ПРОИЗВОДИТЕЛЬНОСТЬ СИЛ ТРАЛОВОЙ СИСТЕМЫ – II: ФИЗИЧЕСКОЕ МОДЕЛИРОВАНИЕ

Abstract

The article is a continuation of scientific research and justification of the possibility of artificial intelligence technologies for the tasks of predictive modeling of the behavior of a trawl system in the process of fishing on a self-learning neural network. The definition of the productivity of forces is introduced - the second time derivative of the work of these forces. The intermediate result of the design of the trawl system is a project - an integrated set of characteristics described in a form suitable for its operation with a given performance of forces. To proceed to predictive modeling, it is necessary to determine the extent of similarity of the trawl system in different areas of its interaction. There is inter-discipline, which is manifested in the formulation of problems, in ap-proaches to their solution, in revealing the connections between theories, in the formation of new disciplines. Interdisciplinarity allows conducting research with the trawl system in its entirety, combining data from various disciplines (hydromechanics, electrodynamics, thermodynamics, acoustics, optics, etc.), leading to the emergence of new postulates and laws that synthesize the sci-entific knowledge necessary for a self-learning neural network of fishing for the trawl system. To combine the knowledge there was chosen the similarity theory as a mathematical modeling method based on the transition from ordinary physical quantities that affect the system being modeled to generalized complex-type quantities composed of original physical quantities, but in certain combinations, depending - from the specific nature of the process under study. The complex nature of these quantities has a deep physical meaning of reflecting the interaction of various influences. The similarity theory studies the methods of constructing and applying these variables and is used in cases of mathematical modeling when an analytical solution of mathematical modeling problems is impossible due to complexity and accuracy requirements. The similarity theory is used in these cases to synthesize relations obtained on the basis of the physical mechanism of the process under study and data of a numerical solution or experiment.