Charging Network Planning for Electric Bus Cities: A Case Study of Shenzhen, China

Yuping Lin, Kai Zhang, Zuo‐Jun Max Shen, Lixin MiaoSustainability 2019 · DOI

TL;DR

This paper tackles the complex challenge of planning charging infrastructure for electric buses (e-buses) in a city like Shenzhen, which has fully transitioned to an e-bus fleet. They developed a novel optimization model that jointly considers both the transportation network and the power grid, proposing a mixed-integer second-order cone programming approach and a specialized "No R" algorithm to improve computational efficiency. This integrated approach is crucial for cost-effective and reliable e-bus operations, especially given the unique demands of e-buses compared to other EVs.

What Problem Does This Paper Address?

The rapid adoption of electric buses, exemplified by Shenzhen's pioneering all-e-bus city status, presents a significant challenge: how to efficiently plan and deploy the necessary charging infrastructure. Unlike private EVs, e-buses have rigid operational schedules and distinct charging behaviors, requiring charging to occur only after completing passenger routes. Furthermore, the sheer scale of e-bus charging demands can place an immense burden on the power grid, potentially exceeding its capacity. Existing planning methods often address transportation and power grid considerations separately, leading to suboptimal solutions. This paper aims to bridge this gap by developing a holistic model that simultaneously optimizes the location and capacity of charging stations, the assignment of buses to these stations, and their connection to the power grid, all while minimizing overall costs.

Methods

The core of this research is a cost-minimization model formulated as a mixed-integer second-order cone programming (MISOCP) problem. This model integrates three key assignment layers:

  1. Origin Assignment: Determining where e-buses start their charging trips (depots, final stops, or initial stops).
  2. Charging Station Assignment: Deciding which charging stations e-buses will use.
  3. Power Grid Connection: Assigning charging stations to specific nodes within the power grid.

To handle the computational complexity of this MISOCP, the authors propose two algorithmic improvements:

  • Relaxation of Binary Variables: Relaxing certain binary variables to continuous ones to potentially speed up computation.
  • "No R" Algorithm: A more significant optimization that preprocesses distance matrices to reduce the problem's dimensionality by pre-selecting the most likely charging trip origins, thereby simplifying the assignment problem and drastically improving computational speed.

Datasets

The study utilizes real-world data from Shenzhen, China, for its case study. This includes:

  • Transportation Network Data: Information on bus lines, initial stops, final stops, and candidate charging station locations.
  • Operational Data: Real operational data from a major bus company (Shenzhen Bus Group), including 333 bus lines and 5698 e-buses.
  • Power Grid Data: A 110-kV distribution network model from a previous study.

The paper also mentions the use of virtual data generation for certain parameters, such as the distance from depots to charging stations, to ensure consistency with real-world proximity.

Key Insights and Ideas

  • Joint Optimization is Crucial: Optimizing the transportation network and power grid independently leads to significantly higher costs and potentially inefficient infrastructure. A holistic approach is essential for true cost-effectiveness.
  • E-bus Characteristics Matter: The unique operational patterns of e-buses (fixed routes, charging only after trips) require specialized planning models, distinct from those for private EVs.
  • Driving Range is King for Cost Savings: Advancements in e-bus driving range have a more substantial impact on reducing overall charging infrastructure costs than improvements in charging speed. This suggests R&D priorities should lean towards battery technology.
  • "No R" Algorithm's Efficiency: The proposed "No R" algorithm dramatically reduces computational time, making it feasible to solve large-scale planning problems that would otherwise be intractable. This is achieved by simplifying the origin selection process.
  • Layout Robustness: The optimal layout of charging stations and their grid connections appears to be relatively robust to changes in charging speed, though more sensitive to driving range variations.

Key Results

  • Cost Reduction: The joint optimization approach achieved a total cost of 8.41 million yuan, saving 18% compared to transportation-first optimization and 33% compared to grid-first optimization.
  • Computational Speedup: The "No R" algorithm demonstrated significant improvements in computational speed, solving large-scale instances in seconds or minutes, whereas the original MISOCP approach could take over 20 hours for comparable problems.
  • Impact of Technology: Increasing driving range from 250 km to 500 km reduced the total cost by 38%, while increasing charging speed by 10 times (from 108 kW to 1080 kW) resulted in only a 13% cost reduction.
  • Shenzhen Case Study: The application to Shenzhen's real-world network successfully identified 11 charging stations and their grid connections, demonstrating the model's practical applicability.

Limitations and Future Work

  • E-bus Scheduling: The current study focuses on charging station location and assignment. Future research should incorporate e-bus scheduling, which involves optimizing both passenger transport and charging trips, considering fleet size and charger numbers.
  • Bus Characteristics: Further emphasis should be placed on understanding how bus electrification impacts other bus characteristics and operations.
  • Battery Lifetime: While the study notes that higher charging power can reduce battery lifetime, this impact is not fully quantified in the cost model. Future work could integrate more detailed battery degradation models.

Summary updated

Abstract

In 2017, Shenzhen replaced all its buses with battery e-buses (electric buses) and has become the first all-e-bus city in the world. Systematic planning of the supporting charging infrastructure for the electrified bus transportation system is required. Considering the number of city e-buses and the land scarcity, large-scale bus charging stations were preferred and adopted by the city. Compared with other EVs (electric vehicles), e-buses have operational tasks and different charging behavior. Since large-scale electricity-consuming stations will result in an intense burden on the power grid, it is necessary to consider both the transportation network and the power grid when planning the charging infrastructure. A cost-minimization model to jointly determine the deployment of bus charging stations and a grid connection scheme was put forward, which is essentially a three-fold assignment model. The problem was formulated as a mixed-integer second-order cone programming model, and a “No R” algorithm was proposed to improve the computational speed further. Computational studies, including a case study of Shenzhen, were implemented and the impacts of EV technology advancements on the cost and the infrastructure layout were also investigated.
BibTeX
@inproceedings{lin2019charging,
  author = {Yuping Lin and Kai Zhang and Zuo‐Jun Max Shen and Lixin Miao},
  title = {{Charging Network Planning for Electric Bus Cities: A Case Study of Shenzhen, China}},
  booktitle = {Sustainability},
  year = {2019},
  doi = {10.3390/su11174713},
}
Figure