Scooter-sharing has been introduced as a popular new transportation mode. However, e-scooters have a limited battery capacity and require frequent manual charging, which adds significnatly to operating costs. Charging stations that connect to the electricity grid require large expenses and effort in planning, permitting and execution, while lacking the flexibility to adapt their locations as demand changes in a city. On the contrary, charging stations that use solar panels could be deployed in a cost-effective and flexible manner as there is no need to connect to utility cables running underground in typical settings.
To show the viability of a solar charging solution for e-scooters, this study simulates the operations of a future service where trips are served by scooters with a limited autonomy, able to relocate themselves to the nearest charging station that is powered by solar panels installed on top of it. Solar charging potential is estimated for multiple scenarios of solar panel size based on a three-dimensional solar irradiation model that gives an accurate assessment of available power at any time of the day after taking into account shading and reflections on buildings in dense urban areas. Using a real-world dataset of historic scooter trips in Singapore, the goal is to serve a maximum number of trips while limiting the fleet size, waiting time for users and the size of installed solar panels.
How does the scooter-sharing services perform when a real-time shareability network is adapted to solar charging platforms? This study runs a simulation on the real trip data that is over 28 days in Singapore, varying key parameters to explore the viability of solar charging under a range of scenarios.
A major concern of the solar charging solution is the resilience of the service when confronting continuous cloudy days that limit the power generated by solar panels.
To investigate the resilience when confronting unfavourable weather, we investigated a scenario where solar charging power is set to zero, and conducted experiments while varying the starting battery level between 150Wh and 900Wh.
It shows that, with the increase of the initial battery capacity, the sharing service can support almost all trips for an increasing number of days without solar power.
As costs for recharging scooters constitute a large amount to operational expenses of scooter-sharing services, the proposed solar charging approach could help in decreasing costs and increasing availability of them.
As a case study in Singapore, solar panels of a size of 1㎡ could support almost all trips, while allowing to reduce the total fleet size, presenting additional savings in costs. The system also supports 90% of on-demand mobility for at least three consecutive days without solar charging, which suggests the resilience of the system and inspires us to promote the proposed solar charging in other global cities.
Rui Zhu Lead Researcher
Dániel Kondor, and Cheng Cheng Researchers
Paolo Santi, Man Sing Wong, and Carlo Ratti Scientific Advisors
Yang Liuhuaying Website & Visualisation
Rui Zhu, Dániel Kondor, Cheng Cheng, Xiaohu Zhang, Paolo Santi, Man Sing Wong, Carlo Ratti (2022).Solar photovoltaic generation for charging shared electric scooters.Applied Energy, 313, 118728.
Zhu, R., Cheng, C., Santi, P., Chen, M., Zhang, X., Mazzarello, M., Wong, M.S., Ratti, C. (2022). Optimization of photovoltaic provision in a three-dimensional city using real-time electricity demand. Applied Energy, 316, 119042.
Zhu, R., Wong, M.S., Kwan, M.P., Min, C., Santi, P., Ratti, C. (2022). An economically feasible optimization of photovoltaic provision using real electricity demand: A case study in New York City. Sustainable Cities and Society, 78, 103614.
Zhu, R., Zhang, X., Kondor, D., Santi, P., Ratti, C. (2020). Understanding spatio-temporal heterogeneity of bike-share and scooter-share mobility. Computers, Environment and Urban Systems, 81, 101483.
Zhu, R., You, L., Santi, P., Wong, M. S., Ratti, C. (2019). Solar accessibility in developing cities: A case study in Kowloon East, Hong Kong. Sustainable Cities and Society, 51, 101738.
The authors thank the funding support from (i) the National Research Foundation, Prime Minister’s Office, Singapore under its Campus for Research Excellence and Technological Enterprise (CREATE) programme, (ii) the Strategic Hiring Scheme (Grant No. P0036221) at the Hong Kong Polytechnic University, and (iii) the General Research Fund (Grant No. 15602619 and 15603920) and Collaborative Research Fund (Grant No. C4023-20GF), the Hong Kong Research Grant Council, Hong Kong, China.