Download full-text


Available from: Steffen Borchers, Sep 27, 2015
28 Reads
  • [Show abstract] [Hide abstract]
    ABSTRACT: Effective fault diagnosis depends on the detectability of the faults in the measurements, which can be improved by a suitable input signal. This article presents a deterministic method for computing the set of inputs that guarantee fault diagnosis, referred to as separating inputs. The process of interest is described, under nominal and various faulty conditions, by linear discrete-time models subject to bounded process and measurement noise. It is shown that the set of separating inputs can be efficiently computed in terms of the complement of one or several zonotopes, depending on the number of fault models. In practice, it is essential to choose elements from this set that are minimally harmful with respect to other control objectives. It is shown that this can be done efficiently through the solution of a mixed-integer quadratic program. The method is demonstrated for a numerical example.
    American Control Conference (ACC), 2013; 06/2013
  • [Show abstract] [Hide abstract]
    ABSTRACT: In this work we focus on unique diagnosability of parametric faults in the presence of measurement uncertainty and model mismatches. Specifically, we formulate a condition for diagnosability of parametric faults in a set-based framework that allows for direct consideration of uncertainty. Based on this condition we present an approach for the analysis and certification of diagnosability. Furthermore, we propose an approach for the redesign of initially given fault classifications in the parameter space. Specifically we compute diagnosable subsets of initially given parameter sets in polynomial discrete-time fault candidates by comparing pairs of fault candidates. Furthermore, we demonstrate the presented approach for a numerical example.
    2013 IEEE 52nd Annual Conference on Decision and Control (CDC); 12/2013
  • [Show abstract] [Hide abstract]
    ABSTRACT: In this article, we consider the problem of parameter identification and state estimation of an uncertain continuous-time linearly parameterized nonlinear system with output dependent nonlinearities subject to exogenous disturbances. A set-based adaptive estimation is proposed in which the parameters and the states of the system are estimated along with an uncertainty set guaranteed to contain the true unknown values. The set-update approach is such that the sets are updated only when an improvement in the precision of the parameter estimates and the state estimates can be guaranteed. The formulation provides robustness to parameter estimation error and bounded disturbances. The adaptive estimation technique can be viewed as an adaptive interval observer. Simulation examples are used to illustrate the effectiveness of the developed procedure and ascertain the theoretical results.
    The Canadian Journal of Chemical Engineering 08/2014; 92(8). DOI:10.1002/cjce.22001 · 1.23 Impact Factor