Quantitative model-based failure detection and isolation (FDI) methods rely on the comparison of a system’s available measurements, with a-priori information represented by the system’s mathematical model. The term quantitative is used here as contrary to the term qualitative, denoting cruder sustem descriptions. There are two main trends of this approach, namely analytical redundancy or residual-generation methods and parameter estimation. This distinction is not universally adopted and some researchers consider these two approaches as belonging to the same category. However, for reasons of clearer presentation the distinction is adopted here, and parameter estimation methods will be presented in Chapter 3.

Analytical Redundancy Methods

A Short Description of a Model Compiler/Interpreter for Supporting Simulation and Optimization of Nonlinear and Linearized Dynamic Systems

Tool Integration by Way of a C omputer- A ided M echatronic L aboratory (CaMeL)

Coach active suspension system design must meet requirements for comfort and safety, and technical requirements. A hydropneumatic system was taken to be the most appropriate spring system. In a hydropneumatic strut, spring and damper elements are replaced by hydraulic ones, the flat spiral spring being replaced by a hydraulic accumulator and an actuator with the damper being replaced by a throttle and an actuator. The stabilizer is abolished altogether. After the statical properties were established a dynamical model of the entire coach, including active components, was made up helping to verify further dynamical loads and demands relating to control engineering. The article describes the design process.

Development of an active suspension system for a Mercedes-Benz coach (O404)

The problems of designing mechatronic systems (e.g. heterogeneous system components and their design methods, complexity) cause the need for an open, integrated design environment. This paper details the process-oriented approach of the software environment CAMeL (Computer-Aided Mechatronics Laboratory) to the integration of different tools and methods. Characteristic concepts and implementation features of experiments in this "laboratory in the computer" are presented. An example from the field of vehicle dynamics (vehicle convoy control) demonstrates typical experiments in the modelling, analysis and synthesis phases of the design cycle of mechatronic systems.

Integration of mechatronic design tools with CAMeL, exemplified by vehicle convoy control design

A design environment specially tailored to problems in control engineering is presented. It contains, apart from plant models and operating environments, the formulation of the experimental environment. Linear controllers of any desired structure are designed with the aid of the characteristic variables of the experiments, using a vector optimization procedure in an interactive design process. The hierarchical description of the model, based on an interpreter solution, was realized in the ADA programming language. A complex problem in controller design, taken from the field of vehicle suspension, proves the efficiency and performance of the design tool.

CACE tool for multi-input, multi-output systems using a new vector optimization method

We study semicontinuous maps on varieties of modules over finite-dimensional algebras. We prove that truncated Euler maps are upper or lower semicontinuous. This implies that $g$-vectors and $E$-invariants of modules are upper semicontinuous. We also discuss inequalities of generic values of some upper semicontinuous maps.

J. Schroer

Papers

CACE tool for multi-input, multi-output systems using a new vector optimization method

Semicontinuous maps on module varieties