Adaptive disturbance attenuation via logic-based switching

The problem of attenuating unknown and possibly time-varying disturbances acting on a linear time-invariant dynamical system is addressed by means of an adaptive switching control approach. Given a family of pre-designed stabilizing controllers, a supervisory unit infers in real-time the potential behavior of each candidate controller and selects the one providing the best potential performance. To this aim, a set of test functionals is defined, which is shown to enjoy favorable inference properties under certain assumptions on the nature of the disturbances. Both persistent-memory and finite-memory test functionals are analyzed. Further, an implementation of the switching controller is proposed which always guarantees stability of the feedback loop, even if the disturbance characteristics are such that the switching is persistent. Simulation results are provided to show the effectiveness of the proposed method.     

Supervisory adaptive fault‐tolerant control against actuator failures with application to an aircraft

In this paper, a supervisory adaptive fault‐tolerant control scheme is proposed for a class of uncertain nonlinear systems with multiple inputs. The multiple inputs are the outputs of an actuator group that may act either on one control surface or on multiple control surfaces and may fail during operation. With some actuator groups as backups, the supervisory adaptive control includes 2 modes: the adaptive compensation mode and the switching mode. The former is used to compensate for the failure of an actuator group as long as at least one actuator of the group works normally, and the latter, to switch the controller from a failed group to a healthy one when the failure is detected by one of the monitoring functions that are constructed to supervise some variables related to system stability. It is shown that with the proposed scheme, all signals of the closed‐loop system are bounded, and prescribed transient and steady state performance of the tracking error can be guaranteed. An aircraft example is used to demonstrate the application of the proposed scheme.     

Safe Supervisory Control Under Observability Failure

This paper discusses about supervisory control under possible loss of observability. The final goal of this work is to design a safe (avoiding disaster states) supervisory control loop structure taking into consideration the uncertainty that stems from the fact that some event(s) may turn into unobservable at some point along the normal system operation. This kind of failure may correspond to the breakdown of some plant sensor. The supervisor finally obtained is in general: a) more permissive than the supervisor obtained assuming those events as unobservable from the beginning, since at some points it shall be better informed, and b) more restrictive than the supervisor obtained assuming that those events shall never fail, since it will have to prevent the system from following some undesirable trajectories that the system could take in presence of an observability failure. This paper presents results to obtain a safe controller that avoids disaster states in presence of the described uncertainty, and also ensures that the system behaviour will not run out from its specifications in absence of failure.     

基于多控制器的直接自适应控制

为解决多模型自适应控制可能造成的模型失配不稳定性问题,利用非伪控制理论,提出了一种直接辨识控制器的方法。由一步超前控制器构造控制器集,通过一个具有一定属性的恰当选择的代价函数,只要控制器集中至少存在一个镇定的控制器,即自适应控制问题是可行的,滞后切换逻辑总能快速将镇定控制器切换到控制回路中,保证闭环系统稳定和输出误差渐近趋于零。给出了闭环系统在L2e意义下输入输出稳定性结论,并给出仿真实例验证其有效性。     

Optimal control during feedback failure

The problem of controlling a perturbed open loop system so as to keep its performance errors within bounds is considered. The objective is to maximise the time during which performance errors remain below a prescribed ceiling, while the controlled system's parameters are within a specified neighbourhood of their nominal values. It is shown that there is an optimal open loop controller that achieves this objective. Conditions under which the optimal controller generates a bang-bang control input signal are characterised. In general, it is shown that the performance of the optimal controller can always be approximated by a bang-bang signal.     

Lyapunov-based switching supervisory control of nonlinear uncertainsystems

The problem of controlling nonlinear noisy systems affected by parametric uncertainties is approached via the introduction of a supervisor which switches on, in feedback to the plant, one controller selected from a finite set of predesigned controllers. A Lyapunov-based falsification criterion allows one to ensure robust stability in the presence of uncertain constant parameters and exogenous bounded disturbances     

Adaptive switching supervisory control of nonlinear systems with no prior knowledge of noise bounds

The problem of controlling nonlinear noisy systems affected by parametric uncertainties is approached via the introduction of a supervisor which, whenever needed, switches on, in feedback to the plant, a controller selected from a finite set of predesigned controllers. A Lyapunov-based falsification criterion allows one to ensure robust stability in the presence of uncertain constant parameters and exogenous bounded disturbances. Simulations are discussed in order to illustrate the merits of the proposed algorithms.     

Data-based supervisory control of uncertain systems with application to automatic drug delivery for anesthesia

Control of nonlinear noisy systems affected by large uncertainties is approached via the introduction of a supervisor which, whenever needed, switches on, in feedback to the plant, a controller selected from a finite set of precomputed controllers. A Lyapunov-based falsification criterion allows one to ensure robust stability in the presence of uncertain constant parameters and exogenous bounded disturbances without a priori information on the system. Applicability of the method is illustrated through simulations on an automatic drug delivery system for anesthesia.     

