Auteurs: Abdesselam Debbah*1,2, Hamid Kherfane1 & Mahieddine Berkani3
1L2RCS Laboratory, Department of Electronics, University of Badji Mokhtar, Po Box 12, 23000, Annaba, Algeria.
2Department of Petrochemical and Process Engineering, University of 20 Aout 1955, Skikda, 21000, Algeria.
3Departement of Electromechanical, University of Badji Mokhtar, Po Box 12, Annaba, 23000, Algeria.
This paper presents a novel robust approach to control the aerodynamics instabilities, surge and rotating stall of a Variable Speed Axial Compressor (VSAC) in gas turbine process. The used model to design the controller is Egeland-Gravdahl who has been commonly recognized as standard when reporting some bifurcation behavior of VSAC which cannot be captured by constant-speed models. The proposed non-linear robust control is reformulated as an optimization problem, the first reformulation is linear matrix inequality (LMI) solved by interior point optimization algorithm, the second reformulation is non-linear constrained optimization solved by genetic algorithm (GA).The two optimization approaches are tested and compared using Matlab software. Promising simulation results were obtained for both optimization approaches, but the robust controller based on genetic algorithm optimization presents better performances to tackle a restricted specification and operating conditions.
Gas turbine, Surge-Rotating stall, Robust control, Constrained Genetic algorithm (GA).