Modelling of unit contact surface of bean seeds using artificial neural networks
Marek Wróbel , Jarosław Frączek , Sławomir Francik , Zbigniew Ślipek , Krzysztof Mudryk
AbstractThe paper proposes a model utilising the Artificial Neural Networks to describe the relationships of the unit contact surface of bean seeds as a function of their geometrical dimensions, load and water content. It has been proven that all factors assumed affect the output variable which is the unit contact surface of the seeds and flat surface. The experiments were performed on beans (Jubilatka specie) for six different water content levels in the seeds. The bean seeds were loaded with eight values of compression force (from 41 to 230 N). Five analyses were performed in parallel for each experimental combination. The neural model developed is a typical unidirectional network – four-layer perceptron with a large number of neurons in the hidden layers. ANN has preserved its generalisation ability – the average relative error 4.7 % for the testing data (not used in the teaching process).
|Publication size in sheets||0.5|
|Book||Malinovska Larisa, Osadcuks Vitalijs (eds.): 12th International Scientific Conference on Engineering for Rural Development - Proceedings, Engineering for Rural Development, vol. 12, 2013, Latvia University of Agriculture Faculty of Engineering , 663 p.|
|Keywords in English||external friction, plant material, energy crops|
|Publication indicators||= 1|
|Oświadczenie||Marek Wrobel, Jaroslaw Fraczek, Sławomir Francik, Zbigniew Slipek, Krzysztof Mudryk|
* presented citation count is obtained through Internet information analysis and it is close to the number calculated by the Publish or Perish system.