China Puts Trust in AI To Maintain Largest High-Speed Rail Network on Earth

China is using AI in the operation of its 45,000km (28,000-mile) high-speed rail network, with the technology achieving several milestones, according to engineers involved in the project. From a report: An AI system in Beijing is processing vast amounts of real-time data from across the country and can alert maintenance teams of abnormal situations within 40 minutes, with an accuracy as high as 95 per cent, they said in a peer-reviewed paper. "This helps on-site teams conduct reinspections and repairs as quickly as possible," wrote Niu Daoan, a senior engineer at the China State Railway Group's infrastructure inspection centre, in the paper published by the academic journal China Railway.

In the past year, none of China's operational high-speed railway lines received a single warning that required speed reduction due to major track irregularity issues, while the number of minor track faults decreased by 80 per cent compared to the previous year. According to the paper, the amplitude of rail movement caused by strong winds also decreased -- even on massive valley-spanning bridges -- with the application of AI technology. [...] According to the paper, after years of effort Chinese railway scientists and engineers have "solved challenges" in comprehensive risk perception, equipment evaluation, and precise trend predictions in engineering, power supply and telecommunications. The result was "scientific support for achieving proactive safety prevention and precise infrastructure maintenance for high-speed railways," the engineers said.

It's pretty slick

[alvinrod]

The AI was able to determine that someone with unacceptable social credit score had slipped on board and automatically rerouted to a re-education camp to drop them off. Quite impressive if you ask me.

Re:

[backslashdot]

Not sure who modded you down, but that was funny. Lighten up people.

95% reliable? Great. For the other

[hdyoung]

5% that results in flaming high-speed crashes, they can quickly dig a hole, bulldoze and bury the fiery wreckage. Without bothering to do a full-blown failure analysis or even to recover all the corpses, of course.

This is not hyperbole. They did exactly that. Look it up. Later, they released an engineering report that was basically “mistakes were made. We fixed it, Nothing to see here, citizen, move along”

Re:

[Anonymous Coward]

Not quiet. Was the train buried? Yes, the rest of your idiotic comment is just conspiracy bullshit. Stop being a gullible, weak minded fool

https://en.wikipedia.org/wiki/... [wikipedia.org]

Re:

[thegarbz]

5% that results in flaming high-speed crashes

That's not how predictive maintenance works. The 5% that is missed will get caught during preventative maintenance later and likely be costlier and slower to repair.

95% is pretty damn good for predictive maintenance systems. I'd question if it is actually that high.

Re:

[gweihir]

Indeed. For a system that supplements regular maintenance and just catches some problems earlier, 95% is perfectly fine. Context and details matter. Too bad most people cannot handle either.

Blaine is a pain and that is the truth.

[mattfosser]

Blaine the Mono was a sentient monorail that went insane over the course of its long lifetime. He became suicidal and attempted to kill Roland and his ka-tet in his suicide, but was ultimately defeated.

Re: Blaine is a pain and that is the truth.

[LindleyF]

Say thankya.

Backwards

[RitchCraft]

I thought China wanted to stop their population decline.

Re:

[sdinfoserv]

it's too late... If you've run out of 30 year old's today to perform work, the time to fix that was 31 years ago. Reaping the rewards of the 1 child policy.

the great firewall of china

[sdinfoserv]

So when it hallucinates resulting in massive causalities, I'm sure we'll hear about it....

Some of the data types that are being analysed

[EreIamJH]

Looks like the AI integrates engine and train instrumentation data with track and environmental instrumentation, identifies patterns and reports anomolies for investigation:

The data comes from various sources and formats, including dynamic waveform values captured by wheel sensors, records of train body movements, rail vibrations, and meteorological records.

According to the paper, fluctuations in power grid current amplitudes – and even electromagnetic spectrum monitoring records – have been used to train the system.

AI’s advantage lies in its ability to analyse diverse data, identify potential problem-related clues and uncover previously unknown connections within seemingly chaotic data sets. This enables more precise fault identification and prediction.

Niu’s team said the technology had improved the efficiency of new data analysis by 85 per cent. Previously, the maintenance management headquarters in Beijing was able to issue nationwide warnings once a week. Regular reports are now issued daily.

Re:40 minutes is an eternity for a high-speed trai

[thegarbz]

Maintenance and operations are not the same thing. An abnormal situation doesn't mean anyone is driving blind, and has little to do with critical faults.

The reality is this system is reacting orders of magnitude faster than it needs to. Prior to this trains would drive until preventative maintenance looks at them, or until they have an unexpected breakdown. 40min to give maintenance teams an alert of what they can expect on their next shift when the train goes back to the depot is more than enough.

Duh

[Muteki719]

Ask any engineerâ¦AI with networked satcom is the solution. Automated âoelast mileâ through robotics.