Neural Network Soft Sensor Model Based on Mechanism Chaotic Analysis and its Application

Abstract:

Ball produce detection and control is an important aspect of ball mill control, however, contribute to the lack of effective and reliable ball detection means in the actual production process, so it is difficult to achieve optimal control. In this paper, secondary variable is selected with this mechanism analysis and information processing technology of choice, the establishment of the ball fast and accurate output of neural network soft sensor model can not only contribute to estimate the steady-state mill and ball mill on the dynamic process can effectively contribute to an online estimate, the soft-sensor model has a strong possibility. At the same time，it provides a reference for the soft measurement of power plant other parameters. This modeling method has portability.

• Periodical: Advanced Materials Research (Volumes 532-533)
• Pages: 508-514
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.532-533.508
• Online since: June 2012
• Authors: Yu Hong Zhao, Yun Hui Zhang, Ze Guang Su
• Keywords: Chaos, Mechanism, Neural Network (NN), Soft Sensor

References

[1] Yongsheng Hao. ball mill pulverizing system functions control and optimizing decision based on soft measurement [J]. Energy Research and Utilization, 2003 (4): 21-24.

[2] Yunhui Zhang. Soft Computing Application in the Thermal Process [D]. [PhD thesis]. Nanjing: Southeast University, 2006: 90-97.

[3] Sugihara G, M ay R M. Nonlinear forecasting as a way of distinguishing chaos from measurement error in time series[J ]. Nature, 1990, 344 (6268) : 734- 741.

DOI: 10.1038/344734a0

[4] Hua-Guang Zhang, Zhiliang Wang, Michael. Chaotic system theory [M]. Shenyang: Northeastern University Press, (2003).

[5] P. Grassberger，I. Procaccia. Measuring the strangeness of strange attractors[J]. Physical D, 1983 (9): 189-208.

DOI: 10.1016/0167-2789(83)90298-1