Abstract
Summary of key findings
We found a greybox model for state estimation and control of the BioDenitro process based on a reduced ASM1. We then applied Maximum Likelihood Estimation on measurements from a real full-scale waste water treatment plant to estimate the model parameters. The estimation method also incorporates the Extended Kalman Filter that provides estimates of any unmeasured states, e.g. the NH4 and NO3 concentrations in both aeration tanks, and more importantly, the NH4 inlet concentration. This will improve control performance without the need for extra sensors and improve forecasts of the load.
We found a greybox model for state estimation and control of the BioDenitro process based on a reduced ASM1. We then applied Maximum Likelihood Estimation on measurements from a real full-scale waste water treatment plant to estimate the model parameters. The estimation method also incorporates the Extended Kalman Filter that provides estimates of any unmeasured states, e.g. the NH4 and NO3 concentrations in both aeration tanks, and more importantly, the NH4 inlet concentration. This will improve control performance without the need for extra sensors and improve forecasts of the load.
| Original language | English |
|---|---|
| Publication date | 2015 |
| Number of pages | 3 |
| Publication status | Published - 2015 |
| Event | 9th IWA Symposium on Systems Analysis and Integrated Assessment (Watermatex 2015) - Gold Coast, Queensland, Australia Duration: 14 Jun 2015 → 17 Jun 2015 Conference number: 9 http://www.awmc.uq.edu.au/conf/watermatex2015 |
Conference
| Conference | 9th IWA Symposium on Systems Analysis and Integrated Assessment (Watermatex 2015) |
|---|---|
| Number | 9 |
| Country/Territory | Australia |
| City | Gold Coast, Queensland |
| Period | 14/06/2015 → 17/06/2015 |
| Internet address |
Bibliographical note
Extended abstract to be presented as poster at WatermatexKeywords
- WWTP
- Greybox
- System Identification