Unlock instant, AI-driven research and patent intelligence for your innovation.

Wind power generating unit drive system fault diagnosis method

A transmission system and fault diagnosis technology, which is applied in the monitoring of wind turbines, machines/engines, and wind turbines, etc., can solve problems such as feature conflicts in feature sets, mismatches between feature fusion and pattern recognition methods, and redundancy, so as to eliminate conflicts. , to achieve the effect of low-dimensional representation

Inactive Publication Date: 2019-03-26
CSIC CHONGQING HAIZHUANG WINDPOWER EQUIP
View PDF7 Cites 6 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

It can solve the problem of feature conflict and redundancy in multi-domain feature set for fault diagnosis of wind turbine transmission system, and also solve the problem of mismatch between feature fusion and pattern recognition methods, thereby improving the reliability and diagnostic rate of wind turbine transmission system fault diagnosis

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Wind power generating unit drive system fault diagnosis method
  • Wind power generating unit drive system fault diagnosis method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0031] A method for fault diagnosis of a transmission system of a wind turbine, the steps of which are:

[0032] A. Extract the time domain, frequency domain, and time-frequency domain information of the vibration signal to construct a multi-domain feature set; extract the time domain, frequency domain, and time-frequency domain features of the rotating machinery vibration signal to construct a multi-domain feature set to analyze the state characteristics of the rotating machinery Comprehensive reflection; the time-domain characteristics of the vibration signal refer to extracting 16 time-domain characteristics that reflect the vibration signal amplitude, energy size and amplitude distribution from the perspective of time-domain feature statistics; the frequency-domain characteristics of the vibration signal It refers to extracting 14 frequency domain features that reflect spectral energy distribution and central frequency band position changes from the perspective of frequency...

Embodiment 2

[0037] like figure 1 As shown, the wind turbine transmission system fault diagnosis method based on deep belief network feature fusion refers to: First, extract features such as time domain, frequency domain, and frequency domain to construct a multi-domain feature set to fully reflect the state characteristics of rotating machinery from different angles ;Secondly, use the deep belief network (Deep Belief Network, DBN) to re-learn the features of the multi-domain feature set, fully mine the essential characteristics of the data, realize feature fusion and eliminate redundant and conflicting information; finally add softmax after the feature output Multi-classifiers, using Back Propagation (BP) algorithm to fine-tune the weights of the deep belief network layer by layer to make the structure optimal and generate appropriate classifiers, so as to realize mechanical fault diagnosis. Diagnosis process such as figure 1 shown.

[0038] (1) Extract the time domain, frequency domain...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a wind power generating unit drive system fault diagnosis method, and belongs to the technical field of electrical equipment. The wind power generating unit drive system faultdiagnosis method comprises the steps of A, extracting time domain information, frequency domain information and time-frequency domain information of a vibration signal, and building a multi-domain feature set; B, adopting a deep belief network for carrying out feature relearning on the multi-domain feature set; C, adding softmax multiple classifiers after outputting features, and utilizing a backpropagation algorithm for layer-by-layer fine-adjusting a weight of the deep belief network so as to achieve an optimal structure, and generating an appropriate classifier; and D, outputting a diagnosis result. According to the wind power generating unit drive system fault diagnosis method, the problems of feature interaction and redundancy in a wind power generating unit drive system fault diagnosis multi-domain feature set can be solved, meanwhile, the problem that feature fusion is unmatched with a pattern recognition method is solved, and the reliability and the diagnostic rate of wind power generating unit drive system fault diagnosis are further improved.

Description

technical field [0001] The invention relates to the technical field of electric power equipment, in particular to a fault diagnosis method for a transmission system of a wind turbine. Background technique [0002] As the core component of the wind turbine, the transmission system's operation reliability directly affects the stability of the unit, so it is of great significance to carry out fault diagnosis on it. At present, for the fault diagnosis of the transmission system, different signal processing techniques are usually used to extract features to construct a fault feature set, and then different pattern recognition algorithms are used to identify and realize automatic diagnosis. In terms of feature extraction, Chen et al. extracted multi-dimensional time-frequency domain features, and performed dimension reduction and feature fusion through linear local tangent space arrangement algorithm; Qi Junde et al. performed wavelet packet decomposition on the signal, and extrac...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): F03D17/00
CPCF03D17/00Y02B10/30
Inventor 母芝验陈薛梅秦鑫张迁聂思宇韩花丽蔡梅园
Owner CSIC CHONGQING HAIZHUANG WINDPOWER EQUIP