Approach to radiation temperature measuring and its application via support vector machine

This paper presents a measuring method based on Support Vector Machine(SVM), which is used to solve the high temperature measuring problem. As we all known, it is difficult to measure directly in complex industrial environment. Thus, the normal support vector machine(NOR-SVM) is improved, and then a new regression algorithm is proposed. Simulation results demonstrate that the improved algorithm has good nonlinear modeling, generalization ability and predictive ability. What´s more, this model needs less Support Vectors(SVs), so it learns more faster.