IFAC Journal of Systems and Control最新文献

In this paper we propose a vertical stabilization (VS) control system for tokamak plasmas based on the extremum seeking (ES) algorithm. The gist of the proposed strategy is to inject an oscillating term in the control action and exploit a modified ES algorithm in order to bring to zero the average motion of the plasma along the unstable mode. In this way, the stabilization of the unstable vertical dynamic of the plasma is achieved. The approach is validated by means of both linear and nonlinear simulations of the overall ITER tokamak magnetic control system, with the aim of demonstrating robust operation throughout the flat-top phase of a discharge and the capability of reacting to a variety of disturbances.

The problem of an autonomous agent moving on a planar surface, such as an aerial drone or a small naval vessel can be treated as navigation between a series of points. While nominally the movement between each pair of points can be treated as a 1D projection of the movement on the vector connecting the two points, in the presence of disturbances, the full problem on the plane must be considered. The time optimal solution is now dependent on the value and direction of the disturbance which in this paper is assumed to be a constant inertial velocity of the medium (wind or current, respectively). We address the minimum time problem of movement on a 2D plane with quadratic drag, under norm state (inertial vessel velocity), and norm control (acceleration) constraints. The structure and properties of the optimal solution are found and analyzed, utilizing the Pontryagin Maximum Principle (PMP) with control and state constraints. Simulations supporting the results are provided and compared with those of the open-source academic optimal control solver Falcon.m.

This paper addresses the $H_2$ and $H_{\infty }$ control of Markov Jump Systems subject to a class of nonlinearities that include them on the Lur’e system framework, with sector slope optimization. The proposed controllers are robust to nonlinearities belonging to a sector. We propose mode-dependent controllers; in other words, the Markov mode $\theta \left(t\right)$ is available to controllers. The design of such controllers considers the optimization of the $H_2$ and $H_{\infty }$ costs as well as the sector slope, which determines the class of accepted nonlinearities. The results are illustrated through two practical examples: coupled electrical machines and the semi-active suspension of a quarter car.

The problem of maximum power point tracking (MPPT) control of uncertain photovoltaic (PV) systems is addressed in this paper, on the basis of an adaptive Kalman-like observer. The system is composed of a photovoltaic generator (PVG) supplying power to a DC centrifugal pump driven by a DC–DC boost converter. The PVG is connected to the converter by a long PV cable. Indeed, many PV power plant situations require that the PV panels be installed at a great distance from the converter for several reasons including the security and the property of the site and its exposure to good daily solar irradiance. This obviously leads to the difficulty of PVG output current and voltage measurement using ordinary sensors. Such quantities measurement being in fact necessary for MPPT algorithms and controllers design. Furthermore, the specific parameters of the long PV cable used, namely its resistance and inductance, could have a significant effect on the MPPT control efficiency if instead of PVG delivered voltage and current measurements, only those accessible at the converter side of the cable are used in MPPT controllers design. Therefore, this work aims at overcoming the two aforementioned issues by proposing an adaptive output-feedback control-based MPPT for PV systems. An adaptive Kalman-like observer providing online estimates of inaccessible state variables as well as of PV cable unknown parameters is firstly designed. Then, a backstepping control law is synthesized to meet the MPPT objective. The convergence of both adaptive observer and closed-loop system control has been established using Lyapunov approach and the effectiveness of the developed adaptive MPPT controller in achieving an accurate and robust MPPT control towards uncertainties of the PV cable parameters, has been validated through numerical simulations.

