Decoupled control using neural network-based sliding-mode controller for nonlinear systems

In this paper, an adaptive neural network sliding-mode controller design approach with decoupled method is proposed. The decoupled method provides a simple way to achieve asymptotic stability for a class of fourth-order nonlinear system. The adaptive neural sliding-mode control system is comprised of neural network (NN) and a compensation controller. The NN is the main regulator controller, which is used to approximate an ideal computational controller. The compensation controller is designed to compensate for the difference between the ideal computational controller and the neural controller. An adaptive methodology is derived to update weight parts of the NN. Using this approach, the response of system will converge faster than that of previous reports. The simulation results for the cart–pole systems and the ball–beam system are presented to demonstrate the effectiveness and robustness of the method. In addition, the experimental results for seesaw system are given to assure the robustness and stability of system.     

Adaptive complementary sliding-mode control for thrust active magnetic bearing system

An adaptive complementary sliding-mode control (ACSMC) system with a multi-input-multi-output (MIMO) recurrent Hermite neural network (RHNN) estimator is proposed to control the position of the rotor in the axial direction of a thrust active magnetic bearing (TAMB) system for the tracking of various reference trajectories in this study. First, the operating principles and dynamic model of the TAMB system using a linearized electromagnetic force model is derived. Then, a conventional sliding-mode control (SMC) system is designed for the tracking of various reference trajectories. Moreover, a complementary sliding-mode control (CSMC) system is adopted to reduce the guaranteed ultimate bound of the tracking error by half while using the saturation function as compared with the SMC. Furthermore, since the system parameters and the external disturbance are highly nonlinear and time-varying, the ACSMC is proposed to further improve the control performance and increase the robustness of the TAMB system. In the ACSMC, the MIMO RHNN estimator with estimation laws is proposed to estimate two complicated dynamic functions of the system on-line. In addition, a robust compensator is proposed to confront the minimum approximated errors and achieve the robustness. Finally, some experimental results for the tracking of various reference trajectories show that the control performance of the ACSMC is significantly improved comparing with the SMC and CSMC.     

Chattering free full-order sliding-mode control

In conventional sliding-mode control systems, the sliding-mode motion is of reduced order. Two main problems hindering the application of the sliding-mode control are the singularity in terminal sliding-mode control systems and the chattering in both the conventional linear sliding-mode and the terminal sliding-mode control systems. This paper proposes a chattering-free full-order terminal-sliding-mode control scheme. Since the derivatives of terms with fractional powers do not appear in the control law, the control singularities are avoided. A continuous control strategy is developed to achieve the chattering free sliding-mode control. During the ideal sliding-mode motion, the systems behave as a desirable full-order dynamics rather than a desirable reduced-order dynamics. A systematic design method of full-order sliding-mode control for nonlinear systems is presented, which allows both the chattering and singularity problems to be resolved. Simulations validate the proposed chattering free full-order sliding-mode control.     

基于分数阶滑模控制技术的永磁同步电机控制

针对传统整数阶滑模控制系统中存在的抖震问题,本文提出了分数阶滑模控制策略并应用到永磁同步电机的速度控制.传统滑模控制器中的开关函数由作用在切换流型或其整数阶导数面推广到其分数阶导数面,利用分数阶系统的特性,缓慢地传递系统的能量,有效地削减抖震.本文采用模糊逻辑推理算法,实现软开关切换增益的自整定.仿真和实验证明,本文提出的分数阶滑模控制系统不但能有效地削减抖震,而且能保持滑模控制器对系统参数变化和外部扰动的鲁棒性.     

Adaptive hybrid control system using a recurrent RBFN-based self-evolving fuzzy-neural-network for PMSM servo drives

In this paper, an adaptive hybrid control system (AHCS) based on the computed torque control for permanent-magnet synchronous motor (PMSM) servo drive is proposed. The proposed AHCS incorporating an auxiliary controller based on the sliding-mode, a recurrent radial basis function network (RBFN)-based self-evolving fuzzy-neural-network (RRSEFNN) controller and a robust controller. The RRSEFNN combines the merits of the self-evolving fuzzy-neural-network, recurrent-neural-network and RBFN. Moreover, it performs the structure and parameter-learning concurrently. Furthermore, to relax the requirement of the lumped uncertainty, an adaptive RRSEFNN uncertainty estimator is used to adaptively estimate the non-linear uncertainties online, yielding a controller that tolerate a wider range of uncertainties. Additionally, a robust controller is proposed to confront the uncertainties including approximation error, optimal parameter vector and higher order term in Taylor series. The online adaptive control laws are derived based on the Lyapunov stability analysis, so that the stability of the AHCS can be guaranteed. A computer simulation and an experimental system are developed to validate the effectiveness of the proposed AHCS. All control algorithms are implemented in a TMS320C31 DSP-based control computer. The simulation and experimental results confirm that the AHCS grants robust performance and precise dynamic response regardless of load disturbances and PMSM uncertainties.     

