Joël Bordeneuve-Guibé

Active Control of a clamped beam equipped with piezoelectric actuator and sensor using generalized predictive control

A predictive control method to perform the active damping of a flexible structure is here presented. The studied structure is a clamped-free beam equipped with collocated piezoelectric actuator/sensor. Piezoelectric transducers advantages lie in their compactness and reliability, making them commonly used in aeronautic applications, context in which our study fits. Theirs collocated placement allow the use of well-known control strategies with guaranteed stability. First an analytical model of this equipped beam is given, using the Hamiltons principle and the Rayleigh-Ritz method. After a review of the experimental setup (and notably of the piezoelectric transducers), two control laws are described. The chosen one, generalized predictive control (GPC), is compared to a typical control law in the domain of flexible structures, the positive position feedback, one of the control law mentioned above. Major befits of GPC lie in its robustness in front of model uncertainties and others disturbances. The results given come from experiments on the structure which performed by a DSP. GPC appears to suit for the considered studys context (i.e. damping of the first vibration mode). Some improvements may be reached. Among them, a more complex structure with more than a single mode to damp, and more uncertainties may be considered.

Robust multivariable predictive control: an application to an industrial test stand

The air conditioning system of an aircraft is used to regulate the cockpit temperature and pressure during flight and usually generates its airflow from the compressor turbine of the jet engine. Testing an air conditioning system requires simulation of the running conditions at ground level. The article concerns the application to such a simulation of /spl alpha/-MPC, a robust extension of the initial multivariable predictive control (MPC) law that improves the disturbance-rejection properties of the closed-loop system, reducing the H/sup /spl infin//-norm of the multivariable sensitivity function with an extra parameter. This augmented algorithm has been chosen to carry out the new tests on the industrial process. Experimental recordings reported here have confirmed significant performance improvement with this new approach relative to the former PID regulation. The article is organized as follows. First we introduce the original MPC. Next we describe the extended /spl alpha/-MPC algorithm and analyze the robustness of the closed-loop system through the H/sup /spl infin// approach. Then we discuss the methodology of the control design task and describe the experimental test stand, focusing on the software and hardware implementation. Finally, we report the results of the /spl alpha/-MPC control law on the actual test stand. Special attention is given to the comparison with the former control system.

Expert supervision of an anti-skid control system of a commercial aircraft

A rule-based supervising system that incorporates fuzzy logic has been designed to back-up a conventional anti-skid braking system (ABS). Expressing the expert knowledge about the ABS in terms of linguistic rules, the supervising fuzzy system adapts the reference wheel slip of the ABS with respect to the actual runway condition. Two approaches are presented: The first uses a simple rule-based decision logic, which evaluates a new reference slip directly from the measured system variables. The second approach employs an explicit identification of the runway condition, which is used as input information of a fuzzy system to evaluate a new reference slip. This application example demonstrates the capabilities of a parallel use of conventional control techniques and fuzzy logic.

Advanced Predictive Control for Aircraft Guidance

Abstract In this paper, a supervised extension of the Generalized Predictive Control (GPC) is proposed. Its main purpose is to improve the robustness towards modeling errors. First, the SISO case is developed and applied to the adaptive control of the flight path angle of a fighter aircraft. A high-performance Recursive Least Squares (RLS) estimator and a model-based estimator are also compared. Finally, the SISO algorithm is extended to the 2 input-2 output case and applied to the lateral control of a large transport aircraft.

ACTIVE CONTROL OF A STRUCTURE WITH SLOSHING PHENOMENA

Abstract In this article, a Generalized Predictive Control (GPC) is used to perform the active damping of a structure with sloshing phenomena. The studied flexible structure has been designed to have the same dynamic behavior than an aircraft wing (the first bending mode has a frequency of 1.59 Hz). It is composed by a free-clamped beam and a tank and is equipped with piezoelectric actuators. The purposes here are on the one hand to perform the damping of the structure i.e. to perform as efficiently as possible the vibratory disturbances rejection and on the other hand to evaluate the performances of the GPC control law, in the domain of flexible structures. The first part of the article deals with the sizing and the modeling of the active structure using the Lagrangian approach and a FEM software. The second part concerns the control law. The used control law – i.e. GPC – has been chosen because of its good control performances, its compactness and its robustness in front of model mismatches. These advantages are particularly interesting in this study, because the control must be performed for several levels of filling of the tank, in a MIMO, multi-modes context. Experiments have been performed (for the control of the first bending mode and the control of the first bending and twisting modes). In each case, GPC is efficient. Moreover, the use of GPC leads to a unique global behavior for the controlled structure, whatever the tanks filling configuration may be.

