Evaluation of diagnostic parameters of complex systems in water transport in terms of the entropy approach

The purpose of the work is to develop and improve the methods of computer modeling and digitalization of the choice of diagnostic parameters of complex technical systems to the maximum informativeness. The entropy method of estimating the technical state of systems is based on the existence of a functional dependence of random variables characterizing the behavior of elements, which is always between two critical states: absolute integrity and complete additivity. The complexity of the interaction of elements in stochastic systems determines the need to search for effective algorithms based on fundamental entropy models and methods for finding the best solutions for monitoring and control. Models and algorithms for estimating the parameters of the technical condition and operational indicators of systems in the process of monitoring are improved on the MATLAB digital platform and its applications. It is noted that for systems of high complexity, a method of control is proposed, based on the entropic assessment of information as the most universal and applicable in technical, biological, economic, social and other systems that are difficult to analyze and formalize. It is shown that the entropic method of selecting the most informative control nodes in the design of diagnostic systems for technical means and technological processes at water transport facilities can significantly increase reliability and ensure the trouble-free operation of controlled facilities. The entropy method of estimating information content, based on the difference in entropies of the initial and subsequent states of the system, is used to solve specific problems given in the examples. An algorithm for ranking diagnostic parameters based on the method of pair comparisons is proposed, on the basis of which the minimax problem is solved: the maximum information content of the selected diagnosable nodes of the system is ensured with a minimum set of controlled parameters.