Zbigniew Ślipek, Sławomir Francik, Jarosław Frączek and Adrian Knapczyk
The article presents the procedure of determination of a structural form of an agricultural machine which encompasses methodological aspects of design related to the division of the basic objective function into member functions, multi-variant solutions and synthesis of astructural form (a virtual prototype of an agricultural machine). The suggested course of proceeding was presented by the case of a design of a tunnel spraying machine. This example reflects well and includes the specificity of the design object in the form of an agricultural machine with specific technical and agro-technical requirements. Attention was paid to the aspects related to methodology of selection of fractional solutions based on previously defined criteria.
Sławomir Francik, Zbigniew Ślipek, Jarosław Frączek and Adrian Knapczyk
The objective of the paper was to carry out a bibliometric quantitative analysis of publications concerning the application of artificial neural networks in the research area - agriculture and a bibliometric quantitative analysis and subject analysis with regard to agricultural engineering. A number of scientific publications devoted to the ANN found in the data base of the Web of Science - in documents published to 2015 was a basis for the quantitative analysis. Research on the use of artificial neural networks in the research area - agriculture is extending systematically. Moreover, a rapidly growing number of citations prove a continuous increase in the scientists’ interest in possibilities of the ANN applications. The quantitative analysis of scientific publications in 5 selected scientific journals and thematically related to agricultural engineering (indexed in the Web of Science) allowed a statement that 236 scientific articles from 1996- 2015 were related to the ANN application. The biggest number of publications was reported in Computers and Electronics in Agriculture - 118 articles. In 2011-2015 there was a growing trend in dynamics of publishing of scientific papers devoted to the ANN application to agricultural engineering. Thus, we may assume that the research related to application of the artificial neural networks to agricultural engineering will be continued and their scope and number will be still growing. The thematic analysis of the most often quoted publications from 2011-2015 in the journal Computers and Electronics in Agriculture, proved that they concern both the issues related to the classification problem as well as to modelling processes and systems. We should suppose that the subjects related to modelling of drying processes and application of neural networks for image analysis will grow dynamically in the following years.
Jarosław Frączek, Sławomir Francik, Zbigniew Ślipek and Adrian Knapczyk
The objective of the research was to create a model which defines the relation between a fundamental contact area of a seed and the pressure force, water content in a seed and its geometrical dimensions with application of artificial neural networks (SSN). Computer program Statistica Neural Networks v. 6.0. was used for formation of a neural model. Tests were carried out on Roma wheat seed and Dańkowskie Złote rye with six various water contents: 0.11 0.15 0.19 0.23 0.28 0.33 (kg·kg-1 dry mass). Caryopses were loaded with eight values of compression force - from 41 N to 230 N. Multiplicity of iterations was 5. Seed material was moistened to obtain a specific water content. Each seed was loaded with compression force with respectively growing values: 41N, 68N, 95N, 122N, 149N, 176N, 203N and 230N. A four-layer network of Perceptron type with 10 neurons in the first and 8 neurons in the second hidden layer was selected as a model which the best defines the contact area of grain seeds loaded with axial force at various moisture levels. This network has 4 inputs (water content, pressure force, thickness and length of caryopses) and one output (elementary contact area of rye and wheat seeds). Comparison of the neural model with empirical formulas obtained from nonlinear estimation proved a considerable higher precision of the first one.