Method and system for abnormal detection of wind turbine gearbox oil temperature based on scada data
A technology for gearbox oil and wind turbines, applied in geometric CAD and other directions, can solve the problems of estimated value deviation, reduce the reliability of abnormal oil temperature detection results, and cannot obtain the error range of gearbox oil temperature, and achieve online detection of abnormal oil temperature. , Realize the effect of offline abnormal oil temperature detection, effectiveness and practicability verification
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0040] In one or more embodiments, a method for abnormal detection of gearbox oil temperature based on SCADA data is disclosed, comprising the following steps:
[0041]Step (1): Data selection. According to the object to be detected, the appropriate monitoring quantity is selected as the input and output of the model according to the professional background knowledge and the correlation analysis method of the data.
[0042] Among them, the data correlation analysis method is Pearson correlation analysis method, and its calculation formula is:
[0043]
[0044] Among them, cov(X,Y) is the covariance of vector X,Y, σ X ,σ Y are the standard deviations of X and Y, respectively.
[0045] Step (2): Data preprocessing. According to the corresponding monitoring data selected by the method of step (1), the historical normal operation data is selected as the training data and the test data, and the data of the period to be detected is selected as the data to be tested. And prep...
Embodiment 2
[0118] In one or more embodiments, a wind turbine gearbox oil temperature anomaly detection system based on SCADA data is disclosed, including:
[0119] A module for selecting the data related to the oil temperature of the gearbox from the numerous state parameters monitored by SCADA as the input quantity of the model;
[0120] A module for filtering and normalizing the data; a module for selecting historical normal operation data as training data and test data, and selecting data in a time period to be tested as data to be tested;
[0121] A module for constructing a probability estimation model of gearbox oil temperature based on sparse Bayesian learning theory, and a module for setting coverage indicators and bandwidth indicators as model evaluation indicators;
[0122] A module used to optimize the bandwidth of the kernel function in the probability estimation model of gearbox oil temperature with the goal of minimizing the bandwidth index under the condition of guaranteed...
Embodiment 3
[0126] In one or more embodiments, a terminal device is disclosed, including a server, the server includes a memory, a processor, and a computer program stored on the memory and operable on the processor, and the processor executes the The program implements the SCADA data-based abnormal detection method for the oil temperature of the wind turbine gearbox in the first embodiment. For the sake of brevity, details are not repeated here.
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


