Home // SIMUL 2021, The Thirteenth International Conference on Advances in System Simulation // View article

Simulating Plug-in Electric Vehicle Charging for AutoMLBased Prediction of Regional Energy Demand

Authors:
Matthias Schneider
Sören Frey

Keywords: Plug-in Electric Vehicle Charging; Simulation; Energy Demand Prediction; Machine Learning; AutoML

Abstract:
We present a system for simulating home and public charging operations of Plug-in Electric Vehicles (PEVs). We model PEV traffic streams that result in corresponding charging operations. The simulation allows to configure many influential factors, such as the number of PEVs, consumption, charging stations, their locations, charging power, working hour distributions, holiday seasons, and the ratio of regular to irregular rides. In this paper, we demonstrate the applicability of our simulation in the context of predicting the short-term, regional energy demand of PEV charging. The prediction can be used to support energy suppliers and charging infrastructure operation, for instance. We use automated machine learning (AutoML) to train a forecasting model based on the simulation output. This combined workflow, integrating discrete-event simulation and machine learning, allows us to build a prediction pipeline where simulation data can be swapped with real data once available.

Pages: 44 to 48

Copyright: Copyright (c) IARIA, 2021

Publication date: October 3, 2021

Published in: conference

ISSN: 2308-4537

ISBN: 978-1-61208-898-3

Location: Barcelona, Spain

Dates: from October 3, 2021 to October 7, 2021