On-line identification of new coal type using joint probability density arbiter

A new approach for on-line identification of new coal type as well as existing coal types by combining the principal component analysis (PCA) technique and the joint probability density arbiter is presented. The flame features were extracted in the time domain and frequency domain from each flame oscillation signal and formed an original feature vector. The principal component analysis technique was utilized to transform the original feature vector into an orthogonal and dimension-reduced feature vector. A joint probability model was established for each known coal type by using the data of the orthogonal feature vector. Then the joint probability density arbiters based on the joint probability density models of the known coal types were used to determine whether the coal being burnt is new and to identify the coal type if it is one of the known types. Experimental data validated the approach and showed that the success rate can reach over 95% under the experimental conditions being conducted.