Prediction of Health Status and Fault of Electrical Equipment Based on the Concept of Digital Twin

Xuan, Yi; Huang, Jinkui; Liu, Shipeng; Lu, Guoming; Yuan, Jian

doi:10.1007/978-981-19-1870-4_126

Cited by 1 publication

(1 citation statement)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the radial feeder model of multiple RG connections, the proposed reactive power control method is simulated by using the actual solar energy output data. Yu et al [27] firstly studies and determines which equipment failure is suitable to be predicted by using the concept of digital twinning and establishes a mathematical and physical model describing the fault or health state, which has undetermined coefficients related to the measurable state characteristics of the equipment. Then, a digital twin model is established on the basis of the mathematical and physical model, and the consistency with the equipment state is realized by adjusting the undetermined coefficients.…”

Section: Introductionmentioning

confidence: 99%

Research on State Detection Method of Electrical Equipment Based on Wireless Sensor Network Signal Processing

Liang¹,

Ding²,

Jin-fa³

et al. 2022

View full text Add to dashboard Cite

False alarms and omissions of key measurement signals may even affect the reliability of decision-making in power IoT system, lead to safety accidents and bring huge economic losses. In order to improve the reliability of system operation and the information management level of electrical equipment, it is necessary to identify and extract suddenly changed signals from massive measurement signals collected by wireless sensor networks, and to detect the working state of electrical equipment by judging the source of signals. Therefore, this article studies the state detection method of electrical equipment based on wireless sensor network signal processing. In the second chapter, a cluster splitting and merging method is designed to solve the problem that the existing detection methods tend to ignore the imbalance of cluster size. In the third chapter, according to the data characteristics of measurement signals collected by wireless sensor networks, a similarity measurement criterion for composite time series of measurement signals is proposed, and the corresponding distance matrix is generated based on this criterion. In the fourth chapter, wavelet decomposition is used to decompose the initial measurement signals collected by wireless sensor networks, and then the signals are compressed twice based on compressed sensing. Then the abnormal signal data information is imported into support vector machine for training to realize the real-time detection of abnormal signals of electrical equipment state. Experimental results verify the effectiveness of the proposed algorithm.

show abstract

Section: Introductionmentioning

confidence: 99%