Government policies, aimed at reducing energy consumption and lowering the production of greenhouse gas emissions, have pushed towards an electrification of the road transport sector. In the near future, pure electric vehicles and hybrid vehicles will replace the current fleet of endothermic engine vehicles. Electric vehicles (EVs) have a strong sensitivity to their operating temperature. The performance of batteries deteriorates quickly if their operating temperature is not properly controlled. Furthermore, given the reduced availability of heat produced by the electrical components, the thermal control of the passenger compartment is also critical, which stresses the battery leading to a further decrease in vehicle driving range. An optimized thermal management system (TMS) is essential to improve driving range. Unlike conventional vehicles, EVs require different and more efficient heating and cooling methods, making the TMS very complex. A simulation model is thus essential to reduce time and helps the developing of these complex systems. The thesis work aims to propose a TMS model using Simscape as simulation software. Simscape is a simulation software within the MATLAB / Simulink environment. Starting from a simplified model present in the Simscape library, it has been improved step by step with more detailed subsystems. Then it was used to simulate an electric vehicle in high and low temperature environmental conditions to evaluate its operation and to analyse the interconnections between the various subsystems and the effectiveness of the TMS. The model was then adapted to simulate the behaviour of a hybrid electric vehicle as well. Finally, the model for the EV underwent a multi-objective constrained optimization process through the use of neural networks. A certain improvement has been achieved with reduction of extracted charged of 1.92% and 0.83% in summer and winter conditions respectively.
Politiche governative mirate a una riduzione del consumo energetico e a una minor produzione di emissioni di gas serra ha spinto verso un'elettrificazione dei veicoli stradali. Nel prossimo futuro veicoli puramente elettrici e ibridi andranno a sostituire l'attuale flotta di veicoli con motore a combustione interna. I veicoli elettrici (VE) presentano una forte sensibilità alla loro temperatura di esercizio. Le performance delle batterie si deteriorano in fretta se la loro temperatura di esercizio non è opportunamente controllata. Inoltre, vista la ridotta disponibilità di calore prodotto dai componenti elettrici, risulta critico anche il controllo termico dell'abitacolo che va a stressare la batteria portando a un'ulteriore diminuzione dell'autonomia. Ciò può portare a una riduzione di efficienza e autonomia fino al 40% in meno rispetto a condizioni ottimali. Risulta fondamentale un sistema ottimizzato di gestione termica (SGT) per migliorare l'autonomia di guida. A differenza dei veicoli convenzionali, i VE necessitano di diversi e più efficienti metodi di riscaldamento e climatizzazione, rendendo il TMS molto complesso. Risulta pertanto necessario l'utilizzo di modelli per poter simulare questi sistemi in modo da ridurne i tempi di sviluppo. Il lavoro di tesi mira a proporre un modello di SGT utilizzando Simscape come software di simulazione. Simscape è un software di simulazione presente nell'ambiente di MATLAB/Simulink. Partendo da un modello semplificato presente all'interno della libreria Simscape, si è proseguito aggiungendo e/o sostituendone alcuni elementi. In seguito, si è passati alle simulazioni in condizioni ambientali di alta e bassa temperatura per valutarne il funzionamento e per analizzare l'interconnessione tra i vari sottosistemi nonché l'efficacia del SGT stesso. Successivamente il modello è stato adattato per poter simulare anche il sistema termico di un veicolo elettrico ibrido. Infine, il modello per il VE è stato sottoposto a un processo di ottimizzazione vincolato multi-obiettivo tramite l'utilizzo di reti neurali. L'autonomia complessiva risulta migliorata presentando una riduzione di carica estratta del 1,92% e dello 0,83% rispettivamente in condizioni estive e invernali.
A simscape model with neural network-based optimization for powertrains of new energy vehicles
CARNAGHI, ANDREA CARLO
2020/2021
Abstract
Government policies, aimed at reducing energy consumption and lowering the production of greenhouse gas emissions, have pushed towards an electrification of the road transport sector. In the near future, pure electric vehicles and hybrid vehicles will replace the current fleet of endothermic engine vehicles. Electric vehicles (EVs) have a strong sensitivity to their operating temperature. The performance of batteries deteriorates quickly if their operating temperature is not properly controlled. Furthermore, given the reduced availability of heat produced by the electrical components, the thermal control of the passenger compartment is also critical, which stresses the battery leading to a further decrease in vehicle driving range. An optimized thermal management system (TMS) is essential to improve driving range. Unlike conventional vehicles, EVs require different and more efficient heating and cooling methods, making the TMS very complex. A simulation model is thus essential to reduce time and helps the developing of these complex systems. The thesis work aims to propose a TMS model using Simscape as simulation software. Simscape is a simulation software within the MATLAB / Simulink environment. Starting from a simplified model present in the Simscape library, it has been improved step by step with more detailed subsystems. Then it was used to simulate an electric vehicle in high and low temperature environmental conditions to evaluate its operation and to analyse the interconnections between the various subsystems and the effectiveness of the TMS. The model was then adapted to simulate the behaviour of a hybrid electric vehicle as well. Finally, the model for the EV underwent a multi-objective constrained optimization process through the use of neural networks. A certain improvement has been achieved with reduction of extracted charged of 1.92% and 0.83% in summer and winter conditions respectively.File | Dimensione | Formato | |
---|---|---|---|
2022_04_Carnaghi_01.pdf
accessibile in internet per tutti
Descrizione: Tesi
Dimensione
4.26 MB
Formato
Adobe PDF
|
4.26 MB | Adobe PDF | Visualizza/Apri |
2022_04_Carnaghi_02.pdf
accessibile in internet per tutti
Descrizione: Extended Summary
Dimensione
712.15 kB
Formato
Adobe PDF
|
712.15 kB | 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/10589/186146