Chinese tech giant Huawei is leveraging large-scale data sets and machine-learning algorithms for AI applications in weather forecasting, drug development, fault identification in trains, and the mining industry.

During his keynote speech at last week’s Huawei Developer Conference in China, Zhang Ping’an, CEO of Huawei Cloud, unveiled the Pangu 3.0 models and Ascend AI cloud services.

These aim to empower industrial users and partners to unlock the potential of artificial intelligence (AI) for transformative growth across a range of sectors.

Pangu 3.0 is a system of pre-trained models that can be quickly adapted to meet scenario-specific needs and address complex problems across multiple industries. By leveraging large-scale data sets and machine-learning algorithms, Pangu 3.0 is set to tackle the industrial application of AI in diverse areas such as weather forecasting, drug development, fault identification in trains, and the mining industry.

“Huawei Cloud Pangu models will empower everyone from every industry with an intelligent assistant, making them more productive and efficient,” said Zhang Ping’an. “We will uphold our mission of AI for industries, and use Pangu models to reshape all industries with AI.”

Science journal Nature recently published a paper describing the potential AI holds for weather prediction. The paper described how Huawei’s Pangu weather AI model team developed a precise and accurate global AI weather forecast system based on deep learning using 43 years of data.

This is the first AI prediction model to demonstrate higher precision than traditional numerical weather forecast methods. The model allows a 10,000 times improvement in prediction speed, reducing global weather prediction time to just seconds. The model focuses on key elements and common time ranges, allowing for more precise weather forecasts. Notably, it has accurately predicted the trajectories and landing times of typhoons, including the recent Typhoon Mawar in May this year.

Next week, Huawei and Shandong Energy Group will jointly launch the first commercial case of the Pangu mining model in Jinan, Shandong province. It should enhance safety, efficiency and productivity. Traditional mining operations are inherently risky, requiring extensive manpower and posing significant technical and safety problems. However, with the implementation of digital technologies, augmented by Pangu, the industry could see a transformative shift. By importing massive amounts of data for model pre-training, the model enables unsupervised self-learning, covering over 300 core scenarios in coal mining operations. From digging and drivage to machinery, transport and communication, the model streamlines AI applications in numerous coal mining scenarios.

Transportation system failures in coal mines are a significant issue, resulting in decreased coal output and financial losses. Addressing this, the model incorporates an AI-based intelligent monitoring system that accurately identifies exceptions in the transportation system, such as large coal blocks and anchors, with an accuracy rate of up to 98%.

The railway industry can also benefit from these AI advancements. Through the implementation of the Pangu railway model, the safety and efficiency of freight trains can be improved. Previous fault identification processes within train freight detection systems (TFDS) have been labour-intensive, inefficient and costly. The model can accurately identify 67 types of freight cars and over 430 types of faults found on railways and freight cars. It can quickly scan millions of images captured by the railway TFDS and filter out 95% of the images that are fault-free. This way, train inspectors can focus on the remaining images, and this helps them improve efficiency and accuracy.