Fault signal reconstruction for multi-sensors in gas turbine control systems based on prior knowledge from time series representation

Yang, Xilian; Zhao, Qunfei; Wang, Yuzhang; Cheng, Kanru

doi:10.1016/j.energy.2022.124996

Cited by 15 publications

(1 citation statement)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The reliability of the sensor and the correctness of the acquisition signal are the foundations of the state sensing system. Most sensors are placed in harsh environments (such as high temperatures, underwater), making them prone to fouling and damages, which will affect the accuracy, stability, and reliability of the system [8][9][10]. An early identification of faults in sensors is the most important aspect of fault detection to ensure data accuracy and reliability.…”

Section: Introductionmentioning

confidence: 99%

Study of the fault diagnosis method for gas turbine sensors based on inter-parameter coupling information

Wang,

Cheng,

Liu

et al. 2024

Meas. Sci. Technol.

Self Cite

View full text Add to dashboard Cite

Correct and reliable measurement data are very important for state monitoring, safe operation, health assessment, and life prediction of integrated energy systems. Sensors are often installed in harsh environments and prone to all kinds of faults, so it is necessary to diagnose sensor faults. In this work, a diagnosis method of sensors faults based on gradient histogram distribution (GHD) combined with light gradient boosting machine (Light GBM) was proposed, and this proposed method effectively utilizes the coupling information between the relevant parameters. The GHD effectively extracted the time-domain characteristics of sensors faults, and reduce the dimension of the eigenvectors. This is beneficial to increasing the diagnostic speed. The kernel density estimation distribution of gradient and the eigenvector for the sensor with strong correlation are similar, but that for the sensor with weak correlation are completely different. The Light GBM classifier trained based on the feature vectors was utilized to diagnose and classify the sensors faults. The diagnosis accuracy and the diagnosis time of this developed method were examined using the multiple-condition practical operation data of gas turbine in the integrated energy system. The experiment results show that the diagnostic accuracy of five sensor faults using this developed method is all above 90%. The diagnostic time is about 0.47~1.34s, and is less than 2s for the gradual faults.

show abstract

Section: Introductionmentioning

confidence: 99%