Note: This content is accessible to all versions of every browser. However, this browser does not seem to support current Web standards, preventing the display of our site's design details.


Design of Fast Model Predictive Control for Electric-Driven Compressors


T. Robbiani

Bachelor Thesis, FS16 (10488)

Centrifugal compressors are widely employed in industry, in chemical processes and to deliver gas in pressure. Oil and gas companies need to pump continuously tons of gas for long distances, and this operation is normally performed by some compressors on the line. Even a slight improvement in the efficiency implies a big saving in terms of absolute losses, considering the high powers involved. Moreover external conditions can lead to process instabilities, as classic and deep surge, that can cause irreversible damages to the machine. Model Predictive Control approaches controlling drive torque actuation and emergency valves have shown promising results, reducing the risk of occurrence of the aforementioned effects and improving the performance of the controller. On the other hand, the compressor model is nonlinear and this influences the MPC computational time. In this project we implement a nonlinear MPC controller with the tool FORCES Pro and a Real-Time Iteration MPC generated with ACADO toolkit and we show that computational times stay within the required sampling time. Moreover, reduced computational times obtained with standard MPCs, makes possible to run offset-free MPCs, that while allowing perfect steady-state tracking of the desired signal, they increase the computational effort burden with additional estimation and stateinput reference computation. In this project we demonstrate that FORCES PRO and ACADO toolkit generated codes solve the nonlinear optimization problem as fast as other approaches and so they can be valid alternatives in industrial embedded systems implementations.

Supervisors: Giampaolo Torrisi, Manfred Morari


Type of Publication:

(13)Semester/Bachelor Thesis

File Download:

Request a copy of this publication.
(Uses JavaScript)
% Autogenerated BibTeX entry
@PhdThesis { Xxx:2016:IFA_5453
Permanent link