A novel actuator failure compensation scheme is proposed for affine nonlinear uncertain systems (not necessarily Lipschitz) subject to persistent/intermittent actuator faults/failures unknown in time, magnitude and pattern. The proposed control methodology satisfies the nonlinear separation principle through a modular backstepping control. The controller is then augmented with multiple estimation models to estimate failure induced parametric uncertainties and unknown system parameters. The output transient performance at start up and post-failure instances, is improved on account of a two layer adaptation which enhances the convergence speed and accuracy of parameter estimates. The proposed fault tolerant control (FTC) method yields a faithful accommodation of uncertain finite as well as infinite/intermittent/persistent actuator failures while ensuring satisfactory output transient and steady state performances. Further, compared to existing multiple model based adaptive fault tolerant control design for nonlinear systems, the proposed methodology circumvents the issues of stability due to switching between different models and utilizes a minimum number of estimation models for parameter estimation without compromising on the output performance. Consequently, the computational burden is also reduced. Compared to multiple model adaptive control based FTC strategies proposed earlier which assume finite actuator failures and Lipschitz nonlinear system, the proposed method is applicable to both Lipschitz and non-Lipschitz nonlinear systems affected by intermittent actuator failures. Using the concepts from stability analysis in random nonlinear impulsive systems, the $L_{\infty }$ and $L_2$ bounds on tracking error for all future time are derived in the case of intermittent/persistent actuator failures obtained using the proposed fault-tolerant controller. The improvement of output transient performance in the proposed control scheme in comparison with controller with single identifier, is theoretically proved and quantified.

Fault detection and isolation on hydraulic systems are very important to ensure safety and avoid disasters. In this paper, a fault detection and isolation method, based on the flatness property of nonlinear systems, is experimentally applied on the three-tank system, which is considered as a popular prototype of hydraulic systems. Specifically, fault indicators, called residues, are generated using flat output measurements, and for the purpose of fault isolation, a definition of the isolability is introduced. This definition allows the characterization of flat outputs that are useful for fault isolation. A sensitivity analysis is proposed in order to improve the robustness of the method. Multiplicative faults are considered on sensors and actuators.

Fault classification is an important part in industrial process for process monitoring and control. As an ensemble learning approach for classification, random forests has been widely used in different areas. Taking into account the performance of individual decision tree, the diversity between trees and the difference between process data, a k nearest neighbors-hierarchical clustering (KNN-HC) method is proposed in this paper for dynamic ensemble selection (DES) in random forests. In addition, a weighted probability fusion strategy is developed as an alternative of majority voting rule. The experimental evaluation of the proposed method is carried out through the Tennessee Eastman (TE) benchmark process. Results show that the proposed method outperforms three conventional methods, the original random forests (RF) and the static selection based random forests.

Truck and N-trailer mobile robots find use in freight transportation, urban transportation, mining as well as in agriculture. The article proposes a nonlinear optimal (H-infinity) control approach for the truck and N-trailer robotic system. The method has been successfully tested so far on the control problem of several types of robotic vehicles and here it is shown that it can also provide an optimal solution to the control problem of the underactuated truck and N-trailer mobile robot. To implement this control scheme, the state-space description of the kinematic model of the truck and N-trailer robotic system undergoes first approximate linearization around a temporary operating point, through first-order Taylor series expansion and through the computation of the associated Jacobian matrices. Next, an optimal (H-infinity) feedback controller is designed. To select the feedback gains of the optimal (H-infinity) controller an algebraic Riccati equation is solved at each time-step of the control method. The global stability properties of the control loop are proven through Lyapunov analysis. Finally, to implement state estimation-based feedback control, the H-infinity Kalman Filter is used as a robust state estimator.

The property of system invertibility is important both from a theoretical standpoint and an important consideration in left and right inversion problems that deal with input estimation and tracking control respectively. Nuances associated with the notion of relative degree in multi-input, multi-output (MIMO) systems on account of input–output delays pose challenges in system inversion problems and are more acute in the case of non-square systems. In this note, using the LTI discrete-time MIMO systems framework, we review existing notions of relative degree and develop new definitions and generalizations for relative degree to help address some of the gaps in system inversion problems. In doing so, we also show through simple but elegant results that existence of well-defined relative degree for an LTI MIMO system implies its invertibility. Furthermore, we also highlight the utility of these ideas in practical tracking control and input estimation situations and problems.

There exist several approaches to design the optimal control strategy to harvest wave energy with a point absorber. However they are generally based on the assumption that the WEC and the PTO dynamics are well-known. In the practical WEC control implementation, this is generally not the case. The objective of this paper is to design a robust optimal control strategy that can take into account the uncertain WEC and PTO dynamics. Our choice is a robust adaptive PI control law. The proposed controller is validated and compared through simulation for irregular sea states.