An efficient DLN2-CRSO approach based dynamic stability enhancement in micro-grid system

Abstract

An efficient regulation of the frequency and voltage of operation of the microgrid in island mode is essential; otherwise, the voltage and frequency deviations in the microgrid will increase the energy imbalance by going beyond the tolerance range. In this manuscript, an efficient hybrid approach to improve the dynamic stability of the micro grid (MG) is proposed. The proposed hybrid approach is the combined performance of Deep learning neural network (DLNN) and chaotic rat swarm optimizer (CRSO), hence it is named as DLN2-CRSO. To improve the searching ability and speed of the convergence, RSO is improved by the chaotic version of RSO. The proposed DLN2-CRSO approach is utilized the low-signal stability analysis under the entire operating points. The major objective of the proposed approach is the stability enhancement. To improve the dynamic stability through the integration of multiple distributed generation units in the micro grid, deep planning neural network is used. The control parameters of the DPNNs are identified by the chaotic rat swarm optimizer (CRSO) approach. The proposed approach reduces the reactive power mismatches among the converters and guarantees the stable operation of MG. The performance of proposed approach is authenticated with simulation outcomes, which is performed with MATLAB/Simulink platform. The performance of the proposed technique is evaluated with statistical measures like mean, median and standard deviation with existing methods. Also the efficiency of the proposed technique reaches 95.6346%.