A data-driven power consumption estimation algorithm of catenary for electrified railway
Power consumption for electrified railway not only reflects the power supply capacity, energy supply efficiency and energy saving level, but also indicates potential defects and risks. However, due to the complexity of the traction network and the randomness of trains, there is a challenge to estima...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-08-01
|
| Series: | International Journal of Electrical Power & Energy Systems |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S0142061525002881 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Power consumption for electrified railway not only reflects the power supply capacity, energy supply efficiency and energy saving level, but also indicates potential defects and risks. However, due to the complexity of the traction network and the randomness of trains, there is a challenge to estimate the power consumption accurately and economically. This paper proposes a data-driven power consumption estimation algorithm of catenary and trains for electrified railway. The unit impedance of catenary is identified via a one-train condition based on the simultaneous real-time electrical data of traction substation and section post. The underdetermined issue of multi-train condition can be converted into a problem within a finite solution domain, depending on either the average speed, speed limit, or both. Considering the time dependence and the displacement continuity of trains, the power consumption can be obtained for conditions with different number of trains operating in a power supply section. Finally, the proposed method is validated by field test and simulation data from traction and power supply calculations. Taking the field test as a reference, the identified result is accurate with an error of −0.75 % compared to the conventional method. No extra monitoring devices and computing power are required in the algorithm. |
|---|---|
| ISSN: | 0142-0615 |