FUEL IDENTIFICATION BY NEURAL-NETWORK ANALYSIS OF THE RESPONSE OF VAPOR-SENSITIVE SENSOR ARRAYS

Neural network analysis of the response of an array of vapor-sensitive detectors has been used to identify sbr different types of aviation f uel. The data set included 96 samples of JP-4, JP-5, JP-7, JP-8, JetA, and aviation gasoline (AvGas). A sample of each neat fuel was injecte d into a continuous stream of breathing air through an injection port from a gas chromatograph. The aspirated sample was then swept from the injection port to the chamber without separation. In the chamber, the sample was exposed to an array of eight vapor-sensitive detectors. Th e analog output of each detector was digitized and stored while the sa mple was swept into and through the chamber. The response of each dete ctor was then averaged and stored as the final response or pattern of each sample. It was clear from a visual inspection of each of the rada r plots that there was a characteristic pattern in the response of the array to five of the sbr different fuel types. This was confirmed usi ng neural network analysis to study the entire data set. A two-step pr ocedure was developed to separate the patterns of all six fuel types i nto their respective classes. In the first step, fuels were separated into one of five groups: JP-4, JP-5, JP-7, AvGas, or a combined JP-8/J etA group. In the second step, the fuels in the combined group were se parated into either JP-8 or JetA groups.