Vibration fault detection system and method for wind turbine generator units

A technology for wind turbines and fault detection, applied in measuring devices, measuring ultrasonic/sonic/infrasonic waves, instruments, etc., can solve the problems of long time consumption and low monitoring efficiency, and achieve the effect of saving time, obvious advantages and improving efficiency

A technology for wind turbines and fault detection, applied in measuring devices, measuring ultrasonic/sonic/infrasonic waves, instruments, etc., can solve the problems of long time consumption and low monitoring efficiency, and achieve the effect of saving time, obvious advantages and improving efficiency

CN104048749AInactive Publication Date: 2014-09-17NORTHEAST DIANLI UNIVERSITY
2 Cites 16 Cited by

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Vibration fault detection system and method for wind turbine generator units
  • Vibration fault detection system and method for wind turbine generator units
  • Vibration fault detection system and method for wind turbine generator units

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0073] First, use the existing 10 sets of fault data and 5 sets of normal data for normalization processing and training. The training data are shown in Table 1, Table 2, and Table 3.

[0074] Secondly, measure the Mahalanobis distance between 1000 sets of normalized normal data and [0]17 matrix as Figure 5 , measure the normal range Mahalanobis distance threshold n=1.9130, n1.9130 is faulty data. Through Mahalanobis distance calculation, a large number of normal data can be excluded, and only a small amount of fault data can be classified.

[0075] Then, the 6 groups of data to be tested are normalized as shown in Table 4, and the affinity is used for fault classification and judgment, including two groups of fault 1 data, two groups of fault 2 data, and two groups of normal data. The judgment results are as follows: Image 6 .

[0076] According to the failure antibody affinity as Figure 7 , the affinity is in the range of [0,0.4] and [0.6,0.7], the fault points are obv...

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 vibration fault detection system and method for wind turbine generator units. Vibration data, including vibration data of front and rear bearings of a gearbox and vibration data of front and rear bearings of a generator, of all parts of the wind turbine generator units are collected through vibration collection nodes by setting up a wireless sensor network, routing nodes passing through all the units are transmitted to a network coordinator node in a multi-hopping mode and transmitted to a monitoring center in the distance through 3G, optical fibers and an etheric multifunctional network, and a negative selection algorithm based on mahalanobis distance is adopted for the monitoring center to achieve remote detection early warning of fault units and fault parts. The vibration fault detection system for the wind turbine generator units comprises the monitoring center, a detection terminal, a database server, the multifunctional network, the network coordinator node, the routing nodes and the vibration collection nodes. According to the system and method, efficiency of the algorithm is improved, the accuracy rate of fault diagnosis is improved, a large amount of time is saved, and the larger the data size is, the more apparent superiority of the algorithm is.

Description

technical field [0001] The invention belongs to the technical field of fault detection methods and equipment, and in particular relates to a vibration fault detection system and method for wind turbines. Background technique [0002] The vibration failure of wind turbines not only reduces the power generation efficiency but also affects its own safe operation, and may even endanger personal safety and cause property losses. Therefore, configuring a real-time wind turbine vibration fault detection and early warning system is an important subject of continuous research and development by researchers. [0003] At present, the research techniques in our country are roughly divided into process parameter estimation method, state estimation method, signal diagnosis method and artificial intelligence method. The process parameter estimation method uses the residual calculated by the mathematical model to judge and diagnose; the state estimation method uses the filter to monitor th...

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
17 Sep 2014
Publication
CN104048749A
IPC
G01H17/00
Inventors
邬春明; 银海燕