China

AI-powered smart train joins Qingdao Metro family

By Zhao Ruixue and Hu Qing | chinadaily.com.cn

Updated: 2026-11-12 20:11

Share - WeChat
An AI-powered, fully automated train for the Qingdao Metro Line 7 rolls off the production line on Nov 12 in Qingdao, Shandong province. [Photo by Zhang Jingang/for chinadaily.com.cn]

A smart train automated entirely by artificial intelligence rolled off the production line on Thursday in Qingdao, Shandong province.

Jointly developed by Qingdao Metro and CRRC Qingdao Sifang, a major high-speed train manufacturer, the train is the first to be built for Qingdao's Metro Line 7, a major north-south route currently under construction.

Powered by CRRC's industrial large model called Zhuolun, the train features GoA4, the highest level of automation in rail transit. It has a maximum operating speed of 100 kilometers per hour and uses a six-car stainless steel Type B body, with a maximum passenger capacity of 1,862, said Zhang Zhenxian, deputy general manager and chief engineer of CRRC Qingdao Sifang.

Equipped with more than 2,600 monitoring points, the train can collect real-time data on its operating status and surrounding environment. AI algorithms analyze data to detect potential faults or abnormalities in advance, he said.

By integrating data from onboard, trackside and ground systems, the technology is expected to improve fault-handling effectiveness by 60 percent and extend routine maintenance intervals by more than 80 percent, he said.

A range of energy-saving technologies has also been employed to reduce the train's overall energy consumption by more than 15 percent. Its flexible design allows train lengths to be adjusted to passenger demand, with longer trains during peak hours and shorter ones during off-peak periods.

Feng Jianghua, an academician of the Chinese Academy of Engineering, said at the follow-up conference on modern industrial chain development that artificial intelligence is profoundly reshaping the rail transit industry, moving from pilot demonstrations toward large-scale deployment.

Copyright 1994 -

Registration Number: 130349

Mobile