Approximating the Minimal Sensor Selection for Supervisory Control

This paper discusses the problem of selecting a set of sensors of minimum cost that can be used for the synthesis of a supervisory controller. It is shown how this sensor selection problem is related to a type of directed graph st-cut problem that has not been previously discussed in the literature. Approximation algorithms to solve the sensor selection problem can be used to solve the graph cutting problem and vice-versa. Polynomial time algorithms to find good approximate solutions to either problem most likely do not exist (under certain complexity assumptions), but a time efficient approximation algorithm is shown that solves a special case of these problems. It is also shown how to convert the sensor selection problem into an integer programming problem.     

A performance optimization algorithm for controller reconfiguration in fault tolerant distributed model predictive control

This paper presents a performance optimization algorithm for controller reconfiguration in fault tolerant distributed model predictive control for large-scale systems. After the fault has been detected and diagnosed, several controller reconfigurations are proposed as candidate corrective actions for fault compensation. The solution of a set of constrained optimization problems with different actuator and setpoint reconfigurations is derived by means of an original approach, exploiting the information on the active constraints in the non-faulty subsystems. Thus, the global optimization problem is split into two optimization subproblems, which enable the online computational burden to be greatly reduced. Subsequently, the performances of different candidate controller reconfigurations are compared, and the better performing one is selected and then implemented to compensate the fault effects. Efficacy of the proposed approach has been shown by applying it to the benzene alkylation process, which is a benchmark process in distributed model predictive control.     

Intelligent adaptive control of bioreactors

Supervised model-based self-tuning control of fermentation processes is addressed. The diversity, nonlinearity and time-varying nature of these processes make their control a challenging task. Conventional linear (PID) controllers with fixed parameters cannot meet the increasing performance requirements over the whole operating range. In the approach pursued in this research, a local linear model is identified at the current working point by using a limited amount input–output data obtained through an identification experiment. A linear controller is then tuned on the basis of this model. To minimize the intervention into the process operation, this tuning procedure is only initiated if the performance of the current controller deteriorates. To this end, a supervisory expert system is designed whose tasks are to monitor the process performance, design an appropriate identification experiment, validate the obtained model and tune the controller. The supervisory system is based on a combination of a state automaton with a rule-based fuzzy inference system. Experimental results have demonstrated the feasibility of this approach.     

Multiple‐model adaptive control using set‐valued observers

A multiple‐model adaptive control methodology is proposed that is able to provide stability and performance guarantees, for uncertain linear parameter‐varying plants. The identification problem is addressed by taking advantage of recent advances in model falsification using set‐valued observers (SVOs). These SVOs provide set‐valued estimates of the state of the system, according to its dynamic model. If such estimate is the empty set, the underlying dynamic model is invalidated, and a different controller is connected to the loop. The behavior of the proposed control algorithm is demonstrated in simulation, by resorting to a mass–spring–dashpot system. As a caveat, the computational burden associated with the SVOs can be prohibitive under some circumstances. Copyright © 2013 John Wiley & Sons, Ltd.     

Switching Supervisory Control Using Calibrated Forecasts

In this paper, we approach supervisory control as an online decision problem. In particular, we introduce ldquocalibrated forecastsrdquo as a mechanism for controller selection in supervisory control. The forecasted quantity is a candidate controller's performance level, or reward, over finite implementation horizon. Controller selection is based on using the controller with the maximum calibrated forecast of the reward. The proposed supervisor does not perform a pre-routed search of candidate controllers and does not require the presence of exogenous inputs for excitation or identification. Assuming the existence of a stabilizing controller within the set of candidate controllers, we show that under the proposed supervisory controller, the output of the system remains bounded for any bounded disturbance, even if the disturbance is chosen in an adversarial manner. The use of calibrated forecasts enables one to establish overall performance guarantees for the supervisory scheme even though non-stabilizing controllers may be persistently selected by the supervisor because of the effects of initial conditions, exogenous disturbances, or random selection. The main results are obtained for a general class of system dynamics and specialized to linear systems.     

Supervisory controller for reduction of wind turbine loads in curtailed operation

There are situations in which wind turbines must curtail their power, i.e. produce less power than is available from the wind. In such cases the wind turbine power can be increased or decreased if required. This gives an opportunity to strike a balance between varying power production and reducing wind turbine structural loading. To that end, a supervisory controller is designed that issues power references to the wind turbine and can be easily installed on already operational wind turbines. The wind turbine with a supervisory controller produces the required mean power, while reducing wind turbine loads by adding power variations. The extensive, realistic simulations are done to evaluate the influence of the proposed controller on the fatigue loads, extreme loads and the overall wind turbine operation. The results indicate that a significant reduction of fatigue loads can be achieved, which can increase the operating life of the structure. Furthermore, the proposed supervisory controller can be utilized as the main building block of a wind farm controller, which meets the grid code requirements and can be easily installed on very large wind farms due to minimal requirements on the farm-wide communication.     

