Genetic Algorithm Tuning of PID Controller in Smith Predictor for Glucose Concentration Control
This paper focuses on design of a glucose concentration control system based on nonlinear model plant of E. coli MC4110 fed-batch cultivation process. Due to significant time delay in real time glucose concentration measurement, a correction is proposed in glucose concentration measurement and a Smi...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Bulgarian Academy of Sciences
2011-07-01
|
| Series: | International Journal Bioautomation |
| Subjects: | |
| Online Access: | http://www.clbme.bas.bg/bioautomation/2011/vol_15.2/files/15.2_03.pdf |
| _version_ | 1811336276694007808 |
|---|---|
| author | Tsonyo Slavov Olympia Roeva |
| author_facet | Tsonyo Slavov Olympia Roeva |
| author_sort | Tsonyo Slavov |
| collection | DOAJ |
| description | This paper focuses on design of a glucose concentration control system based on nonlinear model plant of E. coli MC4110 fed-batch cultivation process. Due to significant time delay in real time glucose concentration measurement, a correction is proposed in glucose concentration measurement and a Smith predictor (SP) control structure based on universal PID controller is designed. To reduce the influence of model error in SP structure the estimate of measured glucose concentration is used. For the aim an extended Kalman filter (EKF) is designed. To achieve good closed-loop system performance genetic algorithm (GA) based optimal controller tuning procedure is applied. A standard binary encoding GA is applied. The GA parameters and operators are specified for the considered here problem. As a result the optimal PID controller settings are obtained. The simulation experiments of the control systems based on SP with EKF and without EKF are performed. The results show that the control system based on SP with EKF has a better performance than the one without EKF. For a short time the controller sets the control variable and maintains it at the desired set point during the cultivation process. As a result, a high biomass concentration of 48.3 g·l-1 is obtained at the end of the process. |
| first_indexed | 2024-04-13T17:37:05Z |
| format | Article |
| id | doaj.art-8b952e1031954bb097c700b57044e6d4 |
| institution | Directory Open Access Journal |
| issn | 1314-1902 1314-2321 |
| language | English |
| last_indexed | 2024-04-13T17:37:05Z |
| publishDate | 2011-07-01 |
| publisher | Bulgarian Academy of Sciences |
| record_format | Article |
| series | International Journal Bioautomation |
| spelling | doaj.art-8b952e1031954bb097c700b57044e6d42022-12-22T02:37:20ZengBulgarian Academy of SciencesInternational Journal Bioautomation1314-19021314-23212011-07-01152101114Genetic Algorithm Tuning of PID Controller in Smith Predictor for Glucose Concentration ControlTsonyo SlavovOlympia RoevaThis paper focuses on design of a glucose concentration control system based on nonlinear model plant of E. coli MC4110 fed-batch cultivation process. Due to significant time delay in real time glucose concentration measurement, a correction is proposed in glucose concentration measurement and a Smith predictor (SP) control structure based on universal PID controller is designed. To reduce the influence of model error in SP structure the estimate of measured glucose concentration is used. For the aim an extended Kalman filter (EKF) is designed. To achieve good closed-loop system performance genetic algorithm (GA) based optimal controller tuning procedure is applied. A standard binary encoding GA is applied. The GA parameters and operators are specified for the considered here problem. As a result the optimal PID controller settings are obtained. The simulation experiments of the control systems based on SP with EKF and without EKF are performed. The results show that the control system based on SP with EKF has a better performance than the one without EKF. For a short time the controller sets the control variable and maintains it at the desired set point during the cultivation process. As a result, a high biomass concentration of 48.3 g·l-1 is obtained at the end of the process.http://www.clbme.bas.bg/bioautomation/2011/vol_15.2/files/15.2_03.pdfE. coli cultivationSmith predictorPID controller tuningGenetic algorithmExtended Kalman filter |
| spellingShingle | Tsonyo Slavov Olympia Roeva Genetic Algorithm Tuning of PID Controller in Smith Predictor for Glucose Concentration Control International Journal Bioautomation E. coli cultivation Smith predictor PID controller tuning Genetic algorithm Extended Kalman filter |
| title | Genetic Algorithm Tuning of PID Controller in Smith Predictor for Glucose Concentration Control |
| title_full | Genetic Algorithm Tuning of PID Controller in Smith Predictor for Glucose Concentration Control |
| title_fullStr | Genetic Algorithm Tuning of PID Controller in Smith Predictor for Glucose Concentration Control |
| title_full_unstemmed | Genetic Algorithm Tuning of PID Controller in Smith Predictor for Glucose Concentration Control |
| title_short | Genetic Algorithm Tuning of PID Controller in Smith Predictor for Glucose Concentration Control |
| title_sort | genetic algorithm tuning of pid controller in smith predictor for glucose concentration control |
| topic | E. coli cultivation Smith predictor PID controller tuning Genetic algorithm Extended Kalman filter |
| url | http://www.clbme.bas.bg/bioautomation/2011/vol_15.2/files/15.2_03.pdf |
| work_keys_str_mv | AT tsonyoslavov geneticalgorithmtuningofpidcontrollerinsmithpredictorforglucoseconcentrationcontrol AT olympiaroeva geneticalgorithmtuningofpidcontrollerinsmithpredictorforglucoseconcentrationcontrol |