Simulation model for the rapid determination of tobacco moisture content

A simulation model was constructed to predict the initial moisture content of burley tobacco leaves based on the outlet air temperature-time data during a forced air drying process. The system consisted of a moist sample of shredded tobacco leaves which were placed in a forced air stream in which heat and mass transfer occurred. A mathematical model was developed for heat and mass balances of both the drying air and the tobacco moisture, resulting in the four partial differential equations. More than 20 parameters of physical and thermal properties were included in the equations which concentrated on the temperature and the moisture of the air and the tobacco. To overcome the complex nature of the equations, they were solved simultaneously by an implicit finite difference method.

A computer model was programmed using a BASIC language for the solution of the four partial differential equations that yielded a drying subroutine which was able to predict the outlet air temperature that changed with time from an assumed initial moisture content of the tobacco. A searching subroutine was programmed to reverse the process of prediction in which the initial moisture content of the tobacco could be obtained if the outlet air temperature-time had been known. The searching subroutine arbitrarily assigned the values of the initial moisture content that produced the temperature-time curves bounding above and below the known temperature-time data. The sum of the differences between the calculated and the reference values was used as a criterion to bring the calculated temperature-time curve, through an iterative process, to match itself with the reference data. The final value of the initial moisture content associated with the matching model curve was identified as the predicted initial moisture content of the known temperature-time data.

Experiments were conducted to verify the model using samples of 3-mm width shredded burley tobacco leaves with initial moisture contents ranging from 14.60 to 59.83% d.b. The samples were dried in an experimental dryer with hot air at temperatures of 100, 120, and 140°C and 0.6096 m/s airflow rate for 5 minutes. The outlet air temperatures were recorded each 5 seconds to produce a reference temperature-time data. After data processing with accompanying instrumentation, the initial moisture content from the simulation model was compared to the standard 24-hour oven method values.

The simulation model was successful in providing a mean for the rapid determination (less than five minutes) of the initial moisture content of the tobacco leaves. Results revealed that there was no significant difference between the calculated and the oven values. The 100°C drying temperature was found to be the most effective. For the 100°C drying temperature, the mean of the difference between the calculated and the oven values was -0.0947 percentage points with a standard error of 0.3610.