Learning FCMs with multi-local and balanced memetic algorithms for forecasting industrial drying processes
| Autor: | Antonio Ruiz-Celma, Jose L. Salmeron, Angel Mena |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2017 |
| Předmět: |
0209 industrial biotechnology
Computer science Process (engineering) Cognitive Neuroscience Evolutionary algorithm 02 engineering and technology Cognitive Maps Machine Machine learning computer.software_genre Fuzzy logic 020901 industrial engineering & automation Local optimum Artificial Intelligence 0202 electrical engineering electronic engineering information engineering Local search (optimization) Industrial drying business.industry Memetic algorithm Fuzzy cognitive map Computer Science Applications 020201 artificial intelligence & image processing Artificial intelligence business computer Fuzzy |
| Zdroj: | Arias Montano. Repositorio Institucional de la Universidad de Huelva instname |
| Popis: | In this paper, we propose a Fuzzy Cognitive Map (FCM) learning approach with a multi-local search in balanced memetic algorithms for forecasting industrial drying processes. The first contribution of this paper is to propose a FCM model by an Evolutionary Algorithm (EA), but the resulted FCM model is improved by a multi-local and balanced local search algorithm. Memetic algorithms can be tuned with different local search strategies (CMA-ES, SW, SSW and Simplex) and the balance of the effort between global and local search. To do this, we applied the proposed approach to the forecasting of moisture loss in industrial drying process. The thermal drying process is a relevant one used in many industrial processes such as food industry, biofuels production, detergents and dyes in powder production, pharmaceutical industry, reprography applications, textile industries, and others. This research also shows that exploration of the search space is more relevant than finding local optima in the FCM models tested. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|