A comprehensive real-time intelligent framework for hybrid power flow algorithm development in smart grids with electric vehicle integration

Document Type

Article

Publication Title

Energy Conversion and Management X

Abstract

Energy is fundamental for efficient system functioning, and with the evolution of power grids, techniques for generation, transmission, and storage have become increasingly sophisticated. The advent of smart grids, integrating renewable energy sources, electric vehicles (EVs), and vehicle-to-grid (V2G) technology, has created opportunities for demand–supply optimization but also introduced challenges such as cyber vulnerabilities, stochastic EV behavior, and renewable intermittency. This research presents the Dynamic-Anomaly Responsive Optimization Framework (DAROF), a mathematically modelled, machine learning-based framework designed to dynamically identify power flow disturbances and generate hybrid power flow algorithms in real time. DAROF combines linear, non-linear, and quadratic programming with evolutionary methods including Genetic Algorithm, Particle Swarm Optimization, and Simulated Annealing, selecting and hybridizing them based on operational context. The framework integrates stochastic EV arrival/departure modelling, battery degradation costs, smart/V2G charging, and real-world constraints like latency and cyber-attacks. Computational benchmarking across IEEE 14-, 57-, and 118-bus systems revealed up to 65 % runtime improvement over conventional methods, while advanced case studies (urban vs rural grids, residential vs fleet EVs, varying renewable penetration) validated its scalability. Sensitivity and robustness analysis confirmed > 90 % stability under high EV penetration, establishing DAROF as a resilient and future-ready solution for smart grid–EV integration.

DOI

10.1016/j.ecmx.2025.101336

Publication Date

10-1-2025

This document is currently not available here.

Share

COinS