Smell Peak Prediction During Black Tea Fermentation Process Using Time-Delay Neural Network on Electronic Nose Data

Fermentation process in black tea manufacturing plays the key role in determining the quality of finished tea. During this process, a complex chain of biochemical changes occurs and the process should be terminated once the optimum fermentation point is reached. Present day practice for detection of optimum fermentation point is purely subjective, and is carried out by experienced industry personnel. Even though chemical methods are available, but they are expensive, time-consuming and offline. A study has been made on real time smell monitoring of black tea during fermentation process using electronic nose and is reported in this paper. Time-delay neural network (TDNN) architecture has been used on time series data obtained from electronic nose for smell peak prediction during the fermentation process. The online predicted result using TDNN seems very promising to detect the optimum fermentation time for black tea manufacturing process