非线性系统的直接自适应输出反馈监督模糊控制

针对一类单输入单输出非线性不确定系统，提出一种稳定的直接自适应模糊输出反馈监督控制算法，该算法不需要系统的状态完全可测的假设条件，监督控制不仅迫使系统的状态在指定的集合内，而且当模糊自适应控制处于良好的工作状态时，监督控制可以关闭，证明了整个模糊自适应输出反馈控制算法可以保证闭环系统稳定。     

Robust H ∞ control of discrete‐time Markovian jump systems in the presence of incomplete knowledge of transition probabilities and saturating actuator

This paper deals with the problem of robust H _∞ control for a class of discrete‐time Markovian jump systems subject to both actuator saturation and incomplete knowledge of transition probability. Different from the previous results where the transition probability is completely known, a more general situation where only partial information on the exact values of elements in transition probability matrix is considered. By introducing some free parameters to express the relationship for the known and the unknown elements of transition probability matrix in stability analysis, a criterion is established to guarantee the stochastic stability of the closed‐loop system as well as an H _∞ performance index. The concept of domain of attraction in mean square sense is used to analyze the closed‐loop stability, and the mode‐dependent H _∞ state‐feedback controller is designed. It is shown that, even in the absence of actuator saturation, the obtained result is less conservative than the existing one. A numerical example is provided to illustrate the effectiveness of the proposed method. Copyright © 2011 John Wiley & Sons, Ltd.     

Observer-based event-driven fault-tolerant control for a class of system with state-dependent uncertainties

The method of designing an event-driven observer-based fault-tolerant controller is addressed for a state-dependent system with external disturbance and fault in this paper. An event-driven criterion is proposed to determine the updating of the controller based on the state of the Luenberger-type state-dependent observer. As a result, communication resources can be saved significantly while the desired H_∞ performance is preserved. The observer error closed-loop system is rewritten as a time-varying delay system. By employing a state-dependent integral function to be a Lyapunov function candidate, the error system is proved to be asymptotically stable. The observer gain, the controller gain and the event parameters in the event condition can be co-designed and obtained in terms of solution to a set of linear matrix inequalities (LMIs). Finally, a numerical example and the tunnel diode circuit model are shown that the proposed method is effective, and the simulation results can reflect that the event-triggered scheme can lead to a larger release period than time-triggering scheme.     

Switched Second‐Order Sliding Mode Control with Partial Information: Theory and Application

This paper proposes a novel switched second order sliding mode (S‐SOSM) control strategy in a partial information setting, i.e., when only the sliding variable is accessible for measurements. Such a control approach allows one to deal with systems characterised by different levels of uncertainties associated with different regions of the state space and to accommodate different control objectives in the different regions by switching among appropriate SOSM controllers. The proposed approach is shown to ensure global finite‐time convergence to the origin of the closed‐loop system trajectory. The braking control of two‐wheeled vehicles is considered as a case‐study to test the controller performance.     

Performance adaptive control General structure and a case study

This paper proposes a supervisory performance tuner for an adaptive controller. The objective of the performance tuner is to automatically adjust tuning parameters of an adaptive controller such that the actual setpoint-tracking and regulatory performance approaches user specifications and is maintained at this level at all times. Experimental evaluation of the prototype performance tuning loop with adaptive generalized predictive and pole placement controllers on a pilot-scale process demonstrates the practicality and utility of this idea.     

Observer‐based H∞ control for systems with repeated scalar nonlinearities and multiple packet losses

This paper is concerned with the H_∞ control problem for a class of systems with repeated scalar nonlinearities and multiple missing measurements. The nonlinear system is described by a discrete‐time state equation involving a repeated scalar nonlinearity, which typically appears in recurrent neural networks. The measurement missing phenomenon is assumed to occur, simultaneously, in the communication channels from the sensor to the controller and from the controller to the actuator, where the missing probability for each sensor/actuator is governed by an individual random variable satisfying a certain probabilistic distribution in the interval [0 1]. Attention is focused on the analysis and design of an observer‐based feedback controller such that the closed‐loop control system is stochastically stable and preserves a guaranteed H_∞ performance. Sufficient conditions are obtained for the existence of admissible controllers. It is shown that the controller design problem under consideration is solvable if certain linear matrix inequalities (LMIs) are feasible. Three examples are provided to illustrate the effectiveness of the developed theoretical results. Copyright © 2009 John Wiley & Sons, Ltd.