| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 22.71 MB | Adobe PDF |
Authors
Abstract(s)
Os veículos híbridos Plug-in estão a começar a ganhar cada mais relevância no mercado
automóvel, apesar de a complexidade inerente ao seu sistema de gestão de propulsão e a
influência dos hábitos de carregamento dos utilizadores dificultarem a antecipação dos
impactos energéticos e das suas emissões da sua utilização em ambiente real.
O presente trabalho desenvolve um método genérico capaz de prever a variação do estado
de carga da bateria como também estimar os diferentes modos de propulsão solicitados pelo
veículo híbrido Plug-in em cada instante, tendo apenas por base dados dinâmicos medidos
em ciclos de condução reais. Através da análise dos dados recolhidos, torna-se possível criar
regras capazes de indicar o modo de propulsão, aferindo em cada instante a fonte energética
em uso, individualmente ou de forma combinada, durante a fase de descarregamento da
bateria. Posteriormente estima-se com base no método desenvolvido (Método ABCD) o
consumo combustível e eletricidade, as emissões de gases poluentes e a proporção de
distância percorrida em modo elétrico relativo à distância total (Utility Factor), sendo
comparado com os resultados reais e a estimativa obtida através do Método VSP Padrão.
Ainda assim, por forma a verificar a capacidade de adaptação do método desenvolvido
(Método ABCD) aplicou-se o mesmo a cinquenta ciclos de condução de diferentes tipologias
(Urbano, Misto e Extraurbano) e compararam-se os resultados com os resultados obtidos
através do Método VSP Padrão.
Os resultados obtidos permitiram obter, de uma forma geral, excelentes correlações entre a
estimativa e a medição do estado de carga da bateria (𝑅
2>0.99) como também se obteve com
base no método desenvolvido uma estimativa dos modos de propulsão com um acerto médio
de 82,2% do período de Charge Depleting do ciclo de condução em estudo.
Conclui-se então que o método desenvolvido, independentemente do ciclo de teste,
demonstrou possuir um excelente desempenho, conseguindo equiparar-se ao desempenho
do Método VSP Padrão, sendo por vezes mais exato que este, acrescentando ainda,
informação sobre a gestão das fontes de propulsão, somente a partir de informação dinâmica.
Plug-in hybrid vehicles are starting to gain more and more relevance in the automotive market, despite the complexity inherent in their propulsion management system and the influence of users' charging habits, making it difficult to anticipate energy impacts and emissions when exposed to a real environment. The present work develops a generic method capable of predicting the variation of the state of charge as well as estimating the different modes of propulsion requested by the plug-in hybrid vehicle at each instant, based only on dynamic data measured in real driving cycles. Through the analysis of the collected data, it becomes possible to create rules capable of indicating the mode of propulsion, checking in each moment the energy source in use, individually or in combination, during the battery discharge phase. Subsequently, based on the developed method (ABCD method) it is estimated the fuel and electricity consumption, the emissions of polluting gases and the proportion of distance traveled in electric mode relative to the total distance (Utility Factor), being compared with the actual results and the estimate obtained using the Standard VSP Method. Even so, in order to verify the adaptability of the developed method (ABCD Method) the same was applied to fifty driving cycles of different types (Urban, Mixed and Extra-urban) and the results were compared with the results obtained through the Standard VSP method. The results obtained made it possible to get, in general, excellent correlations between the estimate of the state of charge and the state of actual charge (𝑅 2> 0.99) as it was also obtained based on the developed method an estimate of the propulsion modes with a correct answer average of 82.2% of the Charge Depleting period of the driving cycle under study. It is concluded that the developed method, being this one of a generic character, demonstrated to have an excellent performance, managing to match the performance of the Standard VSP method, being sometimes more accurate than this, adding even more information about the management of the sources of propulsion, only from dynamic information.
Plug-in hybrid vehicles are starting to gain more and more relevance in the automotive market, despite the complexity inherent in their propulsion management system and the influence of users' charging habits, making it difficult to anticipate energy impacts and emissions when exposed to a real environment. The present work develops a generic method capable of predicting the variation of the state of charge as well as estimating the different modes of propulsion requested by the plug-in hybrid vehicle at each instant, based only on dynamic data measured in real driving cycles. Through the analysis of the collected data, it becomes possible to create rules capable of indicating the mode of propulsion, checking in each moment the energy source in use, individually or in combination, during the battery discharge phase. Subsequently, based on the developed method (ABCD method) it is estimated the fuel and electricity consumption, the emissions of polluting gases and the proportion of distance traveled in electric mode relative to the total distance (Utility Factor), being compared with the actual results and the estimate obtained using the Standard VSP Method. Even so, in order to verify the adaptability of the developed method (ABCD Method) the same was applied to fifty driving cycles of different types (Urban, Mixed and Extra-urban) and the results were compared with the results obtained through the Standard VSP method. The results obtained made it possible to get, in general, excellent correlations between the estimate of the state of charge and the state of actual charge (𝑅 2> 0.99) as it was also obtained based on the developed method an estimate of the propulsion modes with a correct answer average of 82.2% of the Charge Depleting period of the driving cycle under study. It is concluded that the developed method, being this one of a generic character, demonstrated to have an excellent performance, managing to match the performance of the Standard VSP method, being sometimes more accurate than this, adding even more information about the management of the sources of propulsion, only from dynamic information.
Description
Trabalho final de mestrado para obtenção do grau de Mestre em Engenharia Mecânica
Keywords
PHEV Estado de carga da bateria State of charge Tipo de propulsão Type of propulsion Gestão energética Energy management Emissão de gases Gas emission Fator de utilidade Utility factor
Citation
CAMPINO, Miguel Alexandre Calha - Avaliação energética de um sistema de gestão de propulsão de veículos híbridos plug-In. Lisboa: Instituto Superior de Engenharia de Lisboa, 2021. Dissertação de mestrado.
Publisher
Instituto Superior de Engenharia de Lisboa