The Evolving Dynamics Of Distributed Modeled Product Systems

As the current popular design methodology, the advanced distributed modeleddesign methodology is generally applied to the development of engineering systems.A system designed by the methodology is called the distributed modeled systemwhich is composed of many basic centralized models that are distributed andassociated with each other. But in the area of engineering, the distributed modeleddesign methodology makes much more use of the formalized-language description inthe process of the real system design. What phenomena and laws in mathematics dodistributed modeled systems include in their architectures? It is still a puzzling butinteresting problem. Thus, in areas of machine and information automation, theturbine systems provided by Hangzhou Turbine Stock Corporation (HTC) and Javasoftware systems provided by SUN and IBM as typical distributed modeled systemsare selected to study in this paper. This paper will strictly comply with the generalresearch way as follows: Find phenomena from practice, and then reveal essentialmathematical laws according to phenomena, and finally put laws into practice.Nowadays, it has been becoming a research hotspot in systems science to studythe evolving dynamics and network features of complex systems based on the theoryof complex networks. The theory of complex networks originated from thecombination of graph theory and probability statistics studies dynamics systemscontrolled by all kinds of mechanisms in the world. Up to now, there have been fewcontributions on distributed modeled product systems. In the process of studying thesmall-world and degree-distribution features of real systems, this paper based on thetheory of complex networks tries to discover the general phenomena andmathematical laws held by distributed modeled product systems on their uchitecturesare discovered, and their othernesses. The contents of this paper are mainly asfollows:Chapter 1 simply introduces all research work in this paper.Chapter 2 introduces the principles of distributed modeled designs and thearchitecture of real distributed modeled product systems that are abstracted intonetworks.Chapter 3 investigates the small-world features of real distributed modeledproduct systems.Chapter 4 demonstrates the features of in-degree, out-degree and degreedistributions of real distributed modeled product systems. Chapter 5 studies the phenomenon that preferential and random mechanismscoexist in real systems.Chapter 6 studies the phenomenon of "accelerating connections" in realdistributed modeled product systems.Chapter 7 studies some other extended mechanisms in real systems.Chapter 8 studies the "rich gets richer" phenomenon of scale-free networks.Chapter 9 discusses the mechanism that evolving models control the diversifieddesign principle and the method of reducing the internal diversification.Chapter 10 makes a summary and prospect.In this paper, the simulation and application to theory results are made, and therelation between theory in this paper and the distributed modeled design methodologyare also discussed. The study of this paper indicates that certain principles and theircriterion contribute to constructing systems that bear certain network laws inengineering areas.In this paper, some new phenomena and laws of distributed modeled productsystems are revealed as follows:●Demonstrations indicate that distributed modeled product systems havesmall-world and scale-free features. For in-degree and out-degreedistributions, it is always that one follows the power distribution but the otherasymptotically follows the exponential distribution, while the reuse anddistributing technology play an important role respectively.●Real distributed modeled product systems are controlled together bypreferential and random mechanisms.●There is the phenomenon of "accelerating connections" in distributedmodeled product systems. It is require of adjusting system's collaborationperformance. It contributes to the design principle of "realizing the externaldiversification as possible as much with the internal diversification aspossible as few".●There are local events and the phenomena of combining decentralization withcentralization in real product systems.●In theory, any scale-free network with the phenomenon of "rich gets richer"has a certain point (area) at the poor and rich interval that is controlled byevolving mechanisms of systems.