Expert supervision of conventional control systems

The objective of this paper is to outline a general concept for the design of supervising fuzzy controllers to back up or monitor a conventional control system. The use of fuzzy logic in an external, hierarchical control structure provides a systematic approach to integrate heuristics in a conventional control loop. Supervising techniques become especially interesting, when the system to be controlled is highly non-linear (parameter variation, saturation of the control surfaces, etc.). By the means of two application examples it is shown, how this method can effectively be used to improve the performance of a conventional control system. Both examples are part of an extended research project that is being carried out at Aerospatiale and ENSICA, in France to study the role of fuzzy control for potential applications in aircraft control systems.

Robust multivariable predictive control: application to an industrial test stand

This paper reports a theoretical extension of multivariable predictive control (MPC). The robustness of an augmented algorithm (/spl alpha/-MPC) for a general M-input N-output system is explored. It is shown that an extra parameter /spl alpha/ in the criterion function can reduce the H/sub /spl infin//-norm of the multivariable sensitivity function, thus improving the disturbance-rejection properties of the closed loop system. This control law is finally applied to a test stand for air conditioning equipment of aircraft with a great improvement of performances regarding the former regulation.

Interactions of Aircraft Design and Control: Actuators Sizing and Optimization for an Unstable Blended Wing-Body

In this paper the problem of integrated design and control for a civil blended wing-body aircraft is addressed. Indeed this configuration faces remarkable challenges related to handling qualities: namely the aircraft configuration in this study features a strong longitudinal instability for some specific flight points. Moreover it may lack control efficiency despite large and redundant movables. Stabilizing such a configuration may then lead to high control surfaces rates, meaning significant energy penalty and installation mass for flight control actuators, as well as challenges for actuators space allocation. Those penalties should therefore be taken into account early in the conceptual design phase, instead of being checked afterwards. Our approach consists in simultaneously designing a stabilizing controller and actuators dynamic characteristics, namely their bandwidth, for a given aircraft configuration. Our method relies on latest developments on nonsmooth optimization techniques for robust control design. For any aircraft configuration, guaranteed stability performance, as well as optimized control surfaces allocation with respect to their relative efficiency and inertia, are obtained. From these results, a segregation among trim and maneuver control surfaces is obtained, with guarantee that the later ones are able to cope with aircraft instability. Then it is checked that remaining trim control surfaces are sufficient for equilibrating the aircraft in any flight condition. This approach allows for a fast prototyping of control surfaces and actuators for unstable configurations. Different control surfaces layouts are evaluated in order to show the exibility of the method.

Adaptive Output Feedback Control Based on Neural Networks: Application to Flexible Aircraft Control

Abstract Abstract One of the major challenges in aeronautical flexible structures control is the uncertain or the non stationary feature of the systems. Transport aircrafts are of unceasingly growing size but are made from increasingly light materials so that their motion dynamics present some flexible low frequency modes coupled to rigid modes. For reasons that range from fuel transfer to random flying conditions, the parameters of these planes may be subject to significative variations during a flight. A single control law that would be robust to so large levels of uncertainties is likely to be limited in performance. For that reason, we follow in this work an adaptive control approach. Given an existing closed-loop system where a basic controller controls the rigid body modes, the problem of interest consists in designing an adaptive controller that could deal with the flexible modes of the system in such a way that the performance of the first controller is not deteriorated even in the presence of parameter variations. To this purpose, we follow a similar strategy as in Hovakimyan (2002) where a reference model adaptive control method has been proposed. The basic model of the rigid modes is regarded as a reference model and a neural network based learning algorithm is used to compensate online for the effects of unmodelled dynamics and parameter variations. We then successfully apply this control policy to the control of an Airbus aircraft. This is a very high dimensional dynamical model (about 200 states) whose direct control is obviously hard. However, by applying the aforementioned adaptive control technique to it, some promising simulation results can be achieved.

A NEURAL PREDICTIVE CONTROL SCHEME FOR NONLINEAR PLANTS

Since the last three decades predictive control has shown to be successful in control industry, but its ability to deal with nonlinear plants is still under research. Generalized Predictive Control (GPC) was one of the most famous linear predictive algorithms. The control law of GPC contains two parameters that describe the system dynamics: system free response ( f ) and system impulse response matrix (G ). Often these parameters are calculated from the discrete linear model. For nonlinear systems, either a nonlinear system model is instantaneously linearized or a nonlinear optimization is used. The validity of the linear model is the shortcoming of the first one and the possibility of non-uniqueness of local minimum is that for the second. In this paper, a neural network (NN) model is used as a predictor to calculate these parameters for GPC. The nonlinear system free response is obtained instantaneously while dynamic response is linearized every batch of time. The method is tested on a benchmark nonlinear model. Results are compared with that of others neural predictive techniques found in previous literature. Also, the method is applied and validated on a realistic multivariable aircraft model. The simulation results show that this method has some good advantages over others neural predictive techniques. In one hand, the system dynamics parameters are calculated more accurately directly from the nonlinear NN model. And in the other hand, the used linear GPC has a cost function with only one global minimum. The method, as a trade-off between nonlinear neural predictive control (NPC)and instantaneous linearization approximate neural linear predictive control (APC), is promising for control of nonlinear systems.