航空发动机递归神经网络分路式解耦控制

针对航空发动机多变量控制中变量之间的耦合问题,提出了一种基于递归神经网络的分路式动态解耦控制方法,给出了发动机双路式解耦控制系统的结构及其解耦原理和算法。利用递归小波网络较强的动态非线性映射能力,在线完成发动机各控制通道的模型辨识,并回馈对应的灵敏度信息;神经网络PID控制器根据回馈的信息在线自适应调整参数,实现发动机各通道的准确跟踪和分路独立控制。仿真表明,该方法在保证控制系统良好的动态和稳态性能的同时,有效地减小了各回路之间的耦合影响,能够成功应用于发动机控制系统的解耦。      

基于小波神经网络辨识的PID神经MRAC研究

提出了一种基于小波神经网络辨识的PID神经网络模型参考自适应控制方法。该方法采用小波神经网络作为辨识器,PID神经网络作为控制器在线调节。由于小波变换具有良好的时频局部特性,神经网络具有强大的非线性映射能力,自学习、自适应等优势,采用规范正交的小波函数作为神经网络的基函数构成小波神经网络,该网络兼有小波函数的紧支性、波动性以及神经网络的非线性映射能力,自学习、自适应能力等优点,仿真结果表明用该方法构成的控制系统收敛速度快,逼近精度高,鲁棒性好,优于一般的BP网络控制。     

基于小波神经网络的自适应飞/推控制系统设计

基于小波神经网络提出了一种H∞自适应控制方法。控制器由等效控制器和H∞控制器两部分组成。用小波神经网络逼近非线性函数,并把逼近误差引入到权值的自适应律中用以改善系统的动态性能。H∞控制器用于减弱外部及神经网络的逼近误差对跟踪的影响。所设计的控制器不仅保证了闭环系统的稳定性,而且使外部干扰及神经网络的逼近误差对跟踪的影响减小到给定的性能指标。最后基于所设计的控制方法对新一代歼击机设计了飞/推控制系统,并对飞机作大迎角机动仿真。仿真结果表明所设计的飞/推控制系统是有效的,同时验证了所设计的非线性控制方法是有效性的。      

基于功率反馈的涡轴发动机神经网络PID控制研究

通过对功率平衡关系进行分析,提出了利用功率反馈设计智能神经网络PID控制器的方法。基于BP（Back Propagation）神经网络,将功率信号作为神经网络的输入层信号,并改进了网络权值的学习规则。通过在线整定PID参数,控制器能够根据功率误差信号的变化实时调整控制参数,从而使系统自主寻找到功率平衡点,具有良好的稳态和动态响应特性。仿真结果表明：该方法可以使涡轴发动机在全包线范围内具有理想的控制性能。     

某于功率反馈的涡轴发动机神经网络PID控制研究

通过对功率平衡关系进行分析,提出了利用功率反馈设计智能神经网络PIO控制器的方法。基于BP(Back Propagation)神经网络,将功率信号作为神经网络的输入层信号,并改进了网络权值的学习规则。通过在线整定PID参数,控制器能够根据功率误差信号的变化实时调整控制参数,从而使系统自主寻找到功率平衡点,具有良好的稳态和动态响应特性。仿真结果表明：该方法可以使涡轴发动机在全包线范围内具有理想的控制性能。     

基于自适应神经网络的非线性飞行控制

结合反馈线性化和神经网络提出了一种新的飞行控制系统设计方法,通过反馈线性化将非线性耦合系统等效转换为线性解耦系统,利用具有在线学习能力的神经网络补偿反馈线性化的误差,建立了基于自适应神经网络的控制结构,并利用李亚普诺夫函数导出了网络权值的自适应调整规则。在巡航弹地形跟踪应用中的结果表明,该控制系统不仅具有精确的跟踪性能而且具有良好的鲁棒性。     

