Specific draft estimation model for offset disc harrows

Abstract Predicting draft requirements of tillage implements is essential from the viewpoint of proper tractor-implement matching and machinery design. It was hypothesized that draft requirement of offset disc harrow (ODH), a widely used tillage implement depends mainly on four major soil and working parameters, i.e. soil cone index (CI), forward speed, working depth and front gang angle (FGA), which could be used to develop a reasonable regression model for estimating its specific draft. To test this hypothesis, investigations were carried out on the draft requirement of ODH in an indoor soil bin available at Indian Institute of Technology, Kharagpur (22°19′ N, 87°19′ E), India during the months of November 2016 to May 2017. The soil available in the soil bin was acid lateritic sandy clay loam and taxonomically grouped under order Alfisol (Oxyaquic haplustalf). The experimental design was completely randomized with draft measured using S-type load cell (2-tonne capacity) at four levels each of FGA and forward speed, three levels each of working depth and soil CI using a 3 × 3 ODH (i.e. 3 discs in each of the two gangs) with the objective to acquire data on the draft force requirement of ODH at various soil and working parameters. The forward speed, working depth, and soil CI were measured with the help of a magnetic-type proximity switch, rotary potentiometer, and hydraulic cone penetrometer, respectively and were found to have a significant effect (p ≤ 0.01) on the draft with CI having highest influence followed by working depth, speed, and FGA. Multiple regression analysis was carried out to develop a model for estimating specific draft for ODH. The efficiency of specific draft model, was assessed by various performance indices such as mean error (ME), value account for (VAF), root mean square error (RMSE), coefficient of determination (R 2 ), and mean absolute percentage error (MAPE) and their values were found to be -3.22, 92.77%, 4.82, 0.90 and 14.75, respectively. A major advantage of the developed specific draft model compared to other researchers’ model is the inclusion of all parameters that affect draft especially FGA and CI which would increase its applicability. A good general agreement between measured and estimated specific draft was found with the data obtained from the soil bin with an average absolute variation (AAV) of 10.76%. Verification of the model was also carried out by comparing its outputs with the model outputs of previous researchers to check its applicability for other soil types and working conditions, and the results were found to be fairly good (Al-Janobi and Al-Suhaibani, 1998 (R 2  = 0.87, AAV = 14.33%), Kheiralla et al., 2004 (R 2  = 0.93, AAV = 11.95%), and Roul, 2014 (R 2  = 0.94, AAV = 23.05%) except the ASABE draft model (R 2  = 0.66, AAV = 50.18%). Therefore, the developed model in this research could be used to predict draft requirements of ODH’s with reasonable accuracy.