Multivariable robust H∞ control for aeroengines using modified Particle Swarm Optimization algorithm

The increasing stringent performance requirements on aeroengines appeal for more facile optimization design approaches to robust control systems. We propose a multivariable robust H_∞ controller optimization design technique for aeroengines using a modified Particle Swarm Optimization (PSO) algorithm. The control structure of aeroengines with 4 inputs and 4 outputs is built according to general principles of aeroengine operation and variable selection, and thus the linearized state-space models of an aeroengine under the condition of small perturbation is established, which fit well with the data of nonlinear model and are suitable for controller design. The robust H_∞ controller design is optimized by using a modified particle swarm optimization algorithm, which is formulated as a multi-objective optimization problem characterized by searching for the optimal parameters of the three weighting functions. An Adaptive mutation based PSO (AMBPSO) algorithm is proposed for the improvement of the search accuracy and convergency of the standard PSO algorithm, which is featured by modification of the inertia weight with gradient descent and adaptive mutation of the velocities and positions of the particles.