AI Predicts Manhole Explosions In New York City
reillymj writes "Every so often, a 300-pound manhole cover blows sky high in Gotham, followed sometimes by a column of flame and smoke. (There are a few hundred 'manhole incidents' per year in the city, not all of them this dramatic.) Researchers from Columbia University applied machine learning algorithms to Con Edison's warren of aging electrical wires and sewage access points around Manhattan. As the system learns where dangerous mixtures of sewer gas and decrepit wiring are likely to come in contact, it makes forecasts about trouble spots, including where the next explosion may occur. The team has just completed rankings for manholes in Brooklyn and the Bronx, and plans to return to Manhattan's grid, armed with the most recent inspection and repair data." The research was published in the July issue of Machine Learning.
I wonder if anybody has ever died from being hit on the head with one of these, seems it is likely. Shouldn't there be a way to secure the covers to the ground with a bolt that would at least cause the cover to not fly up but just turn over in case of an explosion?
You can't handle the truth.
That's a joke, right? Exploding manhole covers? In pre-Snake Plissken-New York? OMG
Say out loud: I'm an Aspie and I'm somewhat proud, I guess. Uh. Can I write an email in all caps instead? Hm...
So we have millions of people desperate for work, and a whole lot of dangerous wiring all over New York City. Why aren't we (and by "we", I mean ConEd or any level of government) investing in training up as many electricians as we can and replacing the bad wiring while it's relatively cheaper to do so?
Oh, wait, maybe because there are no financial consequences to any organization if a manhole cover gets launched 300' up and lands on some 3-year-old.
I am officially gone from
Southern California Edison requires two ventilating pylons for each manhole. They are much bigger than you might expect, and need to be spaced apart. It greatly increases planning complexity.
Methane build-up is only one cause though. Venting that causes ...odors... that people tend to not want to be near. The more common cause is failure of oil-filled equipment ranging from link switches to transformers to oil-insulated cables. When these go you need someplace for the explosion to expand to... or you will destroy everything in the manhole.
This is an interesting solution to the problem, but I have trouble understanding how it is more effective than root-cause analysis and post-incident review of data they already have. It isn't like the combination of factors is the problem... more like aging and over-burdened equipment that should already be on a predictive-maintenance plan.
It's sewage pipes. There's some water in there, of course, and it could overflow with the next thunderstorm.