A novel comprehensive evaluation method of the draft tube pressure pulsation of Francis turbine based on EEMD and information entropy. (1st February 2019)

• Record Type: Journal Article
• Authors: Wang, Weiyu; Chen, Qijuan; Yan, Donglin; Geng, Dazhou
• Is Part Of: Mechanical systems and signal processing. Volume 116(2019)
• Page Start: 772
• Page End: 786
• Publication Date: 2019-02-01
• Subjects: Francis turbine -- Pressure pulsation -- State evaluation -- EEMD -- Information entropy; Structural dynamics -- Periodicals; Vibration -- Periodicals; Constructions -- Dynamique -- Périodiques; Vibration -- Périodiques; Structural dynamics; Vibration; Periodicals; 621
• DOI: 10.1016/j.ymssp.2018.07.033
• Languages: English
• ISSNs: 0888-3270

Abstract

Graphical abstract: Highlights: EEMD is used to analyse the nonlinear and non-stationary DTPP signal. An targeted EEMD noise reduction method is designed for the DTPP signal. Information entropy theory is introduced for the state evaluation of DTPP. A comprehensive evaluation method of the DTPP is proposed. Abstract: Undesirable pressure pulsations will be induced by flow instabilities in the draft tube when the Francis turbine operates at off-design conditions, which have multiple effects on the steady operation, and the effects vary with the operating condition. Therefore, the establishment of a comprehensive state evaluation index of the draft tube pressure pulsation (DTPP) is of great practical significance. In this study, a comprehensive evaluation method of DTPP has been proposed based on Ensemble Empirical Mode Decomposition (EEMD) and information entropy. First, the frequency components of DTPP are separated by the EEMD method which is more suitable for the analysis of non-stationary pressure pulsation signal. Second, to eliminate the effects of noise, a targeted EEMD noise reduction method which is combined with autocorrelation analysis and wavelet soft-threshold de-noising method is designed for the DTPP signal. Then, index energy values of all frequency components are calculated and normalized, and the values of the signals' characteristic entropy are obtained by using the normalized index energy vector as the input vector. Finally, characteristic entropy values of all pressure pulsation signals measured from the draft tube are integrated into an index, which is named as the comprehensive characteristic entropy of DTPP (E_DTPP). The state of DTPP is comprehensively evaluated by E_DTPP. This method was evaluated by the actual measuring data of a 200 MW hydroelectric generating unit. The case study indicates that this single index can be utilized to evaluate the actual state of DTPP, and pressure pulsations will have distinct influence on the operation stability when E_DTPP increases obviously. The efforts of this study provide a novel and useful method for the comprehensive state evaluation of DTPP, and the E_DTPP could be used as an index for the stability of Francis turbine.