Modeling and Optimizing Performance of Pumpkin Seed Separator Machine Using Response Surface Methodology

Separation is one of the important processes in harvesting agricultural products. Finding appropriate methods for improved separation of the seeds and reducing loss of the seeds in the separation devise are important challenges of this field. The aim of this research is to model and optimize performance of pumpkin seed separator machine. Experiments were conducted using response surface methodology and central composite design Results revealed that diameter of the separator shaft gear, width of the frictional belts and speed of the PTO shaft, at 1% probability level, and feed rate, at 5% probability level, have significant effects on seed loss percentage and purity of the seeds. Second-order regression models were presented for predicting response variables. Values of 3 days, 184 mm, 12.9 cm, 18.33 tons h-1 and 450 rpm were outlined as the optimum values of rest after harvest, gear diameter, width of the frictional belt, feed rate and speed of the PTO shaft, respectively.