It has been recognized that an increased penetration of electric vehicles (EVs) may potentially alter load profile in a distribution network. As EVs are regarded as a diversely distributed load so a deterministic method, to predict EV charging load, may not account for all possible factors that could affect the power system. Thus, a stochastic approach is applied that takes into account various realistic factors such as EV battery capacity, state of charge (SOC), driving habit/need, i.e., involving type and purpose of trip, plug-in time, mileage, recharging frequency per day, charging power rate and dynamic EV charging price under controlled and uncontrolled charging schemes. A probabilistic model of EVs charging pattern associated with residential load profile is developed. The probabilistic model gives an activity based residential load profile and EV charging pattern over a period of 24 h. Then, the model output is used to assess the power quality index such as voltage unbalance factor under different electric vehicle penetration levels at different nodes of the system. An uneven EV charging scenario is identified that could cause the voltage unbalance to exceed its permissible limit.

Introduction: Convolutional neural networks (CNNs) have maintained their dominance in deep learning methods for human action recognition (HAR) and other computer vision tasks. However, the need for a large amount of training data always restricts the performance of CNNs. Method: This paper is inspired by the two-stream network, where a CNN is deployed to train the network by using the spatial and temporal aspects of an activity, thus exploiting the strengths of both networks to achieve better accuracy. Contributions: Our contribution is twofold: first, we deploy an enhanced spatial stream, and it is demonstrated that models pre-trained on a larger dataset, when used in the spatial stream, yield good performance instead of training the entire model from scratch. Second, a dataset augmentation technique is presented to minimize overfitting of CNNs, where we increase the dataset size by performing various transformations on the images such as rotation and flipping, etc. Results: UCF101 is a standard benchmark dataset for action videos, and our architecture has been trained and validated on it. Compared with the other two-stream networks, our results outperformed them in terms of accuracy.