Cases studies of Chebyshev functional link networks in engineering applications

The sensors operating in lumber drying kiln with harsh environment are easily interfered by ambient factors. Aimed to eliminate the noise, a novel data fusion algorithm based on function link network (FLN) based support vector machines (SVM) has been introduced to compensate for the nonlinear response characteristics in the parameters automatic measuring system of lumber drying process. In the proposed algorithm, FLN eliminates the hidden layers of conventional neural networks by expanding the input pattern into a high order dimensional space. The experimental simulation results show that optimum FLN construction could be uniquely obtained by SVM through solving a quadratic programming. The experimental research proves the improved functional link network used in the lumber drying measuring system can compensate the ambient temperature interference effectively and it has certain project.