Loading...
Publication
Aplicação de controlo preditivo ao conversor NPC para carregamento de baterias
| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 10.21 MB | Adobe PDF |
Authors
Advisor(s)
Abstract(s)
Atualmente verifica-se um aumento nos veículos elétricos em circulação nas
estradas, esperando-se que o mesmo se mantenha durante os próximos anos. Este
aumento, acoplado às potências de carregamento cada vez mais elevadas, terá
inevitavelmente de ser compensado pela rede elétrica de energia através da
instalação de capacidade de geração adicional. No entanto existem tecnologias de
suporte, como o V2G e o smart-charging, que permitem reduzir o impacto que estes
veículos têm sobre a rede elétrica de energia.
Neste trabalho implementou-se o carregamento de baterias com recurso a um
conversor NPC de 3 níveis, utilizando um controlador preditivo. Verificou-se que o
conversor consegue realizar a transferência de potência em ambos os sentidos,
permitindo assim implementar o carregamento e tecnologia V2G.
Em termos de controlo implementou-se um controlador preditivo para o controlo
das correntes AC e também um controlador PI (Proporcional Integral), numa malha
de controlo externa, para o controlo da tensão no link DC do conversor.
Inicialmente o conversor foi testado em ambiente de simulação, através da
plataforma Simulink do MATLAB, permitindo verificar o correto funcionamento dos
controladores implementados. Posteriormente o controlo do conversor foi
implementado num protótipo laboratorial, recorrendo a um microcontrolador de baixo
custo, e os resultados foram comparados com os de simulação.
Os resultados obtidos permitem verificar que o controlador de corrente
consegue seguir corretamente as referências pretendidas, com baixo nível de THD,
e também responder rapidamente a variações das referências. O controlador da
tensão do link DC mostrou-se também capaz de seguir as referências pretendidas,
trabalhando em conjunto com o controlador preditivo numa malha de controlo. O
carregamento das baterias realizou-se corretamente segundo o protocolo de
carregamento implementado e verificou-se também que o estimador de estado de
carga está a funcionar corretamente.
Nowadays we can observe an increase in electric vehicles on the roads, expecting this trend to remain in the future. This increase, couple with the ever-growing charging power, will inevitably have to be compensated by the electric power grid with the installation of additional generation capacity. However, there are supporting technologies, such as V2G and smart-charging, which allow the impact of these vehicles on the electric power grid to be reduced. On this assignment battery charging was implemented by using a 3 level NPC converter, with predictive control. The bidirectionality of the power transfer on the converter was verified, which in term allows the implementation of the charging and V2G technologies. In terms of control, a predictive controller was implemented to control the AC currents, and a PI (Proportional Integral) controller was introduced to control the DC link voltage, in an external control loop. The converter was initially tested in a simulation environment, using Simulink from MATLAB, verifying the correct operation of the implemented controllers. Afterwards the converter controllers were implemented in a lab prototype, using a low-cost microcontroller, and the results were compared with the ones obtained in the simulations. The results obtained allowed us to verify that the current controller was able to correctly follow the references as intended, with low THD, and that it was able to rapidly respond to changes in the references. The DC link controller was also able to follow the references, working together with the predictive controller. The charging was managed correctly with the charging protocol, and it was also verified that the charge state estimator worked accordingly.
Nowadays we can observe an increase in electric vehicles on the roads, expecting this trend to remain in the future. This increase, couple with the ever-growing charging power, will inevitably have to be compensated by the electric power grid with the installation of additional generation capacity. However, there are supporting technologies, such as V2G and smart-charging, which allow the impact of these vehicles on the electric power grid to be reduced. On this assignment battery charging was implemented by using a 3 level NPC converter, with predictive control. The bidirectionality of the power transfer on the converter was verified, which in term allows the implementation of the charging and V2G technologies. In terms of control, a predictive controller was implemented to control the AC currents, and a PI (Proportional Integral) controller was introduced to control the DC link voltage, in an external control loop. The converter was initially tested in a simulation environment, using Simulink from MATLAB, verifying the correct operation of the implemented controllers. Afterwards the converter controllers were implemented in a lab prototype, using a low-cost microcontroller, and the results were compared with the ones obtained in the simulations. The results obtained allowed us to verify that the current controller was able to correctly follow the references as intended, with low THD, and that it was able to rapidly respond to changes in the references. The DC link controller was also able to follow the references, working together with the predictive controller. The charging was managed correctly with the charging protocol, and it was also verified that the charge state estimator worked accordingly.
Description
Keywords
Carregamento de baterias Conversor multinível NPC Controlador preditivo V2G Estimador de estado de carga Battery charging Multilevel converter Predictive controller State of charge estimator Engenharia Eletrotécnica - Telecomunicações . Faculdade de Ciências Exatas e da Engenharia