LSTM deep learning model-based hydropower unit fault diagnosis method and system

A technology for hydroelectric generating units and fault diagnosis, applied in neural learning methods, motor generator testing, biological neural network models, etc. The effect of reduced computation, high efficiency and accuracy

Active Publication Date: 2018-06-22
HUAZHONG UNIV OF SCI & TECH
View PDF3 Cites 66 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, due to the relatively shallow model structure of these traditional data-based fault diagnosis methods, they face problems such as the curse of dimensionality and limited learning ability for complex nonlinear objects a

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
  • LSTM deep learning model-based hydropower unit fault diagnosis method and system
  • LSTM deep learning model-based hydropower unit fault diagnosis method and system
  • LSTM deep learning model-based hydropower unit fault diagnosis method and system

Examples

Experimental program
Comparison scheme
Effect test

Example Embodiment

[0071] Example 1:

[0072] The embodiment of the present invention proposes a method for diagnosing faults of a hydropower unit based on an LSTM deep learning model. Refer to figure 1 As shown in the diagnostic model diagram, the specific implementation steps are as follows:

[0073] Step 1: Obtain the sampled values ​​of the vibration signals of the N different signal channels of the hydropower unit. Variational modal decomposition is performed on the vibration signal of each signal channel, and the decomposed K IMF components are obtained.

[0074] Among them, the premise of VMD decomposition is to construct a variational problem. Assuming that each ‘mode’ is a finite bandwidth with a center frequency, the variational problem can be described as seeking K IMF components u k (t), minimize the sum of the estimated bandwidth of each mode, and the constraint condition is that the sum of each mode is the original input signal. The construction process of the variational problem is as f...

Example Embodiment

[0133] Example 2:

[0134] The embodiment of the present invention also provides a fault diagnosis system of a hydropower unit based on the LSTM deep learning model, such as Figure 4 As shown, the system includes:

[0135] The IMF component acquisition unit is used to acquire the sampling sequences of N different signal channels of the hydropower unit, and perform variational modal decomposition on each time sequence to obtain K IMF components;

[0136] The training set and the set to be diagnosed acquisition unit are used to normalize each IMF component and construct the corresponding training set and set to be diagnosed;

[0137] The feature extraction unit is used to construct a long and short-term memory network model for the training set of each IMF component, and perform feature extraction for each eigenmode function through at least two LSTM layers;

[0138] The training module is used to connect the K LSTM layer outputs of the same signal channel to a Dense layer, and then co...

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 an LSTM deep learning model-based hydropower unit fault diagnosis method and system. The method comprises the steps of obtaining sampling sequences of N different signal channels of a hydropower unit, and performing VMD on each time sequence to obtain K IMF components; constructing corresponding training sets and to-be-diagnosed sets; building an LSTM model for the trainingset of each IMF component, and performing feature extraction on each IMF component through two LSTM layers; connecting outputs of K LSTM layers of the same signal channel to a Dense layer; through aSoftmax layer, performing feature classification on outputs of multiple Dense layers; and training a deep learning neural network model through an RMSProp gradient descent algorithm, and diagnosing the to-be-diagnosed sets by the trained model. According to the method and the system, the relatively good signal-noise separation effect of the VMD is combined with the processing advantage of an LSTMto the time sequences, so that the hydropower unit fault diagnosis accuracy is effectively improved.

Description

[0001] The invention belongs to the technical field of fault diagnosis of hydroelectric units, and more specifically relates to a method and system for fault diagnosis of hydroelectric units based on an LSTM deep learning model. Background technique [0002] For a long time, the regular planned maintenance system has played an important role in the normal operation of hydropower stations, but this maintenance system cannot meet the growing demand for equipment maintenance, and the cost of maintenance is still very expensive. Driven by social needs and science and technology, the rapid development of fault diagnosis technology and systems provides the possibility for real-time monitoring and maintenance of hydroelectric units based on their operating status. very necessary. [0003] With the development of machine learning, pattern recognition, signal processing, artificial intelligence, etc., diagnostic methods based on data analysis and processing have begun to develop. This ...

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
IPC IPC(8): G06K9/62G06N3/04G06N3/08G01R31/34
CPCG06N3/084G01R31/34G06N3/048G06F18/2415G06F18/214
Inventor 李超顺王若恒涂文奇陈昊陈新彪
Owner HUAZHONG UNIV OF SCI & TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products