Hybrid adaptive fault-tolerant control algorithms for voltage and frequency regulation of an islanded microgrid

This paper presents new design methodology and performance comparison of two hybrid fault-tolerant control (FTC) schemes applied to the regulation of the frequency and voltage amplitude of a diesel engine generator installed in a microgrid. Both of them are based on a unique combination of a model reference adaptive control (MRAC) with a proportional-integral-derivative (PID) controller and artificial intelligence techniques, i.e. artificial neural networks (ANN) and genetic algorithms (GA). Since in an islanded microgrid, the frequency of the system is determined by the shaft speed of the diesel engine (DE), while the voltage amplitude is set by the synchronous generator (SG) field voltage, therefore, two FTC systems for frequency and voltage regulation have been implemented in each proposed control scheme. The first scheme consists of an MRAC system with a PID controller tuned by a GA for controlling the speed of the DE and a classic MRAC system for controlling the voltage amplitude of the SG. In the second scheme, an MRAC system with a GA-tuned PID controller is used for the DE, and a hybrid controller in which the MRAC is combined with an ANN and a PID controller tuned by a GA is designed for the SG. The dynamic models of the microgrid components are presented in detail, and the proposed microgrid and its FTC systems are implemented and tested in the Simpower Systems of MATLAB/Simulink® simulation environment. All results indicate high effectiveness and robustness of the MRAC-based FTC schemes in both normal and emergency/faulty operations of the microgrid in comparison with a benchmark baseline controller, IEEE Type 1 AVR for the SG and a PID controlled governor for the DE.