COMPARATION OF MATHEMATICAL MODELING AND ARTIFICIAL NEURAL NETWORKS TO PREDICT THE OUTPUT THE CAPACITY OF MATERIALS ON MODIFIED SAGO STRACH PNEUMATIC CONVEYING RING DRYER

Modified starch can be obtained through the process of fermentation and UV light irradiation. To dry the modified starch, a pneumatic conveying ring dryer (PCRD) equipped with dewatering and stirring fermentors and irradiation with UVC lamps. One of the important factors in designing the dryer is to know the capacity of the dryer. PCRD dryer capacity can be known through mathematics and artificial neural network (ANN). The purpose of this research is to compare between mathematical models based on dimensional analysis and ANN to predict the output capacity of the material in PCRD type modified wet sago starch dryers. Mathematical models and ANNs are analyzed, tested and trained using experimental data obtained from PCRD performance testing. The comparison results between the mathematical models with ANN shows that the ANN model is the best for predicting material output capacity (Qob) on PCRD. Each input variable has a very valid influence on the material's output capacity