透過您的圖書館登入
IP:18.222.0.213
  • 學位論文

智慧型保修系統之排程系統

Application-level Scheduling System for Intelligent Equipment Maintenance System

指導教授 : 鍾添曜
若您是本文的作者,可授權文章由華藝線上圖書館中協助推廣。

摘要


智慧型保修系統為超高壓電源武器裝置提供了一套監測及預警系統。前端的監測系統使用安置於武器上的感應器來擷取武器系統的即時狀態,並當下做出系統狀態的判斷,達到即時監測的目的。此外,智慧型保修系統還會利用網路通訊將資料傳回資料庫。後端的故障預警決策系統會定期的擷取資料庫中的資料,然後做故障預警分析。故障預警分析可以得到系統發生問題時各個數據之間的關係模式,一旦察覺到數據間再次發生相同的模式時,本系統將會對維修人員發出警告。維修人員就能提早對該武器系統維修或檢測,如此可以提升系統的可靠性及完善率。 本論文的目的是解決智慧型保修系統中HP之類比數位轉換器所產生的兩個問題。第一是提供太多取樣樣本而導致網路頻寬需求變高,資料庫也無法負荷如此大量的資料。第二是同一部類比數位轉換器中的感應器會得到相同的取樣頻率,無法根據感應器監視的目標特性來設定不同的取樣頻率。於是我們設計了一個軟式取樣系統和應用程式層級排程系統(Application-level Scheduling)來解決這兩個問題。其中,排程系統會視資料串流的重要性來做為排程的依據,使得武器監測器與資料庫能有效運用系統資源。

並列摘要


A key element in an Intelligent Equipment Maintenance system(IEMS)is to monitor system parameters of an equipment in real-time. With real-time system parameter samples, IEMS can help operators in diagnosing equipment before severe damage or system break down occur. Real-time monitoring, however, generates huge sample data that must be saved in database for later processing. This thesis proposes an application-level scheduling scheme, called significance-based proportional service(SPS), to effective allocate system resource, such as network bandwidth, so that high priority sample data can achieve high throughput. SPS is simple and suits well for a variable rate server. It decouples data significance from bandwidth. Extensive simulation has been performed to study the behavior of SPS. Also, we compare the performance of SPS with that of WRR and PDS. The simulation results show that SPS outperform WRR in data loss rate reduction for more significant data flow and outperform PDS in ill-behaved traffic prevention.

參考文獻


[2] V. Jacobson, “Congestion Avoidance and Control,” Proceedings of ACM SIGCOM’88, Stanford, CA, pp. 314-329, Aug. 1998.
[3] W. R. Stevens, TCP/IP Illustrated, vol 1. Addison-Wesley Publishing Company, 1994.
[4] M. Shreedhar and G. Varghese, “Efficient Fair Queuing using Deficit Round Robin,” Proc. Of ACM SIGCOM’95, pp.231-242, 1995.
[5] A. Demers, S. Keshav, and S. Shenker. Analysis and Simulation of a Fair Queueing Algorithm. In Proceedings of ACM SIGCOM, pp. 1-12, Sep. 1989.
[6] A. K. Parekh and R. G. Gallager, “A generalized processor sharing approach to flow control in integrated services networks: the single node case,” IEEE/ACM Trans. Network, vol. 1, no. 3, pp. 344-357, Jun. 1993.

延伸閱讀