With the rising awareness about the impact of vehicular emission on human health as well as the environment, more and more people are switching from conventional vehicles to modern electric vehicles and plug-in hybrid vehicles. The increasing number of EVs makes the use of distributed charging coordinating strategies preferable to the centralized ones due to scalability and simplified communication. In this thesis we will analyze Additive-Increase Multiplicative-Decrease (AIMD) algorithms which have been successfully used in communication networks for congestion control. We will develop number of algorithms based on AIMD in order to maximize quality of service to the EV owner whilst maximizing the utilization from the grid. At the end we will present the simulation results to show the performance of these algorithms compared to a centralized optimal solution. In all the scenarios examined, the performance of tuned AIMD algorithm is very close to centralized optimal solution but AIMD has an advantage over centralized solution because it requires very limited communication and due to its distributed nature, scalability is considerable high as compared to the centralized optimal solution.
La crescente attenzione sull’impatto delle emissione dei veicoli, sulla salute dell’uomo e sull’ambiente, spinge sempre più persone dall’uso di veicoli tradizionali all’uso di veicoli elettrici o ibridi. Il crescente numero di veicoli elettrici (VE) rende preferibile l’uso di strategie di coordinazione distribuita piuttosto che centralizzata, grazie anche alla loro scalabilità e alla semplicità del sistema di comunicazione. In questa tesi, considereremo i così detti algoritmi Additive-Increase Multiplicative-Decrease (AIMD), utilizzati con successo nelle reti di comunicazione per il controllo del traffico di informazioni. Svilupperemo diversi algoritmi basati su AIMD per massimizzare la qualità del servizio dei proprietari dei veicoli, massimizzando l’utilizzo della rete. Infine presenteremo risultati in simulazione per mostrare le prestazioni di questi algoritmi paragonati alla corrispondente soluzione ottima centralizzata. In tutti gli scenari esaminati le prestazioni degli algoritmi AIMD sono molto simili a quelle nel caso centralizzato, ma con il vantaggio che gli algoritmi AIMD richiedono comunicazioni limitate e, trattandosi di algoritmi distribuiti, consentono un’alta scalabilità rispetto alla soluzione ottima centralizzata.
AIMD distributed resource allocation with an application to electric vehicles
SHAH, SAQIB NISAR
2017/2018
Abstract
With the rising awareness about the impact of vehicular emission on human health as well as the environment, more and more people are switching from conventional vehicles to modern electric vehicles and plug-in hybrid vehicles. The increasing number of EVs makes the use of distributed charging coordinating strategies preferable to the centralized ones due to scalability and simplified communication. In this thesis we will analyze Additive-Increase Multiplicative-Decrease (AIMD) algorithms which have been successfully used in communication networks for congestion control. We will develop number of algorithms based on AIMD in order to maximize quality of service to the EV owner whilst maximizing the utilization from the grid. At the end we will present the simulation results to show the performance of these algorithms compared to a centralized optimal solution. In all the scenarios examined, the performance of tuned AIMD algorithm is very close to centralized optimal solution but AIMD has an advantage over centralized solution because it requires very limited communication and due to its distributed nature, scalability is considerable high as compared to the centralized optimal solution.File | Dimensione | Formato | |
---|---|---|---|
2018_04_Shah.pdf
non accessibile
Descrizione: Thesis text
Dimensione
3.52 MB
Formato
Adobe PDF
|
3.52 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/140087