An empirical wavelet transform based fault detection and hybrid convolutional recurrent neural network for fault classification in distribution network integrated power system

The penetration of distributed renewable energy sources degrades the protection of microgrids, which leads to incorrect data flow in the energy systems. It is critical to detect faults, types of defects and location of faults in order to improve the protection and security of microgrids. To cater this issue in hybrid renewable energy system, a novel fault detection scheme is adopted using artificial intelligence. The renewable energy based microgrid system is implemented in the IEEE 13 bus power network to obtain the normal and faulty voltage and current data.. The system is simulated using MatLab/Simulink platform. From the time series data, the features are decomposed using empirical wavelet transform (EWT). First, EWT evaluates the frequency components in the signal, then calculates the bounds and gets the basis of the oscillating components. The obtained samples are classified using a Hybrid Convolutional Recurrent Neural Network (HCRNN) and optimized by the Pelican Optimization Algorithm. The 11 types of faults are identified along with the location of fault in the system is obtained. The results are compared with the existing methods and found that the proposed method has improved the fault sample detection accuracy by 1.56%.