OPTIMIZATION AND FABRICATION OF A DUAL THERMOPILE SENSOR-BASED ON THEBEM

A thermopile sensor containing a dual black-body system for the compen sation of significant environmental temperature change has been develo ped. The boundary-element method (BEM) and genetic algorithm (GA) have been adopted for analysis and optimization of the sensor. The object function of optimization is mainly focused on sensitivity. The present dual sensor is composed of one radiation filter, one diaphragm and tw o radiation absorption areas. One receiving area senses the radiation energy from the object and the other area compensates the environmenta l temperature variation. A reflection mirror focuses on the sensing ar ea and the compensation area can exchange radiation energy with the en vironmental atmosphere. The analytical solution is compared with the B EM solution in simple sensor geometry. It is found that modelling the sensor is easier with the BEM than with the finite-element method (FEM ). In addition, the present BEM modelling provides a more realistic so lution than the previously presented analytical solution. The fabricat ed dual thermopile sensor shows good agreement with the simulated resu lt using the BEM. (C) 1998 Elsevier Science S.A.