Torsten Jeinsch

A unified approach to the optimization of fault detection systems

In this paper, problems of optimizing observer-based fault detection (FD) systems in the sense of increasing the robustness to the unknown inputs and simultaneously enhancing the sensitivity to the faults are studied. The core of the study is the development of an approach that simultaneously solves four optimization problems. Different algorithms are derived for the application of this approach to the optimal selection of post-filters as well as optimization of fault detection filters, and to the systems with and without structure constraints. The achieved results also reveal some interesting relationships among the optimization problems considered. Copyright

Feedback Control Structures, Embedded Residual Signals, and Feedback Control Schemes With an Integrated Residual Access

Driven by the increasing needs for the integration of model-based fault diagnosis into the electronic control units (ECUs) with limited computation capacity and motivated by the recent study on the fault tolerant controller architecture, we investigate feedback controller structures aiming at accessing the residuals embedded in the control loops. For this purpose, we first develop an observer-based realization of the Youla parameterization. This result reveals a new interpretation of control signals as a composite of the residual and reference signals. From this viewpoint, different control schemes are studied and useful relationships between the controller structures and embedded residual signals are established. It leads to the development of two kinds of schemes: 1) extracting residual signals from an existing control loop and 2) configuring control loops with an integrated residual access. The achieved results are demonstrated by two examples of the feedback control loops in engine management systems.

Quality-Related Fault Detection in Industrial Multimode Dynamic Processes

Multivariate statistical process monitoring (MSPM) methods are powerful tools for detecting faults in industrial systems. However, industrial processes are often subjected to dynamic changes. This dynamic behavior is mainly due to set-point changes and nonlinearities. Because of the nonlinearity of processes, the performance of the classical MSPM methods, which are mainly based on the linearity assumption, becomes unsatisfactory, since the process characteristics will change from one operating point to another. The main objective of the work is to develop an efficient fault detection technique for complex industrial systems, using process historical data and considering the nonlinear behavior of the process. In the proposed approach, the nonlinear system is assumed to be linear around the operating points and therefore considered as a piecewise linear system corresponding to each operating mode. The performance and effectiveness of this approach are demonstrated using data obtained from a paper machine and compared with an available method.

A survey of the application of basic data-driven and model-based methods in process monitoring and fault diagnosis

Abstract Basic data-driven and model-based process monitoring and fault diagnosis methods are surveyed from the application viewpoint. The main objective is to study the needed modifications and/or combined use of these methods under different real operating conditions.

An Approach to Analysis and Design of Observer and Parity Relation Based FDI Systems

Abstract This contribution deals with problems related to observer and parity relation based fault detection and isolation (FDI). New relationships among the design parameters of parity space, observer based and factorization approaches are established. Using them a unified approach is proposed, which can be used to solve some typical problems met by designing parity relation and observer based residual generators.

Fault detection system design based on a new trade-off strategy

Problems related to fault detection in dynamic systems with unknown inputs are studied. Instead of designing fault detection systems from the viewpoint of increasing the system robustness against unknown inputs and the sensitivity to the faults, an approach is proposed, which allows us to design fault detection systems based on a trade-off between the false alarm rate and the missed detection rate. A further study on the relationships between the approach proposed and the existing H/sub 2/, H/sub /spl infin///spl infin// and H/sub /spl infin//min/ optimization approaches demonstrates that the approach proposed in the paper provides us not only with a unified solution to the existing approaches but also with the best solution among these approaches in view of minimizing the missed detection rate under a given false alarm rate.

An LMI approach to robust fault detection filter design for discrete-time systems with model uncertainty

Deals with the design of a fault detection filter for discrete-time systems with both model uncertainty and disturbances. We propose an approach to the solution, which consists of two steps: (a) selection of a stable weighting function matrix, optimized in the sense of the maximum sensitivity from the faults to residual signal; (b) formulation of the design of fault detection filters as a model-matching problem and solving the optimization problem using the LMI technique. The achieved results are illustrated by a numerical example.

A numerical approach to optimization of FDI systems

Following the idea of integrated design of residual generation and evaluation, a numerical approach to time domain optimization of observer based fault detection and isolation (FDI) systems is proposed. The discussion on interconnections among different FDI approaches shows that the proposed method can also be used to approach the optimization problem for other kinds of FDI systems.

A numerical approach to analysis and design of FDI systems

This contribution deals with problems related to fault detection and isolation (FDI). A numerical approach is proposed, which can be used, for instance, for the design of low-order FDI systems or for solving robust FDI problem.

LMI-BASED INTEGRATION OF ROBUST H∞-CONTROL AND RFD FOR LTI SYSTEMS

Abstract In this contribution, problems related to the integrated design of observer-based robust H∞ controller and robust fault detection (RFD) system is studied for uncertain LTI systems with both modelling errors and unknown input. We first propose to use an ideal solution, which is in fact an optimization solution of fault detection filter achieved on the assumption of the system model is perfectly known (i.e. without modelling errors), as the reference model of robust fault detection filter design for uncertain systems. The integrated design can then be formulated as a two-objective H∞–optimization problem which is solved using LMI techniques. The basic idea behind our study is that the robust H∞ controller share a common observer with the fault detection filter; by introducing a reference residual model, the feedback controller and robust fault detection filter can be designed together via solving a two-objective H∞–optimization problem; finally an LMI approach to integrated design is developed. An example is also presented to illustrate the proposed approach.