Identification of Switching Operation Based on LSTM and MoE

Authors

Xiaoqing Zhang, 1.Institute of Computing Technology, Chinese Academy of Sciences, Beijing 100190, China;2.College of Computer, Beijing University of Posts and Telecommunications, Beijing 100876, China;Follow
Wanfang Xiao, 3.State Grid Beijing urban power supply company, Beijing 110102, China;
Yingjie Guo, 1.Institute of Computing Technology, Chinese Academy of Sciences, Beijing 100190, China;Follow
Bowen Liu, 3.State Grid Beijing urban power supply company, Beijing 110102, China;
Xuesen Han, 3.State Grid Beijing urban power supply company, Beijing 110102, China;
Jingwei Ma, 3.State Grid Beijing urban power supply company, Beijing 110102, China;
Gao Gao, 3.State Grid Beijing urban power supply company, Beijing 110102, China;
He Huang, 3.State Grid Beijing urban power supply company, Beijing 110102, China;
Shihong Xia, 1.Institute of Computing Technology, Chinese Academy of Sciences, Beijing 100190, China;

Abstract

Abstract: Aiming at the individual differences of different personnel in the same operation and differences of the same person in the same operation at different times, a switching operation recognition model(MoE-LSTM) based on Mixture of experts model (MOE) and long short-term memory network(LSTM) is proposed. Based on MoE, LSTM is integrated to learn the feature distribution of different sources data. The acceleration data is collected to build the switching operation dataset and the action sequence is segmented and aligned based on sliding window. The action sequence is input to MoE-LSTM, and the temporal dependencies of different actions are independently learned by different LSTMs. The gating network selects the output of LSTM that classifies the current input better as the action recognition result. The result of model learning is that for action data from different time and space, different LSTMs perform better in a certain feature area than other LSTMs. The experiments on the switching operation dataset demonstrate superior performance of the proposed method compared to other existing action recognition algorithms.

Recommended Citation

Zhang, Xiaoqing; Xiao, Wanfang; Guo, Yingjie; Liu, Bowen; Han, Xuesen; Ma, Jingwei; Gao, Gao; Huang, He; and Xia, Shihong (2022) "Identification of Switching Operation Based on LSTM and MoE," Journal of System Simulation: Vol. 34: Iss. 8, Article 22.
DOI: 10.16182/j.issn1004731x.joss.21-0288
Available at: https://dc-china-simulation.researchcommons.org/journal/vol34/iss8/22

First Page

1899

Revised Date

2021-05-08

DOI Link

https://doi.org/10.16182/j.issn1004731x.joss.21-0288

Last Page

1907

CLC

TP391.9

Recommended Citation

Xiaoqing Zhang, Wanfang Xiao, Yingjie Guo, Bowen Liu, Xuesen Han, Jingwei Ma, Gao Gao, He Huang, Shihong Xia. Identification of Switching Operation Based on LSTM and MoE[J]. Journal of System Simulation, 2022, 34(8): 1899-1907.

Corresponding Author

Yingjie Guo,guoyingjie@ict.ac.cn

DOI

10.16182/j.issn1004731x.joss.21-0288