Journal of Systems Engineering and Electronics ›› 2018, Vol. 29 ›› Issue (6): 1308-1316.doi: 10.21629/JSEE.2018.06.18

• Reliability • Previous Articles     Next Articles

Evidence combination method in time domain based on reliability and importance

Chengkun LUO*(), Yunxiang CHEN(), Huachun XIANG(), Weijia WANG(), Zezhou WANG()   

  • Received:2017-09-26 Online:2018-12-25 Published:2018-12-26
  • Contact: Chengkun LUO E-mail:afeulck@163.com;cyx87793@163.com;32134717@qq.com;visionwng@foxmail.com;350276267@qq.com
  • About author:LUO Chengkun was born in 1990. He received his B.S. degree of management engineering in 2013 and M.S. degree of management science and engineering in 2015 from Air Force Engineering University. Now he is a Ph.D. candidate of Equipment Management & Safety Engineering College, Air Force Engineering University. He has published more than 10 research papers. His research interests are equipment system engineering and equipment development & demonstration. E-mail: afeulck@163.com|CHEN Yunxiang was born in 1962. He received his M.S. degree from Air Force Engineering College in 1989 and Ph.D. degree from Northwestern Polytechnical University in 2005. Now he is a professor of Equipment Management & Safety Engineering College, Air Force Engineering University. He has published five books and more than 50 research papers. He is an expert of Air Force in reliability, maintenance and support. His research interests are reliability assessment, equipment maintenance support and equipment development & demonstration. E-mail: cyx87793@163.com|XIANG Huachun was born in 1980. He received his M.S. degree of management science and engineering in 2005 and Ph.D. degree of system engineering in 2009 from Air Force Engineering University. Now he is a vice professor and director of Equipment Management & Safety Engineering College, Air Force Engineering University. He has published two books and more than 20 research papers. His research interests are reliability assessment and equipment system engineering. E-mail: 32134717@qq.com|WANG Weijia was born in 1990. He received his B.S. degree in 2013 and M.S. degree in 2015 from Air Force Engineering University. Now he is a Ph.D. candidate of Equipment Management & Safety Engineering College, Air Force Engineering University. His research interests are cooperative control of UAVs, sensor networks for radar and electronic warfare applications. E-mail: visionwng@foxmail.com|WANG Zezhou was born in 1992. He received his B.S. degree in 2014 and M.S. degree in 2016 from Air Force Engineering University. Now he is a Ph.D. candidate of Equipment Management & Safety Engineering College, Air Force Engineering University. His research interests are reliability assessment and equipment system engineering. E-mail: 350276267@qq.com
  • Supported by:
    the National Natural Science Foundation of China(71571190);the National Natural Science Foundation of China(71601183);the National Natural Science Foundation of China(L1534031);the Shanxi Province Natural Science Foundation of China(2014JQ2-7045);This work was supported by the National Natural Science Foundation of China (71571190; 71601183; L1534031) and the Shanxi Province Natural Science Foundation of China (2014JQ2-7045)

Abstract:

In order to achieve the information fusion in the time domain based on the evidence theory, an evidence combination method in the time domain based on reliability and importance is proposed according to the idea of evidence discount. Firstly, the distortion of the time-domain evidence is judged based on single exponential smoothing. The real-time reliability of the evidence at the adjacent time is obtained by the real-time reliability assessment method of the evidence based on the credibility decay model. Then, the relative importance of the evidence at the adjacent time is obtained by comprehensively considering improved conflict degree and uncertainty. Finally, based on the criterion of evidence discount and the Dempster's rule of combination, the evidence combination is carried out to achieve the sequential combination of time-domain evidence. The numerical simulation and analysis show that this method has fully embodied the dynamic characteristics of time-domain evidence combination, and it has strong processing ability for conflict information and anti-disturbing ability. The proposed method has good applicability to information fusion in the time domain.

Key words: evidence theory, time-domain evidence combination, evidence discount, reliability, importance