Open AccessOptimal location of PEVCSs using MAS and ER approach
Author(s) -
Jiang Changxu,
Jing Zhaoxia,
Ji Tianyao,
Wu Qinghua
Publication year - 2018
Publication title -
iet generation, transmission and distribution
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.92
H-Index - 110
eISSN - 1751-8695
pISSN - 1751-8687
DOI - 10.1049/iet-gtd.2017.1907
Subject(s) - taxis , computer science , variety (cybernetics) , convergence (economics) , grid , quality (philosophy) , electric vehicle , power (physics) , simulation , operations research , transport engineering , engineering , artificial intelligence , mathematics , philosophy , physics , geometry , epistemology , quantum mechanics , economics , economic growth
The location of public electric vehicle charging stations (PEVCSs) has a great influence on the operational efficiency of charging stations, charging behaviours of EVs and the power quality of grids. To optimise the PEVCS locations for plug‐in electric taxis (PETs), this study proposes to utilise the multi‐agent system (MAS) and evidential reasoning (ER) approach. First, an MAS simulation framework for PET operation is proposed to dynamically simulate the PETs’ daily operation and estimate the charging demands of PETs, where a variety of agents are built to simulate not only the operation related players but also the operational environments. To accelerate the convergence rate and provide better operational strategies for PETs, a multi‐step Q ( λ ) learning is developed to make decisions for PET agents whether to find passengers or to charge under various situations. Moreover, a multi‐objective model for optimising the location of PEVCSs is developed considering the benefits of PETs and the power grid. Finally, an ER approach is applied to determine the final optimal siting considering the uncertainties of the assessor's cognition. Simulation results have demonstrated that the proposed MAS simulation framework and ER approach can effectively optimise the PEVCS locations.