TY - GEN
T1 - Research on Operational Condition Monitoring Strategy for Experimental Equipment of Space Environment Simulation and Research Infrastructure
AU - Tong, Weiming
AU - Chu, Xu
AU - Shen, Zhixiong
AU - Pang, Long
AU - Wang, Xiaoye
N1 - Publisher Copyright:
© 2023, Beijing Paike Culture Commu. Co., Ltd.
PY - 2023
Y1 - 2023
N2 - For the space environment simulation test field of complete sets or individual experimental equipment, the condition monitoring is necessary to ensure the normal operation of the equipment. In this paper, we propose a condition monitoring strategy for experimental equipment based on power load characteristic analysis technology. First, the typical state of the experimental equipment is analyzed and the data on the electrical load of the equipment is collected. Secondly, the K-means clustering algorithm was used to classify the collected data and construct a library of features corresponding to each typical state; after that, a neural network model was built and model optimization was carried out to achieve the function of equipment condition monitoring; finally, the feasibility of the proposed strategy is verified by two types of equipment in the EMBED dataset. The simulation results show that the proposed equipment condition monitoring strategy can realize the condition monitoring of experimental equipment to a certain extent, and the condition monitoring effect is better for the equipment with rapid state switching.
AB - For the space environment simulation test field of complete sets or individual experimental equipment, the condition monitoring is necessary to ensure the normal operation of the equipment. In this paper, we propose a condition monitoring strategy for experimental equipment based on power load characteristic analysis technology. First, the typical state of the experimental equipment is analyzed and the data on the electrical load of the equipment is collected. Secondly, the K-means clustering algorithm was used to classify the collected data and construct a library of features corresponding to each typical state; after that, a neural network model was built and model optimization was carried out to achieve the function of equipment condition monitoring; finally, the feasibility of the proposed strategy is verified by two types of equipment in the EMBED dataset. The simulation results show that the proposed equipment condition monitoring strategy can realize the condition monitoring of experimental equipment to a certain extent, and the condition monitoring effect is better for the equipment with rapid state switching.
KW - Analysis of power load characteristics
KW - Clustering algorithm
KW - Equipment state monitoring
KW - Neural networks
UR - https://www.scopus.com/pages/publications/85152620786
U2 - 10.1007/978-981-99-0451-8_108
DO - 10.1007/978-981-99-0451-8_108
M3 - 会议稿件
AN - SCOPUS:85152620786
SN - 9789819904501
T3 - Lecture Notes in Electrical Engineering
SP - 1071
EP - 1078
BT - The Proceedings of the 17th Annual Conference of China Electrotechnical Society
A2 - Li, Jian
A2 - Xie, Kaigui
A2 - Hu, Jianlin
A2 - Yang, Qingxin
PB - Springer Science and Business Media Deutschland GmbH
T2 - 17th Annual Conference of China Electrotechnical Society, CES 2022
Y2 - 17 September 2022 through 18 September 2022
ER -