Design Evolution of Mechatronic Systems

SUMMARY This chapter presents a system framework for design evolution of mul­ tidomain engineering systems or mechatronic systems. The described framework inte­ grates machine health monitoring with an expert system to monitor the performance of an existing mechatronic system and to make a decision as to whether a design improve­ ment is necessary. If a design improvement is necessary, the expert system will also assist in identifying the section of the system that needs to be improved. Modeling of multi­ domain engineering systems using bond graphs and linear graphs is explained in detail with examples. A technique to evolve a system model represented by a bond graph or a linear graph using genetic programming is introduced. This technique is combined with a fitness function that represents a set of desired performance specifications to obtain an optimum design. The developed approach allows exploration of the design space to find the optimum design solution in an automated manner. In order to control the arbi­ trary exploration of the design space, domain knowledge, expertise, and input from the

CONTENTS

10.1 Introduction ........................................................................................................................360 10.2 Modeling Multidomain Systems ..................................................................................... 361

10.2.1 Bond Graph Modeling ........................................................................................... 361 10.2.1.1 Causality Assignment ............................................................................363 10.2.1.2 State­Space Formulation .........................................................................365