公开(公告)号：US20240316536A1

公开(公告)日：2024-09-26

申请号：US18603743

申请日：2024-03-13

申请人： HEESUNG CATALYSTS CORPORATION ,  SAMSUNG ELECTRONICS CO., LTD.

发明人： Jong-San Chang ,  Seongyun Ryu ,  Seungjun Lee ,  Joungwoo Han ,  Woosung Choi ,  Yongjae Lee ,  Jungje Park ,  Taewoo Lee ,  Hyunsik Han

IPC分类号： B01J23/30 ,  B01D53/86 ,  B01J23/06 ,  B01J37/02 ,  B01J37/04 ,  B01J37/08

CPC分类号： B01J23/30 ,  B01D53/8662 ,  B01J23/06 ,  B01J37/0201 ,  B01J37/0236 ,  B01J37/04 ,  B01J37/08 ,  B01D2255/20776 ,  B01D2255/20792 ,  B01D2255/2092 ,  B01D2257/2066 ,  B01D2258/0216

摘要： Proposed are a catalyst for decomposing perfluorocompounds (PFCs) and a method of preparing the same. The provided catalyst for decomposing PFCs and the method of preparing the same are as follows. Zinc as an active component for performance improvement and tungsten (W) as an auxiliary component are added to alumina selected from at least one of gamma alumina, aluminum trihydroxide, boehmite, and pseudo-boehmite, and a weight ratio of Al, Zn, and W is at 100:30 to 100:1 to 11. The catalyst for decomposing PFCs not only has an effect of having durability against fluorine generated by decomposition of PFCs but also has a synergistic effect of improving reaction activity. Furthermore, the catalyst decomposes PFCs at a lower temperature than conventional catalysts for decomposing PFCs. Thus, it is possible to reduce operating costs and secure the durability of the system during continuous operation.

公开(公告)号：US11967308B2

公开(公告)日：2024-04-23

申请号：US17425211

申请日：2021-07-08

申请人： SAMSUNG ELECTRONICS CO., LTD.

发明人： Taewoo Lee ,  Taegyoon Kang ,  Hogyeong Kim ,  Minjoong Lee ,  Seokyeong Jung ,  Jiseung Jeong

IPC分类号： G10L15/16 ,  G10L15/06 ,  G10L15/18

CPC分类号： G10L15/16 ,  G10L15/063 ,  G10L15/18

摘要： Disclosed is an electronic device including processor and memory operatively connected to the processor and storing language model. The electronic device may enter data into the language model, generate an embedding vector in the input embedding layer, add position information to the embedding vector in the positional encoding layer, branch the embedding vector based on domain information, normalize the branched embedding vectors, enter the normalized embedding vectors into the multi-head attention layer, enter output data of the multi-head attention layer into the first layer, normalize pieces of output data of the first layer, enter the normalized pieces of output data of the first layer into the feed-forward layer, enter output data of the feed-forward layer into the second layer and normalize pieces of output data of the second layer, and enter the normalized pieces of output data of the second layer into the linearization layer and the softmax layer to obtain result data. In addition, various embodiments as understood from the specification may be also possible.