Fast and Adaptive Loss of Field Detection in Synchronous Generators Using Real-Time Monitoring and Genetic Algorithm

Loss of Field (LOF) is a common fault in synchronous machine operation, often resulting from short circuits in the field winding or inadvertent tripping of the field breaker. Effective protection of synchronous generators against LOF is crucial due to its potential to cause generator and grid instability, reduced operational efficiency, and significant machine damage. This paper presents an advanced method for initiating a trip signal to the Loss of Excitation (LOE) relay in synchronous generators, utilizing real-time data monitoring and adaptive techniques for reliable detection. The proposed method leverages variations in rotor angle, generator voltage, and reactive power to promptly identify LOF conditions, enabling rapid implementation of protective actions. A Genetic Algorithm (GA) is employed to optimize the fitness function, while MATLAB/Simulink is used to model both the generator and its excitation system (ST1A exciter). Simulation results demonstrate that this method significantly outperforms existing approaches in terms of detection speed, accuracy, and reliability. This advancement enhances the protection of synchronous generators against power swing phenomena and other system disturbances, thereby contributing to the robustness and stability of power networks.