Stochastic modelling and energy management method to support electrical network using parking lots incorporating EV charging infrastructure
The increase in market participation of low carbon zero-emission electric vehicles (EVs) and renewable energy sources provides an opportunity to address predominant issues, such as air pollution, depletion of energy resources, climate change, and global warming. The rapid adoption of EVs will turn parking lots (especially shopping centre parking lot (SCPL)) into charging stations. The unmanaged cluster charging of EVs at PLs may create overloading, voltage and frequency stability issues. If properly sized and managed, PLs equipped with bidirectional EV chargers, a stationary battery energy storage system (BESS) and/or a roof top photovoltaic (PV) system can provide grid support such as peak shaving, voltage and frequency regulation which eventually results in mitigating stability issue.
EVs are moveable loads, and it is challenging to estimate their intermittent usage and random charging behaviour. Therefore, this thesis's first contribution is to develop a stochastic model that can emulate EVs' intermittent charging behaviour at SCPL. The proposed method considered the arrival and departure time, irregular energy consumption per kilometre, EVs’ battery degradation, daily distance travelled and rate of charging and discharging of EVs. Uncertainty is added by mapping the above-mentioned EV parameters into their best-fitted probability distribution functions with quantified uncertainty. The proposed stochastic model precisely estimates a day-ahead EV charging demand in SCPLs, which can help evaluate the impact of EV charging on the grid's reliability.
This thesis's second contribution is to develop an on-site grid-connected BESS sizing method for an SCPL equipped with EV charging infrastructure in a constrained grid. The proposed method depends on the intermittent EV charging demand, irregular PV output and grid power constraints. A region reduction method (RRM) is formulated to reduce the search space for the optimisation problem. Based on the computed EV charging demand and PV output, an optimisation problem combined with RRM is proposed to calculate the BESS's accurate size such that the capital cost and operational cost of the parking lot operator are minimised.
This dissertation's third contribution is to develop a two-stage optimisation-based voltage regulation method by managing the power transfer between multiple SCPLs and the electrical grid. The first stage maximises the parking lot operators’ power-selling capacity by managing the energy resources available in the SCPL such that the intermittent EV charging demand in the SCPL was fulfilled. The second stage regulates the system voltage by prioritising multiple SCPLs such that the power transfer losses and the operational cost of the electrical network were minimised.
The methods proposed in this thesis were analysed by performing extensive case studies characterised by real household travel survey data, SCPL occupancy data, and real weather data of the Sydney region. The efficacy of the proposed methods was demonstrated by comparing the results with the literature. The analysis of the results was discussed in detail. The results show that the SCPLs were found to be potentially more significant candidates for providing ancillary services to the grid than workplace or home parking lots, with their scale and throughput compensating for their relatively low occupancy of EVs.
