Keeping the car clean, on the electrification of private transport

The electrification of private road vehicles - and the provision of a low carbon generation mix that supplies the energy for their motion - is likely to be a key contributor to meeting net zero targets and limiting the disastrous effects of anthropogenic climate change.The work presented in this thesis surrounds two aspects of this transition. Firstly,as EVs are fundamentally different to internal combustion vehicles (ICVs), in that their energy storage capacity is far smaller and the rate at which it can be replenished is much slower, there has been consumer resistance to their adoption due to the perception that their charging carries inconvenience compared to ICV fuelling. Secondly, as the energy demand of private road vehicles is shifted from the petrochemical supply chain to the electricity grid, there are potential i) issues surrounding the resilience of the grid and ii) opportunities resulting from the flexibility and 'schedulability' of EV charging in enabling the further decarbonisation of the power sector.With regards to the first aspect, this thesis presents an investigation of the inconvenience of EV charging versus ICV fuelling and how this is likely to change depending on vehicle parameters - battery capacity and charger power - and the set of locations at which it can be charged. It was found that if the EV can be charged while parked at home, the majority (80-95%) of individuals could make the transition to EVs without suffering any increase in time penalty associated with charging compared to ICV fuelling with modest battery capacities at the lower end of the market. However, for drivers who cannot charge at home this convenience parity is harder to achieve.This can be made easier by increasing battery capacity and charger power and ensuring the provision of public destination and en route charging infrastructure. This thesis presents characterisation of the likely demand at such infrastructure based on the analysis of smartphone app users' locational data. It was found that this kind of EV charging demand is likely to vary significantly in time and by location, depending on what kind of public amenity it is installed at; such an approach could be invaluable to transport and power system planners given the projected rapid uptake of EVs. This thesis presents analysis into the likely impact of EV charging on residential distribution networks, taking into account the effect of i) the way in which drivers schedule charge events, ii) the social demographics (and hence travel habits) of the individuals served by a network and iii) the effect of the rapidly evolving EV sector and the resulting changes in technical parameters and level of charging infrastructure. By means of case studies on real distribution networks in Glasgow's South side, it was found that all three of the above factors have a significant effect on the impact seen from EV charging and that, if their charging is uncontrolled, residential distribution networks in better-off areas with high rates of car ownership are unlikely to be able to cope with charging load when over 40% of vehicles are replaced with EVs. Based on the above finding, this thesis presents an investigation of the potential of EVs to interact positively with the grid, by charging selectively to i) avoid network peaks and thus enable a higher penetration of EVs before the network must be reinforced and ii) enable the further decarbonisation of the power system by charging when grid carbon emissions are low and local renewable energy is in surplus. Techniques to schedule charge events are presented: a 'valley filling' optimisation approach based on the controller having perfect foresight of the arrival of vehicles to represent the best case; and a set of low-information heuristics that could operate using data from 'smart' EV chargers.Though some of the heuristics could perform similarly to the formal optimisation, both methods resulted in the network being operated outwith statutory GB voltage limits for 100% penetration of EVs. Finally, it was found that there is significant potential for EVs to complement the decarbonisation of the power system. If 'smart charged',their resulting carbon intensity (gCO2/km) can be reduced to around a fifth of that of a new petrol or diesel car if charged from the current GB grid. Furthermore, a penetration of 20% EVs in Scotland could be smart-charged to absorb up to 75% of curtailed generation at Scotland's largest onshore wind farm.The electrification of private road vehicles - and the provision of a low carbon generation mix that supplies the energy for their motion - is likely to be a key contributor to meeting net zero targets and limiting the disastrous effects of anthropogenic climate change.The work presented in this thesis surrounds two aspects of this transition. Firstly,as EVs are fundamentally different to internal combustion vehicles (ICVs), in that their energy storage capacity is far smaller and the rate at which it can be replenished is much slower, there has been consumer resistance to their adoption due to the perception that their charging carries inconvenience compared to ICV fuelling. Secondly, as the energy demand of private road vehicles is shifted from the petrochemical supply chain to the electricity grid, there are potential i) issues surrounding the resilience of the grid and ii) opportunities resulting from the flexibility and 'schedulability' of EV charging in enabling the further decarbonisation of the power sector.With regards to the first aspect, this thesis presents an investigation of the inconvenience of EV charging versus ICV fuelling and how this is likely to change depending on vehicle parameters - battery capacity and charger power - and the set of locations at which it can be charged. It was found that if the EV can be charged while parked at home, the majority (80-95%) of individuals could make the transition to EVs without suffering any increase in time penalty associated with charging compared to ICV fuelling with modest battery capacities at the lower end of the market. However, for drivers who cannot charge at home this convenience parity is harder to achieve.This can be made easier by increasing battery capacity and charger power and ensuring the provision of public destination and en route charging infrastructure. This thesis presents characterisation of the likely demand at such infrastructure based on the analysis of smartphone app users' locational data. It was found that this kind of EV charging demand is likely to vary significantly in time and by location, depending on what kind of public amenity it is installed at; such an approach could be invaluable to transport and power system planners given the projected rapid uptake of EVs. This thesis presents analysis into the likely impact of EV charging on residential distribution networks, taking into account the effect of i) the way in which drivers schedule charge events, ii) the social demographics (and hence travel habits) of the individuals served by a network and iii) the effect of the rapidly evolving EV sector and the resulting changes in technical parameters and level of charging infrastructure. By means of case studies on real distribution networks in Glasgow's South side, it was found that all three of the above factors have a significant effect on the impact seen from EV charging and that, if their charging is uncontrolled, residential distribution networks in better-off areas with high rates of car ownership are unlikely to be able to cope with charging load when over 40% of vehicles are replaced with EVs. Based on the above finding, this thesis presents an investigation of the potential of EVs to interact positively with the grid, by charging selectively to i) avoid network peaks and thus enable a higher penetration of EVs before the network must be reinforced and ii) enable the further decarbonisation of the power system by charging when grid carbon emissions are low and local renewable energy is in surplus. Techniques to schedule charge events are presented: a 'valley filling' optimisation approach based on the controller having perfect foresight of the arrival of vehicles to represent the best case; and a set of low-information heuristics that could operate using data from 'smart' EV chargers.Though some of the heuristics could perform similarly to the formal optimisation, both methods resulted in the network being operated outwith statutory GB voltage limits for 100% penetration of EVs. Finally, it was found that there is significant potential for EVs to complement the decarbonisation of the power system. If 'smart charged',their resulting carbon intensity (gCO2/km) can be reduced to around a fifth of that of a new petrol or diesel car if charged from the current GB grid. Furthermore, a penetration of 20% EVs in Scotland could be smart-charged to absorb up to 75% of curtailed generation at Scotland's largest onshore wind farm