The Process of Industrial Bioethanol Production Explained by Self-Organised Maps
Bioethanol is produced on an industrial scale by means of fermentation of a sugar substrate by Saccharomyces cerevisiae. Models for the detection of anomalies and their possible evolution are difficult to elaborate due to the biological nature of the ferm
- PDF / 162,970 Bytes
- 11 Pages / 439.37 x 666.142 pts Page_size
- 21 Downloads / 154 Views
The Process of Industrial Bioethanol Production Explained by Self-Organised Maps Miguel A. Sanz-Bobi, Pablo Ruiz, and Julio Montes
Abstract Bioethanol is produced on an industrial scale by means of fermentation of a sugar substrate by Saccharomyces cerevisiae. Models for the detection of anomalies and their possible evolution are difficult to elaborate due to the biological nature of the fermentation process. This paper describes a method able to characterize patterns for explaining industrial bioethanol production using self-organised maps. Also, this method allows for an estimation of the probabilities of evolution to any pattern that the process may have from its last recognized state, therefore helping to take measures to correct a possible problem as soon as possible.
1.1
Introduction
Bioethanol is produced on an industrial scale by means of fermentation of a sugar substrate by Saccharomyces cerevisiae [1]. This is a biological process which is difficult to model with accuracy in order to detect some anomaly in the process as soon as possible. Industrial bioethanol production from sugar based products is a biochemical process that can be affected by different events that may result in a decrease of its performance and contamination which has to be prevented. All these possible problems can cause important costs and undesired shortages. New methods for a better monitoring of the process are being proposed [2]. This paper describes a method based on the analysis and investigation of real data taken from the process of bioethanol production in order to detect a deviation as soon as possible with respect to the normal evolution expected for the process. The method proposed is based on the identification of typical patterns using self-organised maps from historical information observed during the normal process of ethanol production.
M.A. Sanz-Bobi (*) • P. Ruiz • J. Montes IIT- Institute of Technological Research, Engineering School, Comillas Pontifical University, Madrid, Spain e-mail: [email protected] A. Madureira et al. (eds.), Computational Intelligence and Decision Making: Trends and Applications, Intelligent Systems, Control and Automation: Science and Engineering 61, DOI 10.1007/978-94-007-4722-7_1, # Springer Science+Business Media Dordrecht 2013
1
2
M.A. Sanz-Bobi et al.
The paper is organised in the following sections. First, a brief description of bioethanol production is included. This is followed by a description of the main source of information used for the identification of performance patterns. The next sections describe the methods followed to identify and use the patterns found in the propagation and fermentation processes. Finally, a set of conclusions has been included in the last section.
1.2
Brief Description of Bioethanol Production
The process for the industrial production of bioethanol is similar to that of any industrial production of an alcoholic beverage, but in order to give some idea to the reader, a brief description follows. Essentially, industrial bioethanol is ob
Data Loading...
