Energy-Saving Optimization Method of Urban Rail Transit Based on Improved Differential Evolution Algorithm

Guancheng Lu, Deqiang He, Jinlai ZhangSensors 2023 · OpenReview

TL;DR

This paper introduces a novel energy-saving optimization method for urban rail transit by focusing on optimizing the train's traction curve, specifically by maximizing "coasting" (inertia motion) to reduce energy consumption. They developed a sophisticated optimization model and employed an improved differential evolution algorithm combined with Q-learning to solve it, achieving a significant 11.2% energy saving in simulations.

What Problem Does This Paper Address?

Urban rail transit systems are significant energy consumers, and reducing this consumption is crucial for economic and environmental sustainability. While upgrading infrastructure and traction equipment can save energy, it's a costly and time-consuming endeavor. This paper tackles the challenge of achieving substantial energy savings without altering existing infrastructure. The core problem is to optimize the train's operation, specifically its acceleration and deceleration profiles, to minimize energy use while adhering to critical operational constraints like passenger comfort, timetable adherence, and safety. Existing methods often focus on regenerative braking or timetable adjustments, which have limitations in practical energy recovery efficiency or oversimplification of operational conditions.

Methods

The authors propose a two-pronged approach:

  1. Optimization Model: They formulate an optimization model that prioritizes maximizing the "idle distance" (coasting or inertia motion) as the primary objective. This is a key innovation, aiming to replace braking with coasting to avoid energy conversion losses. The model incorporates several crucial constraints:
  2. Maximum allowable running speed.
  3. Passenger comfort (smooth acceleration/deceleration).
  4. Train timetable adherence.
  5. Maximum allowable acceleration.
  6. Kinematic equations governing train motion.
  7. Solution Algorithm: To solve this complex, multi-objective optimization problem, they developed an "adaptive dynamic multimodal differential evolution algorithm" (ADMD-DEA). This algorithm is an enhancement of the standard differential evolution algorithm, incorporating:
  8. An adaptive perception mechanism for better fitness gradient tracking.
  9. A dynamic multimodal mechanism to handle complex solution landscapes.
  10. Hierarchical mutation and adaptive knowledge learning for improved convergence and exploration.
  11. A Q-learning algorithm is then integrated to select the most satisfactory solution from the set of non-dominant solutions generated by the ADMD-DEA. This hybrid approach aims to balance global search with local refinement.

Datasets

The study utilizes historical operation data from the Nanning Urban Rail Transit system in China for its simulation experiments. Specific parameters from this system, such as distance between stations, speed limits, and historical energy consumption, were used as input for the simulations. No new datasets were introduced.

Key Insights and Ideas

  • Maximizing Inertia Motion: The central idea is that by strategically increasing the time trains spend coasting (using inertia), energy consumption can be significantly reduced, avoiding the losses associated with regenerative braking and conventional braking.
  • Holistic Optimization: The model moves beyond single-objective optimization by integrating multiple critical factors like speed, comfort, timetable, and acceleration into a unified framework.
  • Adaptive Algorithm Design: The development of the ADMD-DEA highlights the need for sophisticated optimization algorithms that can adapt to dynamic and multimodal problem spaces, which are common in real-world transportation systems.
  • Hybrid Approach Synergy: Combining a powerful evolutionary algorithm (ADMD-DEA) with a reinforcement learning technique (Q-learning) provides a robust mechanism for navigating complex solution spaces and selecting optimal outcomes.

Key Results

The numerical experiments demonstrated the effectiveness of the proposed method:

  • Energy Savings: The method achieved approximately 11.2% energy savings compared to historical operation data. This is a significant improvement over other compared methods, which showed savings of around 7.7% to 9.7%.
  • Operational Constraints Met: The optimized speed and acceleration curves remained within the defined limits, ensuring safety and passenger comfort.
  • Running Time: While the proposed method resulted in a slightly longer running time (around 170.5 seconds compared to the historical 168 seconds), this difference was deemed acceptable and within the allowable range for train schedules.
  • Economic Impact: Extrapolating the savings to the entire Nanning Metro Line 1, the authors estimate potential annual savings of about 5 million RMB (or 6 million RMB based on a slightly different calculation in the conclusion).

Limitations and Future Work

  • Running Time Trade-off: The proposed method, while saving energy, does lead to a slightly longer running time. Future work could explore further optimizing this trade-off.
  • Algorithm Refinement: The authors acknowledge that the adaptive dynamic multimodal differential evolution algorithm, particularly its hierarchical strategy and adaptive threshold implementation, still has room for improvement.
  • Multi-Factor Optimization: The paper focuses on optimizing the traction curve. Future research will explore optimizing combinations of multiple factors that influence urban rail transit operation to further enhance energy savings.
  • Real-World Deployment: While simulations are promising, real-world implementation and validation would be the next logical step.

Summary updated

Abstract

The transformation of railway infrastructure and traction equipment is an ideal way to realize energy savings of urban rail transit trains. However, upgrading railway infrastructure and traction equipment is a high investment and difficult process. To produce energy-savings in the urban rail transit system without changing the existing infrastructure, we propose an energy-saving optimization method by optimizing the traction curve of the train. Firstly, after analyzing the relationship between the idle distance and running energy-savings, an optimization method of traction energy-savings based on the combination of the inertia motion and energy optimization is established by taking the maximum idle distance as the objective; and the maximum allowable running speed, passenger comfort, train timetable, maximum allowable acceleration and kinematics equation as constraints. Secondly, a solution method based on the combination of the adaptive dynamic multimodal differential evolution algorithm and the Q learning algorithm is applied to solve the optimization model of energy-savings. Finally, numeric experiments are conducted to verify the proposed method. Extensive experiments demonstrate the effectiveness of the proposed method. The results show that the method has significant energy-saving properties, saving energy by about 11.2%.
BibTeX
@inproceedings{lu2023energy,
  author = {Guancheng Lu and Deqiang He and Jinlai Zhang},
  title = {{Energy-Saving Optimization Method of Urban Rail Transit Based on Improved Differential Evolution Algorithm}},
  booktitle = {Sensors},
  year = {2023},
  url = {https://openreview.net/forum?id=zRUJVvoQrg},
}
Figure