A Time-Delay Smart Grid Communication Optimization Model for Transient Fault Tracing and Load Management

This paper developed faster communication grid architecture for signaling and information sharing, with robust distributed cloud management architecture that links the processes for both faults and load management. Experiments were carried out on smart grid (SG) layered time-delay optimization model (LTDOM) using schemes such as SG Neural Network Algorithm (proposed technique), SG stackelberg game algorithm (SGSGA), SG chaos-flower pollination algorithm (SGCFPA), SG cuckoo search algorithm (SGCSA), SG differential search algorithm (SGDSA), and SG cournot algorithm (SGCA) were used for validation of the study. Riverbed Modeller software academic version 17.5 was used to setup the experimental design for LTDOM architecture and the various algorithms were developed with C++ to achieve the simulation test-bed. Smart grid metrics such as energy data received, service delays, media access delays and service throughput were used for performance evaluation. For example, the service throughput results showed that the proposed SGNNLA, SGSGA, SGCFPA, SGCSA, SGDSA and SGCA had 24.09%, 21.68%, 16.86%, 15.66%, 12.04% and 9.63% respectively. This implies that as load demands in the peak periods is being shifted to the off-peak periods, the proposed SGNNLA utilized optimum resources while delivering satisfactorily on the grid network when compared to other schemes.