Articles

Predictive Model of Wood Dyeing Pigment Formula

  • GUAN Xuemei;GUO Minghui;CAO Jun
Expand
  • 1. College of Mechanical and Electrical Engineering, Northeast Forestry University, Harbin 150040, China2. Key Laboratory of Bio-based Material Science and Technology of Ministry of Education, Northeast Forestry University, Harbin 150040, China

Received date: 2012-07-29

  Revised date: 2013-04-03

  Online published: 2013-06-18

Abstract

Wood color is an important factor determining consumer first impression. In order to improve the decorative role and value of wood products, wood and wood materials need to be coloring. Applying computer color matching method to wood dyeing for speeding up the generation of dyeing formula would greatly improve work efficiency and save costs. A kind of prediction model for wood dyeing pigment formula is built by using Dynamic Fuzzy Neural Network (DFNN). The word "dynamic" refers to the fact that the network structure of fuzzy neural network does not preset, it changes dynamically; that is to say, there is no predeterminate fuzzy rule before learning, its fuzzy rules gradually increase and form during the learning process. The output is concentration values of reactive brilliant red X-3B, reactive yellow X-R, and reactive blue X-R, input is color difference, namely, Δ<i>L</i>, Δ<i>a</i>, and Δ<i>b</i>. The relative error of the prediction model is 0.52% and its training time is 128s. The results are comparatively satisfactory. The method provides a new way for wood dyeing and color matching and a new idea for the applications of its theories in color matching system; therefore it has certain value for theoretical research and practical applications.

Cite this article

GUAN Xuemei;GUO Minghui;CAO Jun . Predictive Model of Wood Dyeing Pigment Formula[J]. Science & Technology Review, 2013 , 31(17) : 29 -32 . DOI: 10.3981/j.issn.1000-7857.2013.17.004

Outlines

/