期刊名称：Journal of Theoretical and Applied Information Technology
印刷版ISSN：1992-8645
电子版ISSN：1817-3195
出版年度：2021
卷号：99
期号：18
语种：English
出版社：Journal of Theoretical and Applied
摘要：Nowadays, the control of physical systems is the result of many steps, from design to implementation, modelling and analysis, Especially in the field of education and innovation. It is often necessary to know in advance the performance of the system under study and in this case a mathematical model is more than useful. For economic and performance reasons, it is often necessary in certain situations to have a precise representation of physical phenomena, which leads to complex models and the mathematical tools are not always adapted to the models obtained. For several decades, the most exploited representation has probably been the state representation, or an extension of this form for non-linear models. For some time now, the algebraic approach has been used to discover other characteristics of the models, or at least a new interpretation of properties well known to automaticians. However, these techniques are difficult to use for the uninitiated, and the representation by a link graph model allows to reconcile all the theoretical concepts, which are necessary in the different design phases. Some analysis techniques for design are proposed by bond graph modelling for linear models, but can be quite easily generalised for more general models. In this paper, the bond graph tool proves its intelligence in the implementation of monitoring systems, and in particular an intelligent design support. The causal properties of the bond graph allow to analyse the necessary monitoring conditions before and after the design to generate diagnostic algorithms in a generic way. The following approach is illustrated on a DC motor.
