A comparative study between Neural Networks (NN)-based and adaptive-PID controllers for the optimal bio-hydrogen gas production in microbial electrolysis cell reactor
The main challenge of the hydrogen production study for the MEC reactor is to obtain a good automatic control system due to the nonlinearity and complexity of the microbial interactions. To address this issue an integrated approach involving process modeling, optimization and advanced control has to...
| Main Authors: | , , , |
|---|---|
| Format: | Conference or Workshop Item |
| Language: | English |
| Published: |
2015
|
| Subjects: | |
| Online Access: | http://eprints.um.edu.my/14129/1/A_Comparative_study_between_Neural_Networks.pdf |
| Summary: | The main challenge of the hydrogen production study for the MEC reactor is to obtain a good automatic control system due to the nonlinearity and complexity of the microbial interactions. To address this issue an integrated approach involving process modeling, optimization and advanced control has to be implemented. This work focus on the
controller’s performance in the control system; neural network (NN)-based and Adaptive-PID controllers. The study has been carried out under optimal condition for the production of bio-hydrogen gas wherein the controller output are based on the correlation of the optimal current and voltage to the MEC. A Ziegler–Nichols tuning method and an adaptive gain technique have been used to design the PID controller, while the neural network controller has been designed from the inverse response of the MEC neural network model. |
|---|