A hybrid machine learning framework for modelling and control of particle processes using on-line/at-line particle analysis and other at-line/on-line process sensors. ParticleModel is implemented in Python, using the TensorFlow framework.
The framework consists of several modules outlined by the following:
- Data module: Data structure for storing time-series data
- Design of experiments module: Module for generating design of experiments
- Domain module: Domain module for discretization of particle distributions
- Reference model module: First principles reference models for testing of framework
- Hybrid model module: Hybrid modelling framework
- Process control module: Process control structures for particle processes
Python > 3.7 is required for this code to work. It is recommended so set up an individual python enviroment for this installation.
The necessary python packages and versions can be found in requirements.txt. To install all packages in one go, use the following pip-command:
pip install -r requirements.txt
- R. F. Nielsen, N. A. Kermani, L. la Cour Freiesleben, K. V. Gernaey, S. S. Mansouri, Novel strategies for predictive particle monitoring and control using advanced image analysis, in: A. A. Kiss, E. Zondervan, R. Lakerveld, L. zkan (Eds.), 29th European Symposium on Computer Aided Process Engineering, volume 46 of Comput. Aided Chem. Eng., Elsevier, 2019, pp. 1435-1440. doi:10.1016/B978-0-12-818634-3.50240-X.
- R. F. Nielsen, N. Nazemzadeh, L. W. Sillesen, M. P. Andersson, K. V. Gernaey, S. S. Mansouri, Hybrid machine learning assisted modelling framework for particle processes, Comput. Chem. Eng., Elsevier, 2020 (accepted)