This paper presents a novel knowledge-based Petri net
approach to mechanical systems and assemblies modeling
within a design with objects environment. A new unified
class of object-oriented knowledge Petri nets, which can
incorporate a knowledge-based system with ordinary Petri
nets, is defined and used for the unified representations
of assembly design and modeling. The object knowledge Petri
nets, as a graphical language and a new knowledge-based
description scheme, can be used to express the qualitative
and quantitative aspects of the assembly design and modeling
process in an interactive and integrated way. The four-level
hierarchy model is proposed and constructed in terms of
function-behaviors, structures, geometries, and features.
The function-behavior-structure description is built on
more abstract concepts so that it can match well top-down
design. The static and dynamic characteristics in the design
of assembly can also be captured. With the help of fuzzy
logic, the incomplete, imprecise knowledge and uncertainty
in the design process can also be dealt with. Therefore,
the hybrid design object model can incorporate product
data model, top-down design process, and assembly process
model using an object-oriented, knowledge-based, feature-based,
parametric, and constraint-based modeling approach, and
can provide a more accurate and more flexible representation.
To verify and demonstrate the effective use of the proposed
hybrid design object model, a prototype system has been
developed. This research provides a knowledge-intensive
framework for intelligent assembly design and modeling.
Reliability Evaluation for Mechanical Systems by Petri Nets
