QoS based scheduling mechanism for electrical vehicles in cloud-assisted VANET using deep RNN

Abstract

A charge scheduling strategy is a robust approach to schedule the charging strategies in electric vehicles (EVs) from a broad perspective with the aim of evading the overloading of charging stations and enhancing energy efficiency. However, devising an effective charging scheduling schemefor attaining optimal energy consumption still prevails as a complicated problem, particularly while considering the synchronized behavior of both charging stations as well as EVs. Here, a robust QoS-based charge scheduling approach was developed, which exploits the vehicular Adhoc networks (VANETs) with the improved functionalities for enabling communication between the vehicle-traffic server, road-side units (RSUs), and various EVs on roads. An optimal routing is performed by the Fractional-social sky driver (Fractional SSD), which is devised by the incorporation of the Fractional calculus (FC) and social sky driver (SSD) optimization. Here, the multi-objectives, namely, distance, battery power, and predicted traffic density are considered where the traffic density is effectively predicted using deep recurrent neural network (Deep RNN). Then, the charge scheduling process is executed by the utilization of the developed optimization technique called Fractional-social water cycle algorithm (Fractional SWCA)-based scheduling algorithm by taking into account the QoS-based fitness objective, likepriority, response time, and latency. Moreover, the proposed Fractional SWCA is developed by the integration of fractional SSD and water cycle algorithm (WCA). The performance of the devisedscheme is evaluated withmeasures, like metrics, delay, traffic density, fitness, total trip time, percentage of successful allocation, and power with the values of 8.429 min, 4.8 per lane, 24.571, 49.421 min, 94.494%, and 14,135.72 J.