Safer freight trains – manual inspection reduced by 95% 

Dongguan, China, July 10, 2023 – Huawei Cloud unveiled its groundbreaking Pangu Railway Model during Huawei HDC Cloud’s “Reshaping Industries with AI” conference as it launched the highly anticipated Pangu Model 3.0 and Ascend AI cloud services. This innovative AI-based solution is revolutionising fault identification processes in the rail freight industry, increasing critical safety standards, boosting efficiency and reducing labour intensity.

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

The railway industry faces significant challenges when it comes to the identification of train faults. Current methods, such as Train Freight Detection Systems (TFDS), rely heavily on manual inspection routines and are labour-intensive, inefficient, and costly. To address these pain points, Huawei’s Pangu Railway Model leverages the power of artificial intelligence to automate the fault identification process, reducing the chances of human error and therefore ensuring safer operations and improved efficiency for railway freight services.

Current TFDS Solution

Train detection systems sit alongside and in-between the tracks of railways systems. When a train passes the detection station, TFDS calculates the train speed using the magnetic steel sensors on wheels. It automatically adjusts the snapshot frequency to match the train passing time, usually one picture taken every few milliseconds. These are then uploaded to a central server for fault detection. Due to technical limitations, traditionally most of the fault identification work is done manually.

A typical depot in China monitors more than 40,000 trains running on 800 lines every day making it extremely difficult to conduct manual checks. The TFDS system takes around 80 pictures for each freight carriage, as each train has around 50 carriages this results in thousands of pictures being taken for each train. Due to the sheer volume of pictures taken, inspectors only have around 10 minutes to manually check each batch for faults and cracks in the undercarriages.

The Pangu Railway Model brings significant benefits to the railway industry

On railways, the Pangu Railway 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 system 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 both efficiency and accuracy. This AI-based solution has revolutionised the fault identification process, guaranteeing safer railway operations and enhancing reliability.

“The capability of the Pangu Railway large model is continuously improved. Now, we can accurately identify 67 types of trucks running on the live network and more than 430 types of faults. The fault detection omission rate is zero, and the screening rate of fault-free images is as high as 95%.” said Zhang Pingan, Huawei Executive Director and CEO of HUAWEI CLOUD.

The Pangu Railway Model also uses continuous learning to improve performance. If abnormalities are detected, they can be manually labelled, and added to the model to improve detection performance.

Additionally, the Pangu Railway Model supports concurrent detection on distributed servers, significantly reducing the time required for fault detection. Using a two-node solution, the detection process for a train is completed within 8 minutes. The faster detection process ensures timely identification of faults, mitigating risks and enhancing overall railway safety.

With its exceptional accuracy, improved efficiency, and faster detection capabilities, Huawei’s Pangu Railway Model solution is setting new standard for fault identification in the railway sector

Pangu Railway is one of several AI models that has been announced to date, including Pangu Drug Molecule Model, Pangu Weather Model, Pangu Government Model, Pangu Finance Model and Pangu Manufacturing Model all with results outperforming traditional models. Most recently research about Pangu Weather Model was published in the highly respected scientific journal Nature magazine detailing how a 10000x improvement in the prediction of weather reducing time to just seconds.

These case studies exemplify the transformative power of AI in industrial applications. By harnessing the capabilities of Huawei’s Pangu 3.0 and Ascend AI cloud services, businesses can unlock new levels of safety, efficiency, and productivity, driving substantial growth and competitive advantage in their respective industries. Huawei’s commitment to innovation and the development of advanced AI solutions underscores its commitment to providing cutting edge solutions for customers.