A Smart Residential Security Assisted Load Management System using Hybrid Cryptography

Abstract A residential load management system equips smart meters (SMs) to measure load utilization at residencies. SM reports electricity usage based on electronic appliances. In this paper, a smart residential load management system is designed by three-fold along with the provisioning of consumer security. Load management encompasses load categorizing by Hopfield neural network, fuzzy-logic based bill payments identification and load state prediction using Markov chain. Residential load is based on three classes: active load, affordable load and inactive load. The estimation of residential load ensures to manage the load at residency either to turn off the load or intimate excess consumption of electricity. The registered consumers of a specific residency are enabled to receive load status by Internet of Things (IoT) devices. SMs at residencies periodically compute the electricity manipulation and those readings are encrypted using hybrid blowfish and elliptic curve cryptography algorithm with considering the behavior information from IoT device and partial key storage assists security, even if the device is theft. This smart residential security assisted load management (SRS-LM) system is developed in network simulator 3 and the results showed improvements in terms of power usage, load power, peak load reduction, total power consumption, power cost and computation time.