Research On Predicted Computing Of Oil Spill Based On Mobile Agent

Mobile agent is the intelligent agent having characteristic of mobility, which can migrate from one computer to another in an isomerous network and complete appointed task of users such as computing, searching , filtrating and gathering information, so much as commercial activity. Mobile agent has the characteristic of migration in addition to response, autonomy and intelligence. These characteristics make mobile agent have the good forecast of application, such as software designings dynamic load balancing of network, distributed computing and so on.In this paper, a method of applying mobile agent to predicted comput--ing of oil spill is put forward. The existing predicted model of oil spill (such as the predicted model of oil spill of Chinese north sea area) is ripe in substance and has been validated in practice, but it has some deficiency in computing speed. Thus this article designs the predicted model of oil spill based on mobile agent. In this model, The task of prediction of oil spill is divided into several parts and every part will be handed to a mobile agent which can migrates to another computer to complete the task. Finally, the outcome will be returned.JAVA is adopted to realize the model on Aglet platform, because Aglet platform is the most important platform of mobile agent, which provides a simple and all-around model of mobile agent, dynamic and valid mechanism of communication and a series of detailed and convenient mechanisms of security for mobile agent and expends less resource. Simulant experiment is executed under the circumstance of LAN and compared with the predicted model of oil spill of Chinese northsea area, and the speed of computing is found improved largely.At the end of the article, the whole work is summed up, which points out that mobile agent has the forecast of broad application and give some suggestions for father exploitation and research.