A study was conducted at Sindhanur and Manvi taluks of Raichur district, Karnataka, India to assess the post harvest grain losses in rice combine harvester, where rice is the predominant crop and the fanners are familiar with Escort (Crop Tiger) type combine harvester. Combine harvester performance and conventional harvesting system were compared by taking samples in fields for two crop seasons in successive years in the study area. Results indicated that the post harvest grain losses were found minimum (2.87%) in rice combine harvester and the cost of operation in conventional harvesting was 2.28 times more. The pay back period was found to be less than one year if the machine could harvest 1000 hectares (ha) per year. A mathematical model is felt necessary to predict post harvest grain losses. A multivariate quadratic type function was found to be appropriate to detect and analyze the joint influence of a set of process variables of the rice combine harvester on another set of crop parameters. The estimated models were used for the quantitative prediction of the harvesting, threshing and separation losses. A "confidence band" was fixed for minimizing the individual losses. The model was confirmed through the Chi-square test for validation. Finally, the model was tested randomly in the study area for different set of parameters. The coefficients were determined using the performance test of observed and modeled values. The model is validated with the Chi-square test and the results were found justifiable in field condition.
展开▼
