Optimize The Route Of Petri Net

When the transitions in Petri Net are bounded up with a delay, we call this kind of Petri Net is timed Petri net (TPN). In TPN system the way which connects the input place and the output place and the total delay is shorter than any other ways is called Backbone. It's the most efficient path in TPN system, and determines the system capability in a way. In the area of project we must study a system by studying the Backbone. Petri Net is not sequence or branch structure but with the structure of sequence, branch, side-by-side and circle, and with weight, conflict and cooperation characters. So the Backbone seemed more difficult to find than any other nets.This article will research timed Petri Net with the clew of finding Backbone.At the beginning, we analyze a Petri Net with fixed delay and explain the reason that choices the Ant Density Algorithm. We talk about and resolve the problems of clear up circle ways, speed correspond, choice strategy, update odor elements and update place elements. And then we show the algorithm of finding Backbone .At last we simulate it by program to prove the correctness.The Petri Net with fixed delay is a kind of ideal net, and the resolve is also ideal too. The transitions are always stochastic and many of transitions are obey the exponential distribution we will research the search strategy in SPN.For SPN, we analyze the distribution rule of every transition. After that we design the Ant Density Algorithm based on the result of analyse.Bring forward the new data structure and deal with the problems above mentioned, and then we give the whole algorithm of finding Backbone in SPN.After this problem, another problem occurs. If the path with shortest delay and low dependability is considered as the Backbone, the efficiency of Petri system will be played down. Therefore, we use the model of Weibull. We combine the algorithm above mentioned and the Weibull model designed a new algorithm. That can make the way of ant appear at the way with best benefit. This can improve the usage area of Ant Density Algorithm .In this part, we suggest a way to use this algorithm more often, which can be funded by scrupulous one.After we use the Petri Net with memory characteristic, we find that SPN is one kind of Petri Net with memory. At last, we describe that kind of Petri Net completely and research many performance parameters, to wide the way of usage of the Petri Net with memory.