Dynamic Compensation Method for Sensors Based on FLANN Constructed by LS-SVM

A novel method of constructing functional link artificial neural networks (FLANN) for sensor dynamic compensator was presented. The design steps and learning algorithm were also addressed. Compared with traditional BP-based FLANN, the new least squares-support vector machine (LS-SVM)-based FLANN had more advantages: 1) the LS-SVM solution solved a set of linear equations instead of an iterative problem; 2) FLANN compensator can be uniquely obtained due to the global maximum in the whole training process. The experiment results show that the presented method is faster in learning speed, higher in accuracy, more robust in noise resistance. So it is more suitable for sensors dynamic system.