Measurement of premixed propane combustion equivalence ratio based on flame image color and support vector machine

In order to realize the detection of premixed propane combustion equivalence ratio, a measurement method of propane combustion equivalent ratio based on flame image color information and support vector machine (SVM) is proposed. Firstly, the flame image data was captured by machine vision. Secondly, the flame image was preprocessed by digital image processing technology to obtain the flame target region. The extracted color features were then used to feed SVM for the combustion equivalent ratio prediction modeling. The parameters in SVM were selected by using particle swarm optimization algorithm. At last, the feasibility of the proposed method was further verified by comparing BP neural network and RBF network. The experimental results show that the proposed method can realize the prediction of premixed propane combustion equivalent ratio effectively.