Research on State Evaluation and Prediction System of Hydraulic Turbine Based on Pressure Pulsation Parameters. Issue 1 (June 2021)

• Record Type: Journal Article
• Authors: Jiang, Xiaoping; Gao, Xiang; Hu, Jiawei; Wang, Ziting; Wang, Lele
• Is Part Of: IOP conference series. Volume 774:Issue 1(2021)
• Publication Date: 2021-06
• Subjects: Hydraulic turbine -- Pressure pulsation -- Clustering -- health condition -- Time series -- Trend prediction; Earth sciences -- Periodicals; Environmental sciences -- Congresses; Environmental sciences -- Periodicals; 550.5
• DOI: 10.1088/1755-1315/774/1/012035
• Languages: English
• ISSNs: 1755-1307

Abstract

Abstract: The pressure pulsation parameter of hydraulic turbine is one of the important indexes reflecting hydraulic stability of hydraulic turbine. Based on the pressure pulsation of the turbine, the health condition of the turbine unit is evaluated and predicted in this paper. Based on the measured data of pressure pulsation at four characteristic position of the hydraulic turbine, clustering was conducted by k-means method, and the health status of the hydraulic turbine unit was evaluated by finding the optimal k value for health status classification. The time series relationship of pressure pulsation was determined, and the evaluation result was taken as the prediction feature to establish the trend prediction model of pressure pulsation of hydraulic turbine set based on the time series change of long and short term memory network (LSTM). To realize real-time health assessment and prediction of unit running state and achieve quantitative assessment of equipment health state can provide technical basis for guiding unit maintenance.