Data prediction for calibration of seed drill using multiple linear regressionPantnagar Journal of Research, Volume - 20, Issue - 1 ( January-April. 2022)
PDF Views - 116 | Downloads - 79
Seed drill with fluted-roller seed metering devices is the mostly used for sowing of wheat crop in India. For obtaining the high yield, it is very essential to drop the wheat seeds in rows maintaining accurate seed rate with minimum damage to seeds during metering. This mainly depends on forward speed of the ground wheel, exposure length of the fluted-roller seed metering mechanism and hopper depth. The relationship between these factors and the dependent parameter, i.e., seed rate can be established using multiple linear regression analysis. Hence, an attempt has been made to develop the multiple linear regression (MLR) model using 3 Factor Completely Randomized Design for the prediction of the performance parameters (seed rate) of the fluted-roller seed metering device using speed of ground wheel, hopper depth and exposure length as input parameters. The data were generated in the laboratory by conducting experiments on a sticky belt test setup. The generated data was used to develop statistical model. All independent parameters such as hopper depth, exposure length and speed of ground wheel were found highly significant on seed rate. The R2 values for MLR model during training, validation and testing were found to be 0.983, 0.988 and 0.986 respectively and RMSE values during training, validation and testing were found to be 17.84, 14.84 and 13.71 respectively Seed drill, fluted-roller seed metering mechanism, multiple linear regression (MLR) model, wheat.
Download Full PDF