Comparison of LMI Solvers for Robust Control of a DC-DC Boost Converter

This work deals with a robust Fault-Tolerant Control (FTC) design for a class of uncertain systems. Fault resilience is associated with a robustness bound generated by a sufficient Linear Matrix Inequality (LMI) condition for static state feedback stabilization. This design control approach is based on solving an optimization problem expressed in terms of LMI with three different programming solvers which are mincx (MATLAB), lmisolver (SCILAB) and cvxopt (PYTHON). Numerical validations were carried out, first on an academic model, then on the model of a PV energy conversion system connected to a DC-DC boost converter. Then, a robustness analysis for fault resilience associated with a control law gains, obtained using the three solvers, was realized to investigate the best performance. This approach was finally validated on an experimental test bench.