| Autor: |
Artemenko, Sergiy, Mazur, Viktor |
| Předmět: |
|
| Zdroj: |
Refrigeration Engineering & Technology; 2021, Vol. 57 Issue 3, p138-146, 9p |
| Abstrakt: |
The interdisciplinary nature of new objectives aimed at the design of the working matters for environmentally friendly technologies requires a more dynamic use of information technology (IT) to ensure proper trade-off decisions under a competitive environment. Machine learning (ML) is the part of artificial intelligence (AI) methodologies that uses algorithms that are not a direct solution to a problem but learning through solutions to innumerable similar problems. Machine learning has explored a new path in the study of the thermodynamic behavior of emerging substances. Various computational tools have been provided with an effective approach to solving the actual problem of predicting the phase behavior of soft substances under strong exogenous influences. The aim of this study is to develop a new perspective on predicting the thermodynamic properties of soft substances using a methodology that provides artificial neural networks (ANN) and a global phase diagram to ensure correlation between structure and properties. In this study, we present applications of machine learning in engineering thermodynamics to predict azeotropic behaviour of binary refrigerants and determine the coefficient of performance (COP) for Organic Rankine Cycle (ORC) working media based on the data on boiling and critical points was studied. A new approach to predicting the formation of an azeotropic state in a mixture, which is developed and presented. This approach uses the synergy of neural networks and the global phase diagram methodology to correlate azeotropic data for binary mixtures based only on the critical properties and the centric coefficient of the individual components in refrigerant mixtures. It does not require intensive calculations. The construction of ANN correlations between the information attributes of working fluids and the criteria for the efficiency of the Rankin cycle narrows the scope of trade-offs in the space of competitive economic, environmental and technological criteria. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
