Stochastic methods for prediction of charging and discharging power of electric vehicles in vehicle-to-grid environment

Abstract

As the penetration rate of the electric vehicles (EVs) increases, their uncontrolled charging could cause undervoltage and network congestion in the electric network. To mitigate these impacts, the controlled charging of the EVs has been investigated by earlier publications. However, controlled charging cannot be easily implemented as it involves multiple customers having individual interests. To overcome these drawbacks, the power prediction of charging and discharging of EVs plays a major role. A new realistic power prediction algorithm that accounts for the requirements of different patterns and consumers is developed in this study. The main objective of the study is to develop a charging and discharging coordination algorithm that effectively addresses the problem of power demand during peak time. Stochastic methods were used to develop the charging-discharging models and estimate the EV usage. The proposed algorithm aims to manage high power demands at peak times using vehicle-to-grid technologies. Intensive computer simulations are performed to test and estimate the power demand by adapting the proposed algorithm. The developed algorithm shows a significant improvement in the comprehensive index with a value of 0.649 which is very high compared with conventional charging strategies. The results depicted an efficient scheduling and power distribution without affecting the performance of the EV or the flexibility of EV owner's trip schedule.