Application of a Novel LSSVR Based FLANN Architecture to Dynamic Compensation for Infrared Thermometer Sensor

A novel functional link artificial neural networks (FLANN) architecture is presented and applied to dynamic compensation for infrared thermometer sensor. Firstly, the identification results between generic FLANN and least squares support vector regression (LSSVR) are verified to be similar. Then a new method to update the FLANN weights are derived from LSSVR. In comparison with the generic FLANN, The improved one differs markedly in solving a set of linear equations instead of an iterative problem. As a result, the more accurate weight evaluations are obtained, and a faster learning can be expected. Lastly, the infrared thermometer sensor dynamic compensator is established based on the principle of inverse model rectification and the improved FLANN is used to describe the compensator. The actual calibration data of the infrared thermometer uIRt/c are used to test. The experimental results show that the improve FLANN is faster in training speed, higher in precision and more robustness.