An improved Markov model for idle time prediction based on spatiotemporal learning
- 195 Downloads
In order to reduce the fuel consumption of idling, lots of vehicles are equipped with idle start–stop system. In the actual traffic situation, heavy traffic result in frequent short time idle stop, which causes two major problems with the existing start–stop system, frequent starting and stopping and invalid stopping. That not only exacerbates component wear but also makes fuel efficiency worse. Aiming at this problem, an improved Markov model for idle time prediction based on spatiotemporal learning is proposed. Based on original Markov model, related training subset for training model is determined according time and space information, which could increase prediction accuracy by reducing the redundant noisy data. Experiments conducted on real urban dataset demonstrate that our proposed model could get better performance, the F1-measure increased by 8.0%. Start–stop strategy using the prediction method can make invalid idling reduced by 35.1%, which can effectively avoid invalid idling and obtain a reduction of frequent starting and stopping.
KeywordsStart–stop system Markov model Idle time prediction Spatiotemporal learning Control strategy
This work is supported by Key Science and Technology Program of Jilin Provincial Science and Technology Department (Grant No. 20150204052GX).
- 1.Jou, R.C., Pai, C.W., Wu, Y.C.: Idling stop fines for exceeding legal idling times—from the driver’s perspective. Transp. Res. A 66, 88–99 (2014)Google Scholar
- 4.Bishop, J., Nedungadi, A., Ostrowski G, et al.: An engine start/stop system for improved fuel economy. In: SAE World Congress and Exhibition (2007)Google Scholar
- 9.Yang, L., He, H., Sun, F., et al.: Research of fuzzy logic control strategy for engine start/stop in dual-clutch hybrid electric vehicle. In: Seventh International Conference on Fuzzy Systems and Knowledge Discovery. IEEE, pp. 912–917 (2010)Google Scholar
- 10.Chen, H.Y., Yu, X.L., Zuo, C.J.: A research on the control strategy of engine smart start/stop system. Automot. Eng. 9, 1145–1150 (2014)Google Scholar
- 11.Li, C.C., Chen, Q., Yu, F., et al.: A start–stop system for micro-HEV and the analysis on its effects on vehicle performance. Automot. Eng. 34, 879–884 (2012)Google Scholar
- 13.He, R., Liu, K., Huang, D.X., et al.: A study on the control strategy of intelligent idling stop and start system of engine. Automot. Eng. 32, 466–469 (2010)Google Scholar
- 14.Yu, Z.S.: Automobile Theory, 4th edn. China Machine Press, Beijing (2006)Google Scholar
- 15.Fathy, H.F., Kang, D.S., Stein, J.L.: Online vehicle mass estimation using recursive least squares and supervisory data extraction. In: American Control Conference (2008)Google Scholar
- 18.Song, H.Y.: Research on the Automatic Image Annotation and Annotation Refinement Algorithms. Jilin University (2012)Google Scholar