China’s Huawei is once again challenging traditional weather forecasting, this time with AI model Zhiji

Since then Released in August last yearPangu has revolutionized weather forecasting, offering faster and more accurate forecasts than traditional meteorological methods.
Pingo – First season Burst on the scene In July 2023, when a paper detailing the AI ​​model was published in the journal Nature. A month later, it was launched on the website of the European Center for Medium Range Weather Forecasts (ECMWF).
WhatsApp Group Join Now
Telegram Group Join Now
Instagram Group Join Now

The AI ​​model reached a major milestone when it was able to complete a seven-day weather forecast in just 10 seconds – 10,000 times faster than traditional methods.

Then on February 29, a few months after its inception, there was Pangu-Weather. At number one Ranked among China’s top 10 scientific developments in 2023 by the National Natural Science Foundation of China (NSFC).
Chinese technology company Huawei Technologies is leading the climate revolution with its fast and accurate weather forecasting AI models. Photo: Reuters
“In its recognition by the NSFC, Pangu had two major achievements: First, it improved the world’s leading ECMWF weather forecasting system by about 0.6 days. This means that it is highly predictive; can Weather Earlier and more accurate,” Science and Technology Daily reported. “The second is 7-day predictions in 10 seconds, 10,000 times faster than numerical predictions.”

According to a Huawei report in late February, Pango made more accurate predictions than numerical simulations for weather elements such as temperature, pressure, humidity and wind speed. In addition, its margin of error for forecasting tropical cyclone tracks was 25% lower than that of ECMWF.

This is quite an achievement for an AI model, which is so fast. Changed the face Global weather forecast. By leveraging AI to predict weather patterns, scientists can bypass the complexities associated with traditional forecasting methods. AI does not require any mathematical physics knowledge or expert experience, which has created a new path for weather forecasting.

Now, researchers have used Pangu as a basis to develop the new regional model Zhiji.

Created in collaboration with the Shenzhen Meteorological Bureau, Zhijie is trained with high-resolution data from southern China.

According to the Huawei team, Zhiji can provide a five-day forecast for Shenzhen and its surrounding areas with an accuracy of 3 kilometers. While the Central Meteorological Bureau already provides hourly forecasts with street-level accuracy, these are usually only available for the next 24 hours.

“Ziji is predictable. Basic climatic elements Like wind speed, temperature, humidity and rainfall. Since its trial operation began in February, it has provided valuable insights to the Shenzhen Meteorological Bureau on multiple occasions,” Huawei reported in late March.


How Huawei’s use of 5G and AI is changing China’s coal mining industry

How Huawei’s use of 5G and AI is changing China’s coal mining industry

Currently, AI and manual predictions each have their strengths and weaknesses.

AI excels in predicting the paths of storms. While numerical simulations are more accurate in determining wind strength values.

“Scientists can now integrate numerical simulation results with predictions provided by Ziji to make the most beneficial decisions,” said a Huawei spokesperson. “This may be a trend in the future.”

According to researchers, this year Flood season Zhiji 1.0 will be the real test. They expect the model to be further refined with subsequent improvements to the algorithm.

Ongoing work on the technology aims to enhance its rainfall forecasting capabilities, including providing specialized forecasts such as heatstroke indices and comfort levels, and improving the resolution of heavy rainfall forecasts to 1 km.

“For example, me Storm conditionsClimate models can predict rainfall on road surfaces, providing early warnings for urban drainage systems, Huawei said.
As with Zhiji, if regional data from other regions are available for training, The scientists These areas could potentially develop localized models, serving more cities.

In December last year, the team announced a collaboration with the Thai Meteorological Department, with related products currently being developed.

WhatsApp Group Join Now
Telegram Group Join Now
Instagram Group Join Now

Leave a Comment