Chongqing University

Assembly Line

Chinese scientists say supersized magnesium parts pave the way for cheaper, lighter cars

๐Ÿ“… Date:

โœ๏ธ Author: Zhang Tong

๐Ÿญ Vertical: Automotive

๐Ÿข Organizations: Chongqing University, Chongqing Millison Technologies, Boao Magnesium Aluminium Manufacturing

Researchers in China say they have developed supersized magnesium alloy auto parts that could significantly reduce the cost of making cars and promote lightweight vehicle designs. The scientists produced the two giant parts โ€“ a car body and a battery box cover โ€“ from a single mould in one casting. Each part measures over 2.2 square metres (23.7 sq ft) โ€“ the first of their size to be made from magnesium alloy, according to a news release from the National Engineering Research Centre for Magnesium Alloys (CCMg) at Chongqing University on June 27.

Chongqing Millison Technologies provided the die casting system used for processing, while Boao Aluminium Manufacturing has experience in developing magnesium alloy dashboard and seat frames. They used high-pressure casting to create the two parts using a technology similar to Teslaโ€™s โ€œgigacastingโ€ process. It involves injecting molten metal into a steel mould and filling it under high pressure before cooling.

Read more at South China Morning Post

A deep transfer learning method for monitoring the wear of abrasive belts with a small sample dataset

๐Ÿ“… Date:

โœ๏ธ Authors: Zhihang Li, Qian Tang, Sibao Wang, Penghui Zhang

๐Ÿ”– Topics: convolutional neural network, predictive maintenance

๐Ÿข Organizations: Chongqing University

According to the analysis of displacement data, a new method for the prediction of abrasive belt wear states using a multiscale convolutional neural network based on transfer learning is proposed. Initially, first-order difference preprocessing is ingeniously performed on displacement data. Then, the network parameters of the model are obtained by pretraining the fault dataset and are directly transferred or fine-tuned according to the preprocessed displacement data. Finally, the preprocessed displacement data corresponding to different abrasive belt wear states are accurately classified. This method verifies the application of transfer learning between cross-domain data in industry and resolves the contradiction between the large sample size required for deep learning and the difficulty of obtaining a large amount of sample data in actual production. The experimental results show that this method can accurately predict the wear status of abrasive belts, with an average prediction accuracy of 93.1%. This method has the advantages of low cost and easy operation, and can be applied to guide the replacement time of abrasive belts in production.

Read more at ScienceDirect