Publication details
Emission prediction of a thermal power plant
| Authors | |
|---|---|
| Year of publication | 2014 |
| Type | Article in Proceedings |
| Conference | Znalosti 2014 |
| MU Faculty or unit | |
| Citation | |
| Field | Informatics |
| Keywords | meta-learning; model prediction; boiler; NOx |
| Description | The task of prediction of emissions is very challenging and also important. We argued that simple learning techniques that learn only one predictive model are not powerful enough in more complex situations. Better predictive results can be achieved by splitting data into smaller parts and for each part to learn a sub-model. We proposed and tested a novel method that combines meta-learning and ensemble learning. We showed that there is significant increase in prediction accuracy. |