摘要
In the traditional Markov chain model (MCM), aleatory uncertainty because of inherent randomness and epistemic uncertainty due to the lack of knowledge are not differentiated. Generalized interval probability provides a concise representation for the two kinds of uncertainties simultaneously. In this paper, a generalized Markov chain model (GMCM), based on the generalized interval probability theory, is proposed to improve the reliability of prediction. In the GMCM, aleatory uncertainty is represented as probability; interval is used to capture epistemic uncertainty. A case study for predicting the average dynamic compliance in machining processes is provided to demonstrate the effectiveness of proposed GMCM. The results show that the proposed GMCM has a better prediction performance than that of MCM.
In the traditional Markov chain model （MCM）, aleatory uncertainty because of inherent randomness and epistemic uncertainty due to the lack of knowledge are not differentiated. Generalized interval probability provides a concise representation for the two kinds of uncertainties simultaneously. In this paper, a generalized Markov chain model （GMCM）, based on the generalized interval probability theory, is proposed to improve the reliability of prediction. In the GMCM, aleatory uncertainty is repre- sented as probability; interval is used to capture epistemic uncertainty. A case study for predicting the average dynamic com- pliance in machining processes is provided to demonstrate the effectiveness of proposed GMCM. The results show that the proposed GMCM has a better prediction performance than that of MCM.
基金
supported by the National Key Basic Research Program of China (973 Program) (Grant No. 2011CB706803)
the National Natural Science Foundation of China (Grant Nos. 51175208, 51075161)
