Project Details
Description
Biological wastewater treatment processes using activated sludge may be operated as a periodic cycled plant to enable better possibilities for manipulating the process. The operational task is complicated by the desire to remove Carbon, Nitrogen and Phosphorous to fulfil requirements upon effluent quality. These requirements render this task a typical multivariable control problem. The number of handles in conventional waste water plants is rather limited, therefore there is considerable interest in exploiting the additional degrees of freedom available in periodically operated plants.
The activated sludge process consists of two reactors alternating between aerobic and anoxic conditions. As this process is operated periodically it is highly non-linear. In a first phase of this project a period-to-period linearized model of the plant will be investigated concerning its ability to predict the inter-periodic plant performance. Subsequently, both model verification and state estimation will be investigated. One key aspect is the model detail that is required to obtain reliable predictions several periods ahead. Subsequently it is intended to investigate the achievable inter-periodic control performance of the plant using a linear state space MPC controller.
The activated sludge process consists of two reactors alternating between aerobic and anoxic conditions. As this process is operated periodically it is highly non-linear. In a first phase of this project a period-to-period linearized model of the plant will be investigated concerning its ability to predict the inter-periodic plant performance. Subsequently, both model verification and state estimation will be investigated. One key aspect is the model detail that is required to obtain reliable predictions several periods ahead. Subsequently it is intended to investigate the achievable inter-periodic control performance of the plant using a linear state space MPC controller.
Status | Active |
---|---|
Effective start/end date | 01/05/1997 → … |
Fingerprint
Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.