China is using synthetic intelligence (AI) to turn a dam on the Tibetan plateau into the world’s largest 3D printer, according to scientists involved in the task.
The 180-meter (590-foot) tall Yangqu hydroelectric power plant will be built piece by piece (excavators, trucks, bulldozers, finishers and unmanned rollers, all controlled by AI) with the same additive production procedure used in 3D printing.
Once completed in 2024, the Yangqu Dam will send about five billion kilowatt hours of electric power a year from the upper Yellow River to Henan, the cradle of Chinese civilization and home to one hundred million people.
The electric power will pass through a 1,500 km (932-mile) high-voltage line built exclusively for green energy transmission.
According to the project’s lead scientist, Liu Tianyun, in a paper published in the peer-reviewed Journal of Tsinghua University (Science and Technology), the dam structure and 3D printing are “identical in nature. “
After years of progressive testing, 3D printing generation for large infrastructures had matured enough for massive programs and would “free humans from heavy, repetitive and harmful work,” he said.
Liu, a research associate at Tsinghua University’s State Laboratory of Engineering and Hydroscience, and his team came up with the idea of “printing” large-scale projects a decade ago.
His idea is that a complete structure can become a giant printer, with a large number of automated machines running in combination seamlessly like other components.
First, the 3D printer evolved as a less expensive way to make parts from valuable fabrics. Printing, or adding, fabrics produces less waste than cutting and grinding.
Since then, some architects have begun to apply generation to buildings, the projects so far have been modest. The first 3D published work building, the headquarters of the Dubai Future Foundation, is only six meters (20 feet) high.
Chinese civil engineers are no strangers to AI, which has been used to build Baihetan, the largest dam of the moment in the world, in just 4 years. But so far, it has basically played a coordinating role in the projects.
Tests of the generation in previous structure projects have indicated that intelligent machines can do a greater task than humans, “especially in some hostile and harmful environments,” Liu and his colleagues said.
Liu did not immediately respond to questions about the progress of the Yangqu Dam, but according to state media, work began late last year in Hainan Tibetan Autonomous Prefecture in Qinghai province.
After “cutting” a computer style from the dam into layers, the AI at the center of the task would assign a team of robots to load one layer at a time, according to the newspaper.
Unmanned excavators will identify and load fabrics from a tank park into a fleet of automated trucks, some powered by electricity.
Following an optimized direction calculated through core AI, trucks will deliver the right fabrics to the right places, at the right time, to pass through robotic excavators and cobblestones and remodel into a layer of the dam structure.
Automated rollers tighten the added layer until it tightens and becomes firm, but are also supplied with sensors. Central AI uses them to monitor construction quality by analyzing floor vibrations and other data.
Advances in AI technology, which add deep reinforcement learning, mean machines can now recognize almost any object in facilities, cope with uncertainties in a conversion environment, and carry out responsibilities flexibly, according to the article.
They also don’t make human mistakes. Liu said truckers delivered the fabrics to the wrong place, while bumps and strong vibrations prevented curling operators from maintaining a perfectly direct trajectory. And the maximum staff could not read the engineering design documents well, he added.
But where machines shine is in their ability to work in a potentially fatal environment, having headaches from lack of oxygen or exhaustion after running non-stop for 24 hours, according to the researchers.
Not all dam paintings will be carried out with machinery. The team said the extraction of filler rocks from nearby mountains would be done manually due to the complexity of the task.
Liu’s team said the generation can also be used in other infrastructure projects, such as the construction of airports and roads.
“Artificial intelligence about knowledge, data and knowledge is a new tool. . . that will shape our future,” they said.
A Nanjing-based civil scientist, who asked to be named because of his role in the technical evaluation of some primary infrastructure projects, said there were limits to 3D printing technology but would find more uses in the future.
ALSO READ: As robot dogs take to the streets of China, can they one day reflect the corporation of man’s friend?
“You can’t print a design made of other materials, such as reinforced concrete made of metal and cement,” the scientist said.
“An army of structure robots can compensate for the sharp drop in manual hard work through the low birth rate,” he added.
This article was first published in the South China Morning Post.