Use of Neuroevolution to Estimate the Melting Point of Ionic Liquids
The P hysical Properties Estimation Problem of Ionic Liquids (PPEP I L s ) arises from the need of designing Ionic Liquids (ILs) for specific tasks . It is important to emphasize that the synthesis of ILs is generally expensive and time - consuming . Furthermore, the number of possible ionic liquids...
| Autores: | , , , , |
|---|---|
| Tipo de recurso: | artículo |
| Estado: | Versión publicada |
| Fecha de publicación: | 2017 |
| País: | México |
| Institución: | Tecnológico Nacional de México |
| Repositorio: | Redalyc-TNM |
| OAI Identifier: | oai:redalyc.org:265253896002 |
| Acceso en línea: | https://www.redalyc.org/articulo.oa?id=265253896002 |
| Access Level: | acceso abierto |
| Palabra clave: | Computación QSPR Melting Point Ionic Liquids |
| Sumario: | The P hysical Properties Estimation Problem of Ionic Liquids (PPEP I L s ) arises from the need of designing Ionic Liquids (ILs) for specific tasks . It is important to emphasize that the synthesis of ILs is generally expensive and time - consuming . Furthermore, the number of possible ionic liquids that can be synthesized is extremely large . The purpose of PPEP I L s is to avoid the experimental synthesis of Ionic Liquids (ILs) estimating their physical properties . Moreover, to estimate the melting temperature is the most difficult task . This problem has attracted the attention of interdisciplinary researchers due to their relevant applications such as their usages as catalysts and solvents. Additionally, the ILs are relevant due to their distinctive characteristics and reduced toxicity. This problem is particularly complex since the behavior of ILs is unconventional a nd the available information may not be accur ate. This paper presents a new approach for the PPEPILs based on neuroevolutionary neural networks using molecular descriptors to predict the melting temperatures of ILs with encouraging results. N euroevolutionary networks had been previously used in diverse areas of knowledge and present advantages over classic Neural Networks. |
|---|