Modellierung des Wärmeübergangs beim Blasensieden in freier Konvektion mit künstlichen neuronalen Netzen : Teil II ; Einstoffsysteme
- An artificial neural network (ANN) type multilayerperceptron with sigmoid activation function and errorbackpropagation learning algorithm was applied to predict heat transfer performances in a wide range of pressure and wall superheat at bubble boiling in free convection of two single component systems. For both systems the ANN prediction results were compared with the commonly used conventional correlation and gave in one case a comparable in the other a better accuracy. To improve the accuracy of ANN prediction it is possible to use interpolated experimental data for training and to optimise the network configuration.
| URN: | urn:nbn:de:kobv:526-opus4-7046 |
|---|---|
| DOI: | https://doi.org/10.15771/0949-8214_2000_2_6 |
| Author: | Gerardo Diaz, Josef Schmadl, Thomas Schutt, Mihir Sen |
| ISSN: | 0949-8214 |
| Parent Title (German): | Wissenschaftliche Beiträge 2/2000 |
| Document Type: | Article in a Periodical of the TH Wildau |
| Language: | German |
| Year of Publication: | 2000 |
| Volume: | 5 |
| Issue: | 2 |
| First Page: | 36 |
| Last Page: | 42 |
| Faculties an central facilities: | Fachbereich Ingenieurwesen / Wirtschaftsingenieurwesen (bis 8/2014) |
| Publishing Institution: | Technische Hochschule Wildau |
| Dewey Decimal Classification: | 0 Informatik, Informationswissenschaft, allgemeine Werke / 00 Informatik, Wissen, Systeme / 006 Spezielle Computerverfahren |
| TH Wildau publications: | Wissenschaftliche Beiträge |
| Licence (German): | Das Dokument ist urheberrechtlich geschützt. |
| Release Date: | 2017/01/17 |



