Multiple Elastic Networks With Time Delays for Early Fault Detection and Prognostics

Dongliang Guo,Bing Song,Wen Yang,Jiale Wang,Lei Sun,Hui Liu,Fengbo Tao

doi:10.1109/access.2020.3009562

Abstract

Aiming at the problem of fault prognostics for the energy storage power station, this paper proposes a novel data-driven method named multiple elastic networks with time delays (MEN-TD). The proposed method can learn the status of the energy storage power station in advance and provide early detection of the fault. First, through the correlation analysis and the mechanism knowledge, the energy storage power station key parameter and corresponding key factors affecting the parameter are determined. Secondly, in order to predict the trend of the key parameter over a period of time and improve the prediction accuracy, the MEN-TD model is constructed. Then, based on the predicted values of the key parameter, compared with the control limit in the healthy status, the fault can be pre-warned in advance. Finally, through testing on the practical energy storage power station in Zhenjiang of China, the effectiveness and superiority of the proposed MEN-TD method are demonstrated.

Highlights

With the further requirement of the system safety of energy storage power stations, do we hope to be able to detect and isolate faults when the status deteriorates, and require that we can provide early fault detection and prognostics before the system deteriorates, to ensure that there is enough time to take measures to prevent deterioration and avoid unnecessary losses [1]–[4]
Aiming at the problem of early detection and prognostics of the fault for the energy storage power station, this paper proposes the multiple elastic networks with time delays (MEN-TD) method
First of all, considering that accurate prediction models only relying on quantitative calculation methods are not enough, quantitative calculation methods based on the mutual information and qualitative methods based on the mechanism knowledge are combined to accurately determine the key factors which affect the energy storage power station key parameter

Summary

INTRODUCTION

With the further requirement of the system safety of energy storage power stations, do we hope to be able to detect and isolate faults when the status deteriorates, and require that we can provide early fault detection and prognostics before the system deteriorates, to ensure that there is enough time to take measures to prevent deterioration and avoid unnecessary losses [1]–[4]. Aiming at the problem of early detection and prognostics of the fault for the energy storage power station, this paper proposes the multiple elastic networks with time delays (MEN-TD) method. First of all, considering that accurate prediction models only relying on quantitative calculation methods are not enough, quantitative calculation methods based on the mutual information and qualitative methods based on the mechanism knowledge are combined to accurately determine the key factors which affect the energy storage power station key parameter. In order to obtain the trend of the energy storage power station key parameter for a period of time, multiple elastic networks with time delays are established for prediction. 3) A novel multiple elastic networks with time delays method is proposed to predict the key parameter over a period of time simultaneously for early fault detection and prognostics.

PRELIMINARIES

PREDICTION MODEL CONSTRUCTION

CONTROL LIMIT DETERMINATION

EXAMPLES AND APPLICATIONS

CONCLUSION