Dynamic Random Network In The Short-circuit

It assumes that the weight of the arcs in traditional network is static and a determinate number, which is not true in many fields such as intelligent transportation system and computer network and communication fields. The shortest path problems in variable-time and stochastic network break through the limit of traditional shortest path problems and become foundation theory in ITS. The new real problems make the shortest path computing to be more difficult.In this paper, compared with the new problem and the old problems existing in the former research, we mainly focus on the research of dynamic shortest path in a variable-time and stochastic network. Our work is divided in to three parts:(1) Summarize the former research result, and indicate the existing problems.(2) For the random network which's weights are continuous functions of time ,we discuss on the relation about the shortest path at a moment and the shortest path of the moment's neighborhood and get the solution for the shortest path from the start node to the end node at a moment using the limited coverage theorem .When solving the dynamic shortest path at a specific time in a stochastic network which's weights are continuous functions of time t, we divide the closed interval into a finite number of tiny proper interval, transform the continuous function into discrete functions using expectation.(3) Based on the Ant Colony Algorithm, we design a new algorithm for the shortest path problems in variable-time and stochastic network. This algorithm is proved to be an effective solution to the real-time shortest path in stochastic network.The study of this paper has important theoretical and practical values. The result of the research can be applied in many fields, such as Intelligent Transportation System (ITS), computer network and communication, etc.