Intelligent mitigation of blackout in real-time microgrids: Neural Network Approach

In this paper, a novel application of Artificial Neural Networks (ANN) is deployed to prevent blackout in a microgrid after N-1-1 contingency condition. In fact, microgrids are vulnerable to disturbances and abnormal conditions due to their inherent small inertia. Therefore, stability of microgrids after a disturbance turns into a challenge in power systems. The key contribution of this paper is to utilize the artificial intelligence concept to prevent cascading failure practically at early stages and to make microgrids more reliable and robust by intelligent and adaptive re-dispatch of power. The proposed ANN control approach is tested on an experimental testbed microgrid. Experimental results verify the robustness, accuracy, and effectiveness of the ANN method for preventing cascading failure in addition to providing voltage and frequency stability after initiation of a disturbance.