Energy consumption prediction and routing for electric commercial vehicles
Doktorsavhandling, 2021

With the recent growing interest for electric vehicles as one of the initiatives to help tackle pollution and climate change, several opportunities and challenges emerge. This kind of vehicle releases no tailpipe emissions, is quieter, more energy efficient in terms of tank-to-wheels and simpler, which can lead to less maintenance. On the other hand, their battery is still the main limitation in terms of energy capacity, time to recharge, weight and cost. One of the main consequences is a limitation in driving range, which especially affects commercial vehicles. In order to adopt electric trucks for urban distribution of goods, there is a need to improve and adapt current planning tools to take into account their constraints. To plan the routes and charging for these vehicles it is necessary to estimate their energy consumption accurately.

This thesis focuses on the development of energy consumption prediction and routing methods for electric commercial vehicles. The first part presents an overall background and short state of the art review. The main contributions are presented in the second part. The included articles are a step-by-step development of the methods, each covering different aspects of the problem. The first paper presents a deterministic energy prediction model integrated into routing models. The second paper proposes a probabilistic energy estimation method based on Bayesian machine learning and adds chance-constraints into the routing problem in order to plan charging within a confidence interval. The third paper covers routing with dynamic customers and stochastic energy consumption, proposing a solution method based on Safe Reinforcement Learning to minimize the risk of battery depletion by planning charging in an anticipative way. All papers are validated with realistic simulations as well as logged data. The results indicate that it is possible to save energy and reduce the risk of running out of energy while en route.

Green Logistics

Vehicle Routing

Bayesian Inference

Eco-Routing

Reinforcement Learning

Electric Vehicles

Machine Learning

Energy Consumption

Landahlsrummet (room 7430), Hörsalsvägen 11, Göteborg
Opponent: Professor Jesus Gonzalez Feliu, Department of Project Purchasing and Supply Chain, La Rochelle Business School, France

Författare

Rafael Basso

Chalmers, Elektroteknik, System- och reglerteknik, Reglerteknik

Electric Vehicle Routing Problem with Machine Learning for Energy Prediction

Transportation Research Part B: Methodological,; Vol. 145(2021)p. 24-55

Artikel i vetenskaplig tidskrift

Energy consumption estimation integrated into the Electric Vehicle Routing Problem

Transportation Research Part D: Transport and Environment,; Vol. 69(2019)p. 141-167

Artikel i vetenskaplig tidskrift

Dynamic Stochastic Electric Vehicle Routing Problem with Safe Reinforcement Learning

With the recent growing interest in electric vehicles as one of the initiatives to help tackle pollution and climate change, several opportunities and challenges emerge. This kind of vehicle releases no tailpipe emissions, is quieter, more energy efficient in terms of tank-to-wheels and simpler. On the other hand, their battery is still the main limitation in terms of energy capacity, time to recharge, weight and cost. One of the main consequences is a limitation in driving range, which especially affects commercial vehicles. Furthermore, their energy consumption depends on several uncertain factors, such as driving behavior and traffic conditions. In order to adopt electric trucks in urban logistics, there is a need to improve and adapt current planning tools to take into account their constraints, especially in more dynamic transport operations.

This thesis introduces energy consumption prediction and routing methods for electric commercial vehicles based on probabilistic machine learning. The included articles present a systematic development, each covering different aspects of the problem. The objective is to minimize energy consumption and minimize the risk of battery depletion while the vehicles are driving, by planning routes and charging in real-time and considering uncertainty. Consequently, these methods can help increase confidence in these vehicles, enabling their integration in current transport operations.

EL FORT - Optimering av elfordonsflotta i Real-Tid - (Fas 2)

VINNOVA, 2018-03-01 -- 2019-12-31.

EL FORT - El Flottor Optimering i Real-Tid

VINNOVA, 2014-07-01 -- 2017-06-30.

Styrkeområden

Transport

Ämneskategorier

Transportteknik och logistik

Reglerteknik

ISBN

978-91-7905-435-9

Doktorsavhandlingar vid Chalmers tekniska högskola. Ny serie: 4902

Utgivare

Chalmers tekniska högskola

Landahlsrummet (room 7430), Hörsalsvägen 11, Göteborg

Online

Opponent: Professor Jesus Gonzalez Feliu, Department of Project Purchasing and Supply Chain, La Rochelle Business School, France

Mer information

Senast uppdaterat

2021-01-21