压电自适应桁架结构智能振动控制

介绍了采用模糊神经网络模型进行振动主动控制的压电自适应桁架结构设计、应用及实验结果. 设计了一种具有自适应结构技术的压电主动构件结构, 并提出了具有5层结构能够自调整隶属函数的模糊神经网络控制模型. 为了验证控制模型的有效性, 搭建了配置压电主动构件的双跨桁架结构试验平台, 通过检测误差信号, 由模糊神经网络控制模型确定主动构件的驱动输出. 试验结果证实了模糊神经网络控制模型在振动抑制方面的有效性.     

Fully distributed adaptive sliding-mode controller design for containment control of multiple Lagrangian systems

In this paper, we study the containment control problem for multiple Lagrangian systems with multiple dynamic leaders in the presence of parametric uncertainties and external disturbances with fully distributed controllers under an undirected graph. We first propose a fully distributed adaptive sliding-mode control algorithm combined with distributed sliding-mode estimators, without requiring the upper bounds of the derivatives of the leaders’ states and any other global information to be known by each follower. To reduce the effect on the varying gain during the adaption mainly caused by the initial error, fully distributed adaptive time-varying sliding-mode control is presented for controller design. To tackle the chattering effect caused by the discontinuous controller, we further propose a fully distributed continuous adaptive controller, under which both the containment errors and the adaptive gains are ultimately bounded. Simulation results are given to illustrate the theoretical results.     

Continuous sliding-mode control for underactuated systems: Relative degree one and two

This paper deals with the design of sliding-mode controllers for the stabilization of some types of underactuated systems. The proposed method takes advantage of the mechanical properties of three different types of underactuated systems instead of using nonlinear transformations. In order to design the controller, some sliding variables with relative degree one and two are introduced. The theoretical differences between these approaches are discussed and some simulation results show the practical differences. Experimental results on a cart–pole system are presented to validate the proposed control strategy.     

Robust sliding-mode formation control and collision avoidance via repulsive vector fields for a group of Quad-Rotors

This paper is focused on solving the collision avoidance problem for a group of Quad-Rotors which are affected by external disturbances when they are moving in a horizontal plane by means of Repulsive Vector Fields (RVFs). The RVFs are included in an attractive potential function control strategy, that allows the Quad-Rotors to reach the desired position in a geometric pattern, together with a Continuous Sliding-Mode Control (C-SMC) strategy based on Sliding-Mode Observers (SMOs) that are used to estimate, in a finite-time, the linear and angular velocities, respectively. For this purpose, a parameter, which depends on the distance among the Quad-Rotors and their velocities, is designed in order to scale the RVFs properly. In this sense, the repulsion force will be proportional to the velocity and acceleration of the Quad-Rotor when it detects any obstacle. A RVF not properly scaled may result in a collision or a Quad-Rotor movement far away from its desired position. The combination between the C-SMC strategy, the SMOs, the attractive potential functions and the RVFs robustly solves the formation control and avoidance collision problem under the presence of disturbances. Numerical simulations illustrate the performance of the RVFs when the Quad-Rotors are in risk of collision.     

智能控制的模拟自然环境研究

创建一种合理的温室智能控制拓扑结构,应用智能控制技术和先进的算法,根据作物生长的数据库,模拟自然环境,实现对温室环境类自然的最优控制,为实现应用于植物生长箱和人工气候室的环境控制提供了可行性依据。     

Design and real-time implementation of perturbation observer based sliding-mode control for VSC-HVDC systems

This paper develops a perturbation observer based sliding-mode control (POSMC) scheme for voltage source converter based high voltage direct current (VSC-HVDC) systems. The combinatorial effect of nonlinearities, parameter uncertainties, unmodelled dynamics and time-varying external disturbances is aggregated into a perturbation, which is estimated online by a sliding-mode state and perturbation observer. POSMC does not require an accurate system model and only one state measurement is needed. Moreover, a significant robustness can be provided through the real-time compensation of the perturbation. Four case studies are carried out on the VSC-HVDC system, such as active and reactive power tracking, AC bus fault, system parameter uncertainties, and weak AC grid connection. Simulation results verify its advantages over vector control and feedback linearization sliding-mode control. Then a hardware-in-the-loop (HIL) test is undertaken to validate the implementation feasibility of the proposed approach.     

