Kai Lui
Using recurrent neural networks to model fed-batch processes.
Email: [email protected]
Supervisors
Project description
Fed-batch processes are used to produce high value products in the chemical, biological, food, pharmaceutical and semiconductor industries. The general features of fed-batch biological processes include:
- strong nonlinearity
- no steady state operation
- instinctive time variation
- batch-to-batch variation
- uncertainty caused by drifting of raw materials
These features complicate modelling and control.
Recurrent neural networks (RNNs) have a dynamic memory. They can process temporal context information. They are highly promising tools used for solving complex temporal, nonlinearity, time variation and uncertainty tasks.
In this project, I will use RNNs to model fed-batch processes. I will optimise them using a covariance matrix adaption evolutionary strategy.
Publications
- Liu K, Zhang J. Optimization of Echo State Networks by Covariance Matrix Adaption Evolutionary Strategy. In: 2018 24th International Conference on Automation and Computing (ICAC) 2018, 1-6. IEEE.