Performance Investigation of Electric Vehicle Charging in a Bipolar DC Microgrid

Abstract

Transportation electrification and charging infrastructure in India has to gain momentum in accordance to development of electric vehicle technology. Charging of electric vehicles is going to be a major electrical load in the near future, as more and more population shift to electric automotives from internal combusted engine-powered vehicles. Integration of electric vehicle charging stations might even burden the existing grid to a point of collapse or grid failure. Thus, an alternate grid structure is a necessity in the near-futures, for the powering the EV loads. Emergence of distributed energy sources and compatible technologies makes dc microgrids more popular compared to ac grids. Besides electric vehicle loads is primarily a battery which is dc in nature. Setting up charging stations integrated to microgrids can avoid the overburdening of the primary utility grid, by benefiting the distributed energy resources. The bipolar DC microgrid is a far better microgrid structure than the unipolar microgrid structure in many aspects like reliability, flexibility, and controllability. It can provide multiple voltage level interfaces according to the load demands, which is very apt for different charging levels of electric vehicles. Thus bipolar dc microgrid is the best answer we can put forward for the new age power grid. Establishing charging infrastructure interfaced with bipolar DC microgrids along the roads and highways is the most realistic and feasible solution to avoid the overburdening of the existing power system. Operation of multiple sources and loads connected to bipolar DC microgrid will affect voltage regulation, and overall stable operation of the grid. Along with this, intermittent nature of renewable energy resources and unpredictable charging and discharging behaviour of EVs aggravate the power imbalance between two poles of dc microgrid. This leads to voltage unbalance issues in bipolar dc microgrid. This thesis proposes bipolar converter configurations and fast acting nonlinear control strategies which can address voltage balancing and power sharing issues efficiently as well as integrate electric vehicle charging stations to bipolar dc grid. A bipolar dc grid in this research work consists of photovoltaic power generation, DC loads and a battery system as auxiliary energy storage. Three level boost converter derived from neutral point clamped converter configuration is used for connecting PV systems to bipolar dc grid. Bipolar converter based on three level buck boost converter is used to connect battery energy storage system to microgrid. This bidirectional converter is operated such that auxiliary battery storage bridges the power gap between load demand power and generated power. Discrete state space modelling of the three level converters is done and model predictive control is developed, which is fast acting and more robust than PI control. With model predictive control, voltage regulation, pole capacitance voltage balancing, MPPT tracking and overall stable operation of the bipolar dc microgrid is ensured. Simulation of the proposed model predictive control with three level converters is done in Typhoon Schematic Editor and Simulink. Hardware in loop testing of the model predictive control of the bipolar dc microgrid is done with Typhoon HIL 402 and validated in virtual and real time hardware in loop environment. EV load profile is developed for a cluster of vehicles considering different driving profiles and different vehicle dynamics. V2G and G2V regulation and control is implemented in integrating this EV load profile to bipolar dc grid. Multi-port three level converter is proposed for connecting electric vehicle charging stations. An overall decentralized model predictive control of multi-node bipolar dc grid with EV load profile, PV generation, BESS and dc loads is implemented effectively and analysed under dynamic grid conditions.