Generalized predictive control using recurrent fuzzy neural networks for industrial processes

This paper presents a design methodology for predictive control of industrial processes via recurrent fuzzy neural networks (RFNNs). A discrete-time mathematical model using RFNN is constructed and a learning algorithm adopting a recursive least squares (RLS) approach is employed to identify the unknown parameters in the model. A generalized predictive control (GPC) law with integral action is derived based on the minimization of a modified predictive performance criterion. The stability and steady-state performance of the resulting control system are studied as well. Two examples including the control of a nonlinear process and the control of a physical variable-frequency oil-cooling machine are used to demonstrate the effectiveness of the proposed method. Both results from numerical simulations and experiments show that the proposed method is capable of controlling industrial processes with satisfactory performance under setpoint and load changes.     

仿人智能控制与模糊控制神经网络融合技术

在分析了仿人智能控制器（ＨＩＣ）的特性基础上,针对其不足,提出将仿人智能控制器与模糊控制相融合的仿人智能模糊控制器（ＨＩＦＣ）,实践表明ＨＩＦＣ的性能大大优于基本ＨＩＣ和一般模糊控制器。最后提出一种正交基函数神经网络模型实现ＨＩＦＣ的方法,为ＨＩＦＣ的自学习提供了一条有效途径。     

智能控制及其在电弧炉除尘系统中的应用

针对炼钢除尘系统的复杂过程提出了智能控制方法。由基本控制器给定一个基本输出值，通过神经网络预测冶炼不同阶段烟气量变化趋势，并以此来改变系统的工作点，及时调整风机转速。采用Ｆｕｚｚｙ推理的方法补偿布袋过滤器对系统的影响。实际应用表明，该方法可获得很好的除尘效果，降低除尘系统的运行成本。     

面向FPGA便捷部署的智能模型预测控制

现场可编程逻辑门阵列(FPGA)具有可编程、易并行化的独特优势, 是实现一体化感知、决策、控制最具前景的人工智能芯片之一, 但其硬件描述语言(HDL)不易掌握. 本文提出了一种基于神经网络的智能MPC及其FPGA便捷部署方法, 使用高层次综合(HLS)生成HDL代码, 并通过MATLAB-Modelsim联合仿真验证代码功能, 可克服人工编写HDL代码的困难, 提高控制算法的部署效率. 该方法利用了深度神经网络的结构特点和FPGA的并行计算优势, 离线训练神经网络在线仅需硬件化正向传播, 在低资源占用的同时具有严格计算时间保证. 将所提方法分别应用于高速、高维控制系统中, FPGA在环测试验证了其有效性.     

汽车智能防撞自适应控制研究与仿真

针对汽车防撞模糊控制模型不能自动调整参数的缺点,建立汽车防撞自适应模糊推理模型。采用混合学习算法对自适应模糊推理模型的前提参数和结论参数进行辨识,以加速收敛。经模拟训练和仿真输出结果证明,该模型能够对汽车防撞模糊控制器隶属函数和模糊规则进行优化,较好地实现紧急报警情况下的汽车防撞自适应控制。     

转炉炼钢终点磷的智能预报

提出了一种用于预报转炉炼钢终点磷含量的智能方法, 在该方法中, 采用模糊推理和遗传算法, 其中模糊推理用于估算转炉熔池的磷含量, 而模糊推理模型中的各个系数则由遗传算法辨识与优化. 为了提高熔池磷的估算精度, 同时还设计了一种神经网络以补偿来自模糊推理过程的误差. 仿真结果表明了该方法的有效性.     

故障诊断专家系统综合智能推理技术研究

推理机在故障诊断专家系统中起着非常重要的作用。提出普通规则、模糊逻辑和模糊神经网络推理相结合的综合智能推理机应用于故障诊断专家系统。综合智能推理机既能提高诊断推理的速度,又可以提高诊断推理的准确程度。通过对某型号导弹故障诊断验证,采用综合智能推理机诊断快速、准确率高,取得了较好的诊断效果。     

Adaptive neural dynamic surface control for servo systems with unknown dead-zone

An adaptive neural controller is proposed for nonlinear systems with a nonlinear dead-zone and multiple time-delays. The often used inverse model compensation approach is avoided by representing the dead-zone as a time-varying system. The “explosion of complexity” in the backstepping synthesis is eliminated in terms of the dynamic surface control (DSC) technique. A novel high-order neural network (HONN) with only a scalar weight parameter is developed to account for unknown nonlinearities. The control singularity and some restrictive requirements on the system are circumvented. Simulations and experiments for a turntable servo system with permanent-magnet synchronous motor (PMSM) are provided to verify the reliability and effectiveness.