Chebyshev FLN-based Nonlinearity Estimation of LMCS

Lumber moisture content (LMC) is a key parameter for regulating and controlling wood drying process. With detailed study, authors found that the precision of LMC is influenced by many nonlinear factors such as noises, intercross-sensitivity existing among multiple environmental factors, and the electric characteristic of ionic groups changing with the ambient temperature fluctuating, etc. A novel functional link network (FLN) based on Chebyshev polynomial is put forward for this purpose to compensate for the nonlinear response characteristics and complex nonlinear dependency of the environmental factors on the sensor characteristics. The hidden layers of conventional neural networks is eliminates by functional expansion. The computational economy to be gained by Chebyshev series increases when the power series is slowly convergent. The simulation studies that the FLN-based algorithm estimated MC errors remain within plusmn0.8 at ultimate stage over a wide range of temperature (24degC~80degC). Therefore, nonlinearity estimation based on FLN is a very effective method for lumber MC measuring