This paper focuses on the prognosis problem in manufacturing of the electronic chips for devices. Electronic devices are of great importance at present, which are popularly applied in daily life. The basis of supporting the electronic device is the powerful electronic chip and its manufacturing technology. Chip manufacturing has been one of the most important technologies in recent years. The etching machine is the key equipment in the etching process of the wafers in chip manufacturing. Due to the high demands for precise manufacturing, monitoring the health state and predicting the remaining useful life (RUL) of the etching system is quite important. However, the task is very hard because of the lack of knowledge of exact onset of failure or degradation and the multiple operating conditions, etc. This paper proposes a novel deep learning-based RUL prediction method for the etching system. The transformer module and random forest are integrated in the methodology to identify the health state of the machine and predict its RUL, through training with the complex data of the etching machine’s sensors and exploring its underlying features. The experiments are based on the subject of the 2018 PHM Data Challenge—for estimating time-to-failure or RUL of Ion Mill Etching Systems in an online fashion using data from multiple sensors. The results indicate the proposed method is promising for the real applications of the prognosis of the etching system for electronic devices.

Lightweight blades with high strength are urgently needed in practical rotor engineering. Sandwich structures with porous core and reinforced surfaces are commonly applied to achieve these mechanical performances. Moreover, blades with large aspect ratios are established by the elastic plate models in theory. This paper studies the vibration of a rotating sandwich pre-twist plate with a setting angle reinforced by graphene nanoplatelets (GPLs). Its core is made of foam metal, and GPLs are added into the surface layers. Supposing that nanofillers are perfectly connected with matrix material, the effective mechanical parameters of the surface layers are calculated by the mixing law and the Halpin–Tsai model, while those of the core layers are determined by the open-cell scheme. The governing equation of the rotating plate is derived by employing the Hamilton principle. By comparing with the finite element method obtained by ANSYS, the present model and vibration analysis are verified. The material and structural parameters of the blade, including graphene nanoplatelet (GPL) weight faction, GPL distribution pattern, porosity coefficient, porosity distribution pattern, length-to-thickness ratio, length-to-width ratio, setting angle and pre-twist angle of the plate are discussed in detail. The finds provide important inspiration in the designing of a rotating sandwich blade.

Freestyle skiing U-shaped field is a snow sport that uses double boards to perform a series of action skills in a U-shaped pool, which requires very high skills for athletes. In this era of deep learning, in order to develop a more scientific training method, this paper combines multitarget tracking algorithm and deep learning to conduct research in freestyle skiing U-shaped venue skills motion capture. Therefore, this paper combines the convolutional neural network and multitarget tracking algorithm in deep learning to study the human action recognition technology, and then uses the LSTM module to study the freestyle skiing U-shaped venue skills. Finally, this paper designs the training method of the action recognition algorithm of the freestyle U-shaped skiing skills multitarget tracking algorithm based on deep learning. This paper also designs multitarget tracking dataset experiments and model updating experiments. Based on the data of experimental analysis, the training method designed in this paper is optimized, and finally compared with the traditional training method. Compared with the traditional freestyle U-shaped skiing skills training method, the experimental results show that the training method of the freestyle U-shaped skiing skills multitarget tracking algorithm action recognition algorithm is based on deep learning designed in this paper and this improves the skill score by 14.48%. Most professional students are very satisfied with the training method designed in this paper.