RBF神经网络对输入未建模动态设计了神经网络逆补偿器,然后利用反演控制技术设计了导弹非线性控制器,最后应用Lyapunov稳定性理论推导出RBF神经网络权重矢量调节律,证明了系统的所有信号均有界且全局指数收敛至原点。最后给出的BTT导弹非线性六自由度数字仿真结果显示了该设计方法的有效性。

基于NARX神经网络的轮重减载率预测方法

介绍了一种基于神经网络的轮重减载率预测方法。以左轨轨向不平顺、右轨轨向不平顺、左轨高低不平顺、右轨高低不平顺为输入,以轮重减载率为输出,采用贝叶斯正则化算法构建了NARX(外部输入非线性自回归神经网络)。仿真试验结果及与BP神经网络的比较表明,采用NARX实现轮重减载率预测是可行而有效的。NARX比BP神经网络更适用于减载率预测。

基于履带车辆车体动态响应的行驶路面不平度识别

建立了基于履带车辆车体动态响应的行驶路面不平度识别的模型。该模型采用带外源输入的非线性自回归神经网络结构,以履带车辆车体动态响应为输入、路面不平度为输出。将相关性系数、均方根误差和绝对误差累计概率密度作为识别效果的评价指标,并给出了上述三个指标的融合方法。基于正交试验设计的思路分析并实现了路面不平度识别模型输入数量和识别效果的平衡,简化了测试系统传感器的布置。分析了不同的路面、采样频率和车速下的路面不平度识别效果。结果表明,提出的不平度识别方法满足工程实际需求。