Paper
25 October 2006 Diameter detection and process quality prediction for optimal control of enameled wires
Hai-tao Su, Hua Dong, Bo Xu
Author Affiliations +
Proceedings Volume 6280, Third International Symposium on Precision Mechanical Measurements; 628030 (2006) https://doi.org/10.1117/12.716317
Event: Third International Symposium on Precision Mechanical Measurements, 2006, Urumqi, China
Abstract
The quality of enameled wires impacts the reliability of motor products, wiring products and electronic products. The diameter of line-shape material usually is measured by means of photoelectricity energy, laser scanning, laser diffraction and projection imaging. A bi-light-source projection method was proposed by Zhao Bin etc., which measured 0.1-1.0mm diameter of enameled wires by using bi-refringent-crystal. A new method, single-light-source projection, is proposed in this paper. A beam from semiconductor laser light-source was disparted into two beams by Fresnel bi-prism. Then, the two beams projected onto an enameled wire and detected by CCD. The accuracy of the new method is much higher due to the magnified projection. Aiming at multi-factor input and nonlinear output of the enameled wires' production process, the RBF neural network was selected to pre-treat the process data obtained by our new method to set up dynamical training-set and predict quality. The simulation training results have proved that RBF neural network prediction method costs less training time, has less prediction errors and less samples. In summing up, it may be stated that single-light-source projection and RBF neural network prediction method are very suitable for real-time optimal controlling on enameled wires' production process.
© (2006) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Hai-tao Su, Hua Dong, and Bo Xu "Diameter detection and process quality prediction for optimal control of enameled wires", Proc. SPIE 6280, Third International Symposium on Precision Mechanical Measurements, 628030 (25 October 2006); https://doi.org/10.1117/12.716317
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Neural networks

Neurons

Control systems

Data modeling

Charge-coupled devices

Diffraction

Error analysis

Back to Top