مروري بر روشهاي تحليل اعتمادپذيري شبكه هاي درون شهري شريانهاي حياتي

A Review on the Method of Reliability Analysis of Networks with Emphasis on Lifeline Network Performance after Earthquake

افزايش اعتمادپذيري شبكه ها مي تواند خسارات وارده به سيستم ها و شبكه ها را پس از بروز حوادثي مانند زلزله كاهش دهد. شناخت شبكه ها و روشهاي تحليل اعتمادپذيري امري ضروري و از اهداف اين مطالعه است. يكي از مشخصه هاي بارز شبكه ها كه آنها را داراي ساختاري ممتاز مي نمايد ارتباط بين هر دو گره از شبكه است كه مي تواند با مسيرهاي زياد و مختلفي تامين شود و اين ارتباط را قابل اعتماد سازد. بدترين حالت زماني است كه با شكست مجموعه اي از مولفه هاي مشخص، شبكه دچار شكست مي شود. در بسياري از تحليل هاي اعتمادپذيري و بهينه سازي مدل ها، دو حالته بودن (داير و غيرداير) مولفه ها فرض مي شود كه در حالت واقعي چنين فرضي مي تواند دور از واقعيت باشد. در اين مقاله، ضمن معرفي گرافها و انواع سيستم ها، دياگرام هاي تصميم دودويي تشريح مي شوند. همچنين شبكه هاي پيچيده مانند شبكه هاي تصادفي و آزاد مقياس معرفي و تحليل آنها مورد بحث قرار مي گيرد. با توجه به گستردگي برخي شبكه ها بايد از روشهايي استفاده كرد كه ضمن دقت كافي، هزينه محاسباتي را نيز كاهش دهد. در اين ميان همان-طور كه در بخش تحليل شبكه هاي پيچيده تشريح مي گردد مي توان به دو روش موسوم به گراف تصادفي و آزاد مقياس اشاره نمود، اگرچه روش گراف تصادفي داراي محبوبيت بيشتري است.

Sufficient knowledge about the method of analyzing and reliability of network is essential to estimate the damage of systems and networks after earthquake. So, the method of reliability analysis networks is important and it is one of the aims of this study. A special characteristic of the current organization of the human life is the growing level of the interconnectivity. A tendency to increase in dimension and to aggregate in complex interconnected systems exists in economical, social, and technological structures. One important property of networks is that the connection between any two nodes of the networks can be typically reached through a number of redundant paths, thus making the connection basically reliable, and it makes them an advantaged structure both in natural and technological systems. Therefore, the reliability of networks has always been a main alarm since the earliest times of reliability engineering. Since the early 60s the exponential growth of the size and number of networks of all types has led many researchers to study the various characteristics, the topology, and the reliability of the networks. Now, the analysis of the network reliability is a major concern for the formation, validation and maintenance of many real systems such as communication, computer or electrical networks. The reliability of these complex systems is of increasing importance since the failure of some components may lead to catastrophic results.The failure of definite sets of components in the worst case may cause the failure of the network itself. In many real-life states, components and/or systems can be in one of more than two states (up and down). Systems that are characterized by different levels of performance are known as multi-state system (MSS).Power systems and computer systems are examples of MSS where the performance of the elements is characterized by generating capacity and data processing speed respectively. Most of reliability analysis and optimization models suppose that system is made of binary-state components, where the states are functioning and failure. Binary-state reliability models don’t allow to satisfactorily describing some aspects of the MSS as MSS operation processes, remaining resources, system state evolution, reasons of system failures and mechanisms of their prevention.The main difficulty in the MSS reliability analysis is the “dimension damnation” since each system element can have many different states(not only two states as the binary-state system). In this study, graphs, systems and binary decision diagram are introduced. Complex networks are described. Also, the specific properties of random networks and scale free networks are illustrated. Finally, weighted multi-state networks have been explained.