Band 61, Heft 2 -

The increasing automation and the resulting complexity require to build distributed control architectures. The complexity of distributed control architectures requires methods which facilitate the developement. The aim of the approach presented here is a developer support through a combination of notation, attributes, and design patterns across multiple levels of an adapted process model.

This article presents an approach to improve the engineering of automated facilities by combining process and plant description. Through the integration of a semi-formal process description language in the well known Microsoft Visio environment and the possibility to transfer the results into AutomationML it is aimed for maximum acceptance. Based on the combination of different engineering models an integrated requirements management process is possible.

The process from design up to operation of a typical chemical plant takes about 5 to 10 years with state-of-the-art technology. Recurring problems should be solved by using standardized modules to reduce the engineering time, become more efficient and to reduce the costs. The case-based reasoning follows the approach of comparing the actual case with similar cases from past experiences to achieve the solution faster.

The development of mechatronic systems is a multidisciplinary task. Common models shall provide a common understanding of the later system in the early design phase. This paper shows how the used models can be abstracted and generalized, which enables an automated consistency check. Additionally, the structure of the rules, necessary for that consistency check, is presented.

For a successful using of model based methods it is necessary to ensure the semantic correctness of the relevant models regarding the specified requirements. The quality of models is an essential parameter for the quality of model based methods. In this paper a method for model analysis is introduced. Establish methods of constraint programming are combined with Petri net techniques in order to analyse open, reactive and event driven system models.

The goal of fault-tolerant control is to influence a faulty process in such a way that it still fulfills a given specification. This paper presents the design of such controllers for systems modeled by nondeterministic input/output automata. Forward and backwards reconfiguration methods are used depending on the fault identification time step. The controller re-design includes an adjustment of its set trajectory and of its input and output signals to counteract the effects of the fault. A simple manufacturing process is used to illustrate the proposed approach.

This paper describes the development of a component library for modeling and simulation of networked automation systems (NAS) based on the OOAD approach (object-oriented analysis and design). During analysis typical properties of NAS are described. The design respects specific constraints of the used modeling language Modelica. The library is realized for the tool Dymola.