A PHET Dispatching Method Considering Customer Demand and Charging Resources

Plug-in electrical vehicles (PEVs) play a significant role in environment protection and attract global attentions. However, with the popularization of PEVs, low-efficiency supporting facilities such as the charging system impede its future development. To improve the charging system, we focus on plug-in hybrid electric taxis (PHETs) as they are the main users of public charging system. In this paper, we first predict the order numbers and mileage consumption of orders with the help of convolutional neural networks (CNNs). We then divide the area into 30 groups using K-means method and plan the charging capacity of station in each area. Two coordinated dispatching and charging strategies are proposed considering the states of charge (SOCs) at vehicle level and considering the real-time effect at region level, respectively. Finally, we test the dispatching effect using order car ratio (OCR) models at region level. The results show that it works quite well when testing on the real dataset. This method provides optimal instructions for PHETs to pick orders, satisfy their charging demand and also meet the order demands for taxis.

[1]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[2]  Toshiyuki Yamamoto,et al.  Fast-charging station choice behavior among battery electric vehicle users , 2016 .

[3]  Walid Saad,et al.  Economics of Electric Vehicle Charging: A Game Theoretic Approach , 2012, IEEE Transactions on Smart Grid.

[4]  Fang He,et al.  Improving the electrification rate of the vehicle miles traveled in Beijing: A data-driven approach , 2017 .

[5]  Michael I. Jordan,et al.  Machine learning: Trends, perspectives, and prospects , 2015, Science.

[6]  Feng Shu,et al.  Short-term traffic flow prediction based on spatio-temporal analysis and CNN deep learning , 2019, Transportmetrica A: Transport Science.

[7]  Kankar Bhattacharya,et al.  Electric Vehicle Charging Facility as a Smart Energy Microhub , 2017, IEEE Transactions on Sustainable Energy.

[8]  Zhenhong Lin,et al.  Charging infrastructure planning for promoting battery electric vehicles: An activity-based approach using multiday travel data , 2014 .

[9]  D.M. Mount,et al.  An Efficient k-Means Clustering Algorithm: Analysis and Implementation , 2002, IEEE Trans. Pattern Anal. Mach. Intell..

[10]  Yong Wang,et al.  Learning Traffic as Images: A Deep Convolutional Neural Network for Large-Scale Transportation Network Speed Prediction , 2017, Sensors.

[11]  Cyrus Shahabi,et al.  Diffusion Convolutional Recurrent Neural Network: Data-Driven Traffic Forecasting , 2017, ICLR.

[12]  Ryan Liu,et al.  A survey of PEV impacts on electric utilities , 2011, ISGT 2011.

[13]  Bin Ran,et al.  A hybrid deep learning based traffic flow prediction method and its understanding , 2018 .

[14]  Ram Rajagopal,et al.  Design and Planning of a Multiple-Charger Multiple-Port Charging System for PEV Charging Station , 2019, IEEE Transactions on Smart Grid.

[15]  Huimiao Chen,et al.  Spatiotemporal Traffic Flow Forecasting for PHETs Based on Data Mining and Deep Learning , 2019, 2019 IEEE International Symposium on Signal Processing and Information Technology (ISSPIT).

[16]  Zhipeng Liu,et al.  Optimal Planning of Electric-Vehicle Charging Stations in Distribution Systems , 2013, IEEE Transactions on Power Delivery.