Gernot Schullerus

Brief paper: Input signal design for identification of max-plus-linear systems

The present contribution addresses the problem of designing an adequate persistent excitation for state space identification of max-plus-linear systems. The persistent excitation is designed using the same techniques that have recently been developed for model predictive control for max-plus-linear systems. The application of this method for input signal design allows to incorporate additional objectives which are desirable for the input signals and the resulting process behaviour such that an optimal persistent excitation is obtained.

A method for estimating the holding times in timed event graphs

This contribution addresses the estimation problem of holding times in timed event graphs. Based on the known Petri net topology the holding times are determined such that an equation error is minimized. The paper investigates the important issue of convergence of the estimated parameters to the true parameter values and shows that this objective can be achieved by applying particularly chosen input signals. A new iterative estimation procedure is proposed using these suitably designed input signals.

Input Signal Design for Discrete Event Model Based Batch Process Diagnosis

Abstract This paper presents a diagnosis approach suitable for batch processes with discrete sensors and actors. The evolution of event times will be modeled using the max-plus algebra. Based on this model, the presented diagnosis algorithm detects the faulty component comparing estimated process parameters with their corresponding nominal values. The focus is first on the influence of input signals on the diagnosis performance and second on an input signal design method for reliable diagnosis results.

Adaptive model predictive control using max-plus-linear input-output models

Model predictive control (MPC) is a popular controller design technique in the process industry. Conventional MPC uses linear or nonlinear discrete-time models. Recently, we have extended MPC to a class of discrete event systems that can be described by a model that is “linear” in the max-plus algebra. In our previous work we have considered MPC for the time-invariant case. In this paper we consider an adaptive scheme for the time-varying case, based on parameter estimation of input-output models. In a simulation example we show that the combined parameter-estimation/MPC algorithm gives a good closed-loop behaviour.

Relations of Timed Event Graphs and Timed Automata in Fault Diagnosis

Abstract This paper investigates the relations between timed event graphs and timed automata that describe discrete-event systems subject to faults. An algorithm is presented for transforming a timed event graph into an equivalent timed automaton. The relations among the models of the faulty systems provide the basis for comparing diagnostic methods that have been developed in the past for these two timed discrete models.

Modelling and hierarchical diagnosis of timed discrete-event systems

A new diagnostic method for hierarchically structured discrete-event systems is presented. The efficiency of this method results from the fact that the complexity of the diagnostic task is reduced by first detecting a faulty component using a coarse process model on a high level of abstraction, and subsequently refining the result by investigating the faulty component with the help of a detailed component model in order to identify the fault with sufficient precision. On both abstraction levels, the method uses a timed discrete-event model of the appropriate part of the system. On the higher abstraction level a timed event graph is used that describes how the temporal distance of the events is changed by component faults. On the lower level of abstraction, timed automata are used to cope with the non-determinism of the event sequence generated by the faulty and faultless components. The approach is illustrated by the diagnosis of a batch process.

On the use of hybrid system identification methods for max-plus-linear system identification

Abstract In recent years, different approaches for identification of hybrid systems and max-plus linear systems that are a special class of discrete event systems, have been developed. This paper introduces a new identification method for piecewise affine hybrid systems and shows that this approach is also easily applicable for identification of max-plus-linear systems.

Diagnose zeitbewerteter ereignisdiskreter Systeme (Diagnosis of Timed Discrete Event Systems)

Abstract Dieser Beitrag befasst sich mit der Diagnose von Prozessen, die als zeitbewertete ereignisdiskrete Systeme modelliert werden. Für die Diagnose dieser Systeme werden zwei Verfahren vorgestellt, die auf zeitbewerteten Synchronisationsgrafen bzw. auf zeitbewerteten Automaten beruhen. Beide Verfahren erkennen und lokalisieren Fehler durch Auswertung der Ereignisfolge und der gemessenen Ereigniszeitpunkte.