Tuning strategies for control of integrating systems with model predictive control
Hakala, Erik (2022)
2022
Julkaisu on tekijänoikeussäännösten alainen. Teosta voi lukea ja tulostaa henkilökohtaista käyttöä varten. Käyttö kaupallisiin tarkoituksiin on kielletty.
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi-fe2022062349009
https://urn.fi/URN:NBN:fi-fe2022062349009
Tiivistelmä
The objective of this thesis was to verify the suitability of Valmet’s own MPC solution, DNAMPC2, when applied to a common system encountered in process industry and biofuels processing. An extensive literature study was conducted to increase the understanding of MPC algorithms. The study comprised of the entire MPC control algorithm, from the modelling to the solvers used for computing the MPC control law.
The MPC was applied to a simulated falling film evaporator used for the drying of crude tall oil in a case study. The evaporator simulator was created as a part of the thesis and was implemented in the Valmet DNA system. The simulator is based on a detailed mathematical model of the evaporator.
For the case study, in total three different controllers, with two different control structures were designed and tested via simulations. For the testing, two separate testing sequences were designed to verify the MPCs ability of disturbance rejection. For a quantitative analysis of the controllers, four different error integrals were calculated based on the simulation results.
The MPC was applied to a simulated falling film evaporator used for the drying of crude tall oil in a case study. The evaporator simulator was created as a part of the thesis and was implemented in the Valmet DNA system. The simulator is based on a detailed mathematical model of the evaporator.
For the case study, in total three different controllers, with two different control structures were designed and tested via simulations. For the testing, two separate testing sequences were designed to verify the MPCs ability of disturbance rejection. For a quantitative analysis of the controllers, four different error integrals were calculated based on the simulation results.