Open AccessApproximation of hydraulic turbine performance characteristics using optimal splines and neural networks
Author(s) -
A. N. Kovartsev,
A. A. Nazarova,
V E Zakharchenko
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1652/1/012015
Subject(s) - turbine , computer science , boundary (topology) , control theory (sociology) , automation , steam turbine , power (physics) , control engineering , artificial neural network , mathematical optimization , control (management) , mechanical engineering , engineering , mathematics , artificial intelligence , mathematical analysis , physics , quantum mechanics
The performance characteristic of hydrounits is used very often in various control systems on hydro power plant for velocity control, power distribution, for displaying at different levels of automation pyramid: technicians, technologists, management and so on. Different systems do not have single unified description of the hydrounit’s performance characteristic. It leads to growth of unnecessary calculations, errors and collisions. The study of the turbine efficiency can be automated by creating an appropriate mathematical model that approximates the initial performance characteristics of the turbine. This paper describes a method of approximation of the turbine performance characteristics using optimal two-dimensional cubic Hermit splines with free boundary conditions. The proposed method of hydrounit efficiency description allows to create single mathematical model and to use it in any control systems. The calculation load tests show high efficiency of the methods and confirm the possibility of their application in PLCs and servers.