DEVELOPMENT OF A REGRESSION MODEL FOR PREDICTING RICE GRAIN BREAKAGES FOR DEUTZ-FAHR M1202 COMBINE HARVESTER
Publication Date : 01/02/2019
A multiple linear regression analysis was used to develop a performance model on grain breakages for Deutz-Fahr M1202 combine harvester with grain moisture content (MC) and cylinder-concave clearance as the two independent variables and grain breakages as the dependent variable. Field experiments were conducted at constant combine harvester forward speed 1.70 km/h. Combination of three different grain moisture contents 11, 13 and 15 % (wet basis) and three different cylinder-concave clearances 12, 20 and 26 mm were used as independent variables to determine the dependent variable of grain breakages. Analysis of variance shows that there was highly significant difference in the mean values obtained for grain breakages at different MC and concave clearances. Theoretical field capacity, effective field capacity and field efficiency were 0.77 ha/h, 0.68 ha/h and 88 % respectively. The predicted values obtained through the developed model were compared to another set of observed field data in order to validate it. High value of R2(0.901) which is significant indicate good agreement between the real values and the predicted values of grain breakages for the combine harvester. The accuracy and precision was calculated as 0.083 while bias value of 0.00077 was obtained. The mean square of predicted error (MSPE) was found to be 0.083.
No. of Downloads :