一种新的干扰力矩补偿方法在测试转台中的应用

为提高转台位置跟踪精度,提出了一种新的复合控制方法:电机中摩擦模型采用摩擦参数为非一致性变化的LuGre动态模型。控制器采用参数自适应律和CMAC神经网络来估计未知LuGre模型参数和辨识位置周期摩擦扰动并给与补偿。该方法保证了闭环系统全局稳定性和对期望位置信号的渐进跟踪,提高了转台位置跟踪精度。     

基于反馈误差学习的神经网络控制

研究了应用神经网络和ＰＤ反馈控制实现非线性系统的自适应跟踪问题。ＰＤ反馈控制器不但保证闭环系统的稳定性,同时其输出又作为训练神网络的参考信号。证明了通过选择适当的初始加权和加权的调节速率可以实现非线性系统的固定点跟踪。     

Robust Control for Static Loading of Electro-hydraulic Load Simulator with Friction Compensation

Load simulator is a key test equipment for aircraft actuation systems in hardware-in-the-loop-simulation. Static loading is an essential function of the load simulator and widely used in the static/dynamic stiffness test of aircraft actuation systems. The tracking performance of the static loading is studied in this paper. Firstly, the nonlinear mathematical models of the hydraulic load simulator are derived, and the feedback linearization method is employed to construct a feed-forward controller to improve the force tracking performance. Considering the effect of the friction, a LuGre model based friction compensation is synthesized, in which the unmeasurable state is estimated by a dual state observer via a controlled learning mechanism to guarantee that the estimation is bounded. The modeling errors are attenuated by a well-designed robust controller with a control accuracy measured by a design parameter. Employing the dual state observer is to capture the different effects of the unmeasured state and hence can improve the friction compensation accuracy. The tracking performance is summarized by a derived theorem. Experimental results are also obtained to verify the high performance nature of the proposed control strategy.     

Neural network modeling of flight control system

Modeling of angle tracking systems in the presence of actuator non-linearity such as angle, position and rate limits is a very significant and difficult task in the design and implementation of aircraft, target-tracking, and missile guided systems. A new recurrent neural network with time-delayed inputs and output feedback is used for the modeling of angle tracking systems, with emphasis on the neural network architecture, principles and algorithms. The neural network controller with modeling units for angle tracking is designed by using TMS320C25 processors. For time and size requirements, limited precision technology and look-up table technology are used in the design of the hardware and software systems. Given a set of input commands, the network is trained to control the system within the constraints imposed by actuators. The results show that the proposed networks are able to model the angle tracking system through learning without separate consideration of the non-linearity of actuators     

基于新型神经网络的航空发动机多变量控制

根据PID控制结构提出了一种新型神经网络控制器,对其基本结构和学习算法等进行了分析。结合某型航空涡喷发动机双变量控制需求,利用2个结构相同、相互联系的神经网络,实现了发动机双变量控制和接耦。在不同飞行条件、不同发动机工作状态下的仿真表明,控制器具有良好的控制性能     

RFNN control for PMLSM drive via backstepping technique

A robust fuzzy neural network (RFNN) control system is proposed in this study to control the position of the mover of a permanent magnet linear synchronous motor (PMLSM) drive system to track periodic reference trajectories. First, an ideal feedback linearization control law is designed based on the backstepping technique. Then, a fuzzy neural network (FNN) controller is designed to be the main tracking controller of the proposed RFNN control system to mimic an ideal feedback linearization control law, and a robust controller is proposed to confront the shortcoming of the FNN controller. Moreover, to relax the requirement for the bound of uncertainty term, which comprises a minimum approximation error, optimal parameter vectors and higher order terms in Taylor series, an adaptive bound estimation is investigated where a simple adaptive algorithm is utilized to estimate the bound of uncertainty. Furthermore, the simulated and experimental results due to periodic reference trajectories demonstrate that the dynamic behaviors of the proposed control systems are robust with regard to uncertainties.     

