Neues Regressionsmodell mit einer Dummy-Variablen und geometrischen Zinkenparametern zur Vorhersage der horizontalen Kräfte einzelner Zinken

Abstract

This paper discusses three different equations of a regression model to predict the horizontal force for single tines. Four standard single chisel plow tines were used (heavy duty, double heart, double heart with wings and duck-foot). The first model is based on the effect of the operational conditions speed and depth for each tine, the second model is employing a statistic dummy variable, also representing each tine. The geometry of tines is base for the third model. A stepwise selection with a multi-linear regression at significance level 5% was used to evaluate these regression models. Experiments were carried out in a sandy loam soil at soil moisture content of 10.3% ± 0.8 (based on dry matter) and soil bulk density of 1.38 g/cm³ ± 0.01 under controlled soil bin conditions featuring varying speeds between 1.1 and 3.6 m/s and varying depth from 5 to 20 cm. Field testing was done in order to validate the regression model obtained from the soil bin. The results show that the horizontal force increases linearly with the speed-depth interaction term and quadratically with the depth for each tine in all regression models. The depth is effecting the force more significantly in comparison to speed (p < 0.05). Eventually it can be stated that there is a good general accordance of observed and predicted values of the horizontal force for all tines by using the dummy and the geometric regression models.