Reliability Analysis And Optimization Of Bridge Network System

With the advancement of society and the development of science and technology, modern industrial system is developing to several aspects just as function synthesis, equipment diversification and information integration. Consequently it results in that system dimensions is being bigger and bigger and system framework is being more complex. Firstly, to analyze such huge and complex system to maintain its normal operation and reduce its trouble rate, it is necessary to scientifically analyze its reliability to satisfy the realization of system function. Secondly, establishing a modern huge system needs a great deal of investments, so it must be based on scientific reliability argumentation that the building and operation of the costly system.One of the primary goals of reliability research is seeking the best method to enhance the system reliability. Reliability optimization can help to achieve this aim. A lot of important systems in practice are complex network systems. This paper mostly researches in the system reliability analysis method of bridge network system, a kind of typical complex system. Moreover, it researches in that employing adaptive genetic algorithm (AGA) to solve the reliability optimization problem of bridge network system.For bridge network system, some methods, such as the binomial expanding method, the network system decompounding method and the minimum circuit method, are used to analyze the system reliability in this paper. The excellences and disadvantages of the above methods are also discussed. Then, the theory, principle, method and characteristic of genetic algorithm (GA) are introduced. Moreover, it explains how to mend GA to improve its performance. AGA, which employ adaptive strategy on cross probability and mutation probability, is introduced as emphases. In the simulating example, the optimization object in the case is to make the investment least so as to obtain supreme economy benefit under certain requirement of system reliability index. GA is used to optimize the indexes of the formed nonlinear mathematical model. In the meantime, effective ranking selection strategy, greatest unit conserved strategy, adaptive technique to crossover operator and mutation operator are adopted to improve the performance of GA largely.In comparison with other optimization methods, AGA is proved to be more fit to solve the reliability problem of the complex system. It also shows that AGA have powerful ability to deal with very complicated nonlinear object programming problem in the practice.