Finite-time fuzzy adaptive output feedback control of electro-hydraulic system with actuator faults

Abstract

This article studies the fuzzy finite-time output -feedback control problem on electro-hydraulic systems (EHSs). The EHSs addressed in this study contain unknown states and are subject to actuator faults. The internal leakage and friction are unknown nonlinear functions. By utilizing fuzzy logic systems (FLSs) to approximate the unknown nonlinear dynamics in the EHS and a fuzzy adaptive state estimators are formulated. Subsequently, the unknown states are obtained via the designed state estimators. Further, under the frameworks of finite-time stability a backstepping fuzzy fault-tolerant control (FTC) method is presented. The stability of the controlled electro-hydraulic system is proved by constructing composite Lyapunov functions. The developed fuzzy adaptive control scheme does not assume that the states of controlled electro-hydraulic system must be fully measured directly and has fast convergence and strong robustness against uncertainties. The simulated results and a comparison with the existing control methods are provided to show the validation of the presented control method.

Introduction

In the past decade, an electro-hydraulic system has found extensive applications in unmanned aircraft systems, marine systems, robotic manipulators and active suspensions. It generally has advantages over electric motors, containing high force to weight ratio, compact size and fast response. With the increasing applications of hydraulic mechanisms, the issue of stabilizing electro-hydraulic systems has attracted tremendous attention in recent years. To handle this issue, many control methods were developed. For instance, [5] presented a robust adaptive controller for a single-rod electro-hydraulic actuators subject to nonlinear parameters. By constructing a novel Lyapunov function, [6] developed a sliding- mode control scheme. In order to attenuate the value fault of an independent metering value, a fault-tolerant controller (FTC) was developed in [4]. The backstepping controller [7] was also used in electro-hydraulic systems in handling mismatched disturbances. In order to solve the uncertain nonlinearity and uncertain parameters in hydraulic systems simultaneously, an adaptive robust control method was presented in [21].

Note that full state information is difficult to acquire for many hydraulic applications because of cost reducing, volume or weight limitations and heavy measurement noises. In fact, only position information is available in practical electro-hydraulic systems. Thus, it is significant to investigate the issue of output feedback control for EHSs. By introducing the extended state observer to estimate the unknown states and via backstepping control technique, [22] presented a observer–based output -feedback control method for the EHSs. In addition, an output-feedback passivity-based controller was also developed in [10] by designing a high-gain observer to obtain the unknown states in electro-hydraulic systems.

Note that the above mentioned control schemes all require precise structural information of the considered hydraulic system. Especially, the frictions and internal leakage are both required to be known. Therefore, they cannot effectively control the hydraulic systems containing unknown nonlinear dynamics [1]. Since fuzzy systems [18] and neural networks [8], [14] have a good ability to approximate continuous functions, they are utilized to solve the control problem on the uncertain nonlinear systems. In [2], the authors integrated fuzzy learning mechanisms into the modeling of EHS, and proposed a kind of fuzzy PI controller. In [8], a neural network adaptive control scheme was developed for a class of EHSs and achieved the better control tracking performance of the cylinder position in presence of lumped uncertainties. In [24], a neural network adaptive backstepping control method was developed for the hydraulic knee exoskeleton with nonlinear disturbance.

Although the fuzzy and neural controllers mentioned above have skillfully solved the tracking control problem for electro-hydraulic systems, they only ensure that the electro-hydraulic systems are asymptotic stability. In fact, for many practical systems like the electro-hydraulic system addressed by this study, it is more desired that state trajectories of the controlled systems converge to the equilibrium points within a finite-time interval instead of an infinite–time interval. Furthermore, the controllers developed by finite-time stability are fast transient and strong robust to the uncertainties. Thus, the finite-time control has been obtained much attention in recent years. For example, [17] developed a fuzzy adaptive finite-time control approach for some nonlinear systems with unknown dynamic functions. In [12], the authors studied adaptive NN finite-time control design problem for strict-feedback uncertain nonlinear systems. In [11], a fuzzy observer-based finite-time output tracking control scheme was presented for a class of strict-feedback unknown nonlinear systems. In [26], a fast finite time adaptive control issue was discussed for a class of uncertain nonlinear systems. Besides, by the finite-time stable theory, [16], [23], [25] proposed fuzzy and neural finite–time controllers for some practical systems, and achieve good control performances. However, by far, to authors’ best knowledge, no finite-time output feedback control methods have been reported for the EHSs subject to the unknown states and the actuator faults, which motivates us to conduct this study.

Inspired by the above reviews, this study considers the fuzzy output-feedback finite-time control problem on the EHSs. The addressed EHSs are subject to immeasurable states and actuator faults. Besides, the friction and internal leakage are nonlinear uncertainties. By utilizing the FLSs to model the uncertain nonlinear EHSs, an adaptive fuzzy state estimators is presented and the unknown states are thus obtained. Further, by establishing composite Lyapunov functions and using the finite time stability concepts, a novel output-feedback finite-time fuzzy FTC scheme is formulated. The main features of this paper are as follows:

(i) This study first presents a fuzzy finite-time output -feedback controller for the uncertain EHSs by designing a novel fuzzy state estimator. It is mentioned that although the literature [10], [22] also address the output feedback control problem, they both require that the nonlinear dynamics of the friction and internal leakage are known and must satisfy the Lipschitz conditions. However, the proposed fuzzy output feedback controller in this study removes the restrictive conditions required by [10], [22].

(ii) Since the presented fuzzy output -feedback finite-time control strategy is designed under the finite-time stability, it can ensure that the electric-hydraulic systems are stability within a finite-time interval, and as well has properties of fast convergence and a strong robustness against the actuator faults compared with [2], [8], [24].