Flatness pattern recognition via EA-ABC cloud inference network implemented by DSP
The present flatness pattern recognition method based on neural network belongs to software simulation, and it can't be directly applied to engineering practice. What's more, the accuracy of present flatness pattern recognition is limited. In order to improve it, flatness pattern recogniti...
Saved in:
Published in | 2016 35th Chinese Control Conference (CCC) pp. 3490 - 3495 |
---|---|
Main Authors | , , , , |
Format | Conference Proceeding Journal Article |
Language | English |
Published |
TCCT
01.07.2016
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Summary: | The present flatness pattern recognition method based on neural network belongs to software simulation, and it can't be directly applied to engineering practice. What's more, the accuracy of present flatness pattern recognition is limited. In order to improve it, flatness pattern recognition via EA-ABC cloud inference network implemented by digital signal processor (DSP) is presented in this paper. Firstly, based on T-S cloud inference network, flatness pattern recognition model is designed in DSP. And it is applied to 900HC reversible cold rolling mill. Later, flatness pattern recognition model optimized by efficient adaptive artificial bee colony (EA-ABC) algorithm is proposed. Finally, the two results of flatness pattern recognition model, which run on MATLAB and DSP respectively, are compared and analyzed. Experimental results confirm that EA - ABC algorithm has such advantages: high optimizing precision, less number of iterations, fast speed of convergence. Flatness pattern recognition model via EA-ABC cloud inference network can identify the defect types of flatness correctly. At the same time, the experimental results verify that the T-S cloud inference network can run on the hardware TI TMS320F2812 in a fast speed and it provide a basis for neural network which is applied to practical engineering. |
---|---|
Bibliography: | ObjectType-Article-2 SourceType-Scholarly Journals-1 ObjectType-Conference-1 ObjectType-Feature-3 content type line 23 SourceType-Conference Papers & Proceedings-2 |
ISSN: | 2161-2927 1934-1768 |
DOI: | 10.1109/ChiCC.2016.7553895 |