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。     

机械手的神经网络自适应滑动模控制器设计

 针对多关节机械手的鲁棒跟随控制器设计问题，提出了一种新的机械手神经网络自适应滑动模控制器设计方法，机械手的动力学非线性假设是完全未知的。在提出的控制结构中，高斯径向基函数神经网络用于在线补偿机械手的动力学非线性，参数学习律由稳定性理论得到。给出了系统稳定性和参数收敛性的证明。最后提出方法的可行性通过仿真得到验证。     

Induction motor servo drive with adaptive rotor time-constantestimation

An indirect filed-oriented induction motor (IM) position servo drive with adaptive rotor time-constant estimation and an on-line trained neural network controller is presented. First, the rotor time-constant is estimated real-time on the basis of the model reference adaptive system (MRAS) theory. Next, a linear model-following controller (LMFC) is designed according to the estimated plant model to allow the state responses of the plant to follow the reference model. Then an augmented signal generated from the proposed neural network controller, whose connective weights are trained on-line according to the model-following error of the states, is added to the LMFC system to preserve a favorable model-following characteristic under various operating conditions     

一种通用神经网络模型及其应用研究

提出了一种新型模拟生物神经网络的性能特征通用人工神经网络模型。该模型含有跨层连接和反馈连接,在这些连接上设有选择性开关,能够控制网络模型做出相应的变换。并给出了相应的学习算法。     

航空发动机神经网络自适应控制研究

本文研究神经网络自适应控制方法及其在航空发动机控制中应用。结合某型航空涡喷发动机,首先研究采用神经网络进行非线性动态系统辨识,包括神经网络模型辨识的格式、输入信号形式等问题。然后,提出了一种神经网络自适应控制方法,阐明了该方法基本结构、原理。最后,在选定的设计点处进行发动机控制系统设计,当偏离设计点时,利用神经网络很强的学习、适应能力,通过在线修正神经网络参数,使控制系统仍保持良好性能。      

基于RBF神经网络的机器人鲁棒自适应控制器设计

在考虑机器人系统中存在的模型不确定性的情况下,提出了一种基于RBF神经网络和反演技术的鲁棒自适应控制器的设计方法。首先,通过状态变换将机器人的模型转换为适用于反演技术的形式;然后,利用反演技术设计了鲁棒自适应控制器,用两个RBF神经网络分别对模型的不确定性进行了处理,并用Lyapunov稳定性理论推导出RBF神经网络的权重矩阵调节律以及相关的鲁棒项,证明了系统的全局稳定性;最后,进行了相应的仿真研究,验证了设计的正确性和有效性。     

基于神经网络的导弹稳定控制回路设计方法

基于径向基函数神经网络（RBFNN）的学习能力和逼近特性，将传统方法中的局部/整体构架蕴涵于神经网络内部，对导弹稳定控制律进行局部（定点）/整体（拟合）设计，以简化计算步骤，提高设计方法的通用性和自动化程度，以防空导弹侧向稳定控制回路设计为背景，对所提出的稳定控制回路设计方法进行了仿真，验证了该方法的可行性和有效性。     

Impedance Calculations for Elements of Sonar Arrays by Neural-Network-Based Integration

In this paper, a technique of neural network based integration is proposed to calculate the self-and mutual-impedances within arrays of sonar transducers. The multi-dimensional integrals appearing in self-and mutual-impedance formulations are transformed into neural-network-based integration and the final results can be found from look-up tables in mathematical handbooks. Initially, the integrand is modeled by a trained neural network. Integration on the integrand then becomes integration on the linear combination of weights and basis functions within the neural network. The results will become the linear combination of error functions which can be looked up in mathematical handbooks. Numerical simulation shows that the results calculated by the proposed method are consistent with those given in other existing studies. The proposed technique requires neither numerical nor artificial integration procedure. Due to the inherent learning and predicting property of neural network, only a small number of sampling points for the integrand are required in the proposed integration technique.