Optimization of a Screw Press Using Linear Programming Techniques

Screw presses are exclusively used for extraction of oil from vegetable seeds; having a mechanical assembly where screw shaft is used to press the seeds at high pressures in order to have a reasonable oil yield. Mostly conventional screw presses are used in the commercial and industrial sectors having high O&M costs and hence less life. For this purpose an efficient small scale screw press was selected for this study. The objective was to apply the optimization techniques in order to have a set of optimized decision variables satisfying certain constraints to get the maximum oil yield. The software used for linear optimization was TORATM Version 2.00. The objective function was to maximize the oil yield. The decision variables comprised of seed input, screw speed, motor power, die opening size, total pressure and temperature. Constraints were accordingly defined for each variable and the data was fed to TORAs input grid. The solver menu used the simplex starting all-slack method to solve the LP model. The output of the model selected the optimized variables which resulted in maximum oil yield. To include the losses incurred in the screw press a loss function was developed including the heat losses, mechanical/inertial losses and seed cake losses. The results calculated depicted an oil extraction rate of as high as 35.42% compared to the conventional oil extraction rates of 28-32%.