The code in this repository is based on the works in [1][2] and is written in Matlab. It is a variation of a modifier adapation algorithm using Gaussian processes (GP-MA). To cite GP-MA please use the publications [1][2].
Download the entire folder containing all Matlab files and add the entire folder and subfolders to the Matlab path. Next run RTO_MA_GP, which should run the pre-defined problem as defined in [2]. Once this works the "true" plant model can be modified in Plant_model with the corresponding objective defined in Plant_data. The "approximate" plant model can be modified in Approx_model with the corresponding approximate objective defined in Approx_data. Note the plotting only works for two dimensional problems.
[1] E.A. del Rio Chanona, J.E. Alves Graciano, E.Bradford, and B.Chachuat, Modifier-Adaptation Schemes Employing Gaussian Processes and Trust Regions for Real-Time Optimization, IFAC-PapersOnLine, vol. 52, no. 1, pp. 52-57, 2019.
[2] T.A. Ferreira, H.A. Shukla, T. Faulwasser, C.N. Jones, and D. Bonvin, Real-time optimization of uncertain process systems via modifier adaptation and Gaussian process, In European Control Conference (ECC) 2018, pp. 465-470.
This project is licensed under the MIT license – see LICENSE.md in the repository for details.