On-line identification of fuel type using joint probability density arbiter and support vector machine techniques

This paper presents a new approach for on-line identification of fuel type by combining the joint probability density arbiter and support vector machine techniques. The flame features are extracted both in the time domain and frequency domain from each flame oscillation signal and form an original feature vector. Orthogonal and dimension-reduced features are obtained by using the principal component analysis technique. In order to identify fuel types, a joint probability density arbiter model and a support vector machine model are established for each known fuel type by using the orthogonal features. Then the joint probability density arbiter model is used to determine whether the type of fuel is new or not and the support vector machine model is used to identify the fuel type if the fuel is one of the known types.