Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Steam turbine rotor fault diagnosis method based on LSTM

A steam turbine rotor and fault diagnosis technology, which is applied to mechanical equipment, engine functions, engine components, etc., can solve problems such as unfavorable industrial promotion, poor diagnostic accuracy, and low diagnostic efficiency

Active Publication Date: 2019-04-02
XI AN JIAOTONG UNIV
View PDF5 Cites 24 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to detect and identify steam turbine rotor faults and ensure the safe operation of steam turbine generator sets. A LSTM-based steam turbine rotor fault diagnosis method is provided to solve the need for steam turbine rotor diagnosis in traditional methods with the help of experience and extremely high signals. Handling skills lead to low diagnostic efficiency, poor diagnostic accuracy, and are not conducive to industrial promotion and other issues

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
  • Steam turbine rotor fault diagnosis method based on LSTM
  • Steam turbine rotor fault diagnosis method based on LSTM
  • Steam turbine rotor fault diagnosis method based on LSTM

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0161] Such as Figure 4As shown, according to the method of the present invention, the vibration signals of four typical faults of steam turbine rotors are firstly given. There are two different sets of simulated signals with a signal-to-noise ratio of -3. The sampling frequency of the simulation signal is 12000. The rotor speed is 3000rpm. The window length of the segmented signal is 2048. Each fault signal can be divided into a training set of 685 samples. One set of simulations is used as a training set after pre-processing, and the other set of simulation signals is used as a test set after pre-processing.

[0162] Build an LSTM neural network with a fully connected layer, set the initial learning rate to 0.001, reduce the learning rate by an order of magnitude every 30 steps, and finally set the learning rate to 0.00001.

[0163] Table 1 is the confusion matrix diagnosed by the model of the present invention on the test set. From the results of the confusion matrix,...

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 steam turbine rotor fault diagnosis method based on LSTM, and belongs to the technical field of mechanical fault diagnosis. Firstly, multi-point acquisition sensors are deployed and controlled, and vibration signals of various typical turbine rotor faults are collected as a training set and a verification set. Secondly, the steam turbine rotor vibration signals are extracted from a power plant DCS system to serve as a testing set. Thirdly, the training set, the testing set and the verification set realize fusion of multi-point signal data and data enhancement throughsignal division, stacking and other operations. Fourthly, a neural network based on the LSTM is constructed, the training set and the verification set are used for completing training of the network,and finally, maintenance of a diagnostic model is achieved in cooperation with an actual diagnostic task, and finally the steam turbine rotor fault diagnosis is realized on the testing set.

Description

technical field [0001] The invention belongs to the technical field of mechanical fault diagnosis, and in particular relates to a method for diagnosing a fault of a steam turbine rotor based on LSTM. Background technique [0002] Turbine generator set is the key equipment for electric power production. It has the characteristics of complex structure, harsh working conditions (high temperature, high pressure, high speed), high requirements for continuous operation, etc., and is prone to failure. In the operation of the unit, the rotor is an important component. Once a failure occurs and cannot be checked in time, it will cause unplanned shutdown due to the vibration exceeding the limit, and cause damage to the unit and casualties in severe cases. Therefore, the fault diagnosis method and experimental research of the turbogenerator rotor are of great significance for ensuring the safe operation of the turbogenerator, reducing major economic losses and avoiding catastrophic acc...

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): F01D21/00G06K9/62
CPCF01D21/00F01D21/003F05D2260/80G06F2218/12G06F18/214
Inventor 张荻王崇宇谢永慧刘天源
Owner XI AN JIAOTONG UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products