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

CNN-LSTM fan fault prediction method and CNN-LSTM fan fault prediction system based on attention mechanism

A technology of fault prediction and attention, applied in neural learning methods, computer components, instruments, etc., can solve problems such as inability to mine time series features between data well, poor performance in dealing with time series problems, and poor generalization ability

Pending Publication Date: 2021-04-09
BEIJING HUADIAN TIANREN ELECTRIC POWER CONTROL TECH
View PDF4 Cites 15 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The data generated by wind turbines is time-series data that is continuously growing on the time axis. Therefore, learning the time-series relationship of wind turbine data is very important for wind turbine fault prediction; however, the current wind turbine fault prediction model has poor performance in dealing with timing problems and cannot perform well Mining the potential time series features between data leads to poor generalization ability. Therefore, a model is needed that can effectively process time series data with strong generalization ability and high accuracy

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
  • CNN-LSTM fan fault prediction method and CNN-LSTM fan fault prediction system based on attention mechanism
  • CNN-LSTM fan fault prediction method and CNN-LSTM fan fault prediction system based on attention mechanism
  • CNN-LSTM fan fault prediction method and CNN-LSTM fan fault prediction system based on attention mechanism

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0091] Example 1: A CNN-LSTM Fan Fault Prediction Method Based on Attention Mechanism

[0092] Such as image 3 As shown, Embodiment 1 of the present invention provides a CNN-LSTM fan failure prediction method based on attention mechanism, comprising the following steps:

[0093] Step 1. Collect fan timing data from the SCADA (Supervisory Control And Data Acquisition) system, including: its own performance data and external environment data, where its own performance data includes pitch angle, pitch speed, generator Current, power generation, etc.; external environment data include wind speed, temperature, humidity, etc.

[0094] It can be understood that those skilled in the art can arbitrarily configure the selected fan real-time data type, a preferred but non-limiting implementation is, in the embodiment of the present invention, in order to predict the icing failure of the fan, from the SCADA system The collected 370,000 pieces of data focus on wind speed, generator spee...

Embodiment 2

[0182] Example 2: CNN-LSTM Fan Fault Prediction System Based on Attention Mechanism

[0183] Such as figure 1 As shown, embodiment 2 of the present invention provides a kind of CNN-LSTM fan failure prediction system based on attention mechanism, including: input module, CNN neural network module, LSTM neural network module, attention mechanism module (Attention part) and output module.

[0184] Among them, the input module is used to input the data set filtered by the features of the random forest algorithm.

[0185] The CNN neural network module is used to compress the original features of the input data step by step, and finally obtain a high-level feature representation. The CNN neural network module includes:

[0186] The first part: the convolution layer, which consists of several convolution kernels. The weight of each convolution kernel is obtained through data-driven learning. These convolution kernels compress the original features of the input data step by step, a...

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 CNNLSTM fan fault prediction method based on an attention mechanism, and the method comprises the steps: firstly collecting the time sequence data of a fan through an SCADA system; secondly, a random forest algorithm is used for carrying out feature correlation degree analysis on the fan time series data, features with high fan fault correlation degree are extracted, a new feature data set is formed, the new feature data set is input into a CNNLSTM model to be trained, and an Attention mechanism is introduced into the model, so that the model pays more attention to the features related to fan faults; interference of non-related features is reduced, so that the accuracy and generalization ability of the fan fault prediction model are improved; and finally, inputting a to-be-predicted fan data set into the trained model to obtain a fault prediction result. The model provided by the invention is a fan fault prediction model which is suitable for processing time series data, high in accuracy, strong in generalization ability and outstanding in comprehensive performance.

Description

technical field [0001] The invention relates to the technical field of wind power failure prediction, in particular to a CNN-LSTM fan failure prediction method and system based on an attention mechanism. Background technique [0002] Wind energy is a pollution-free renewable energy with almost endless energy and wide distribution. Wind power eases the supply of energy and is of great significance to the protection of the environment. Wind power equipment is generally installed in areas with rich wind resources but inconvenient transportation, such as mountainous areas, plateau areas, grassland pastures, etc. The environment in these areas is harsh, and lightning, rainstorms or typhoons may cause failures of wind turbines, and regular inspections of wind turbines are required. To ensure the normal operation of wind power equipment, but the wind turbines are located in remote areas, the inspection workload is heavy, which increases the workload of maintenance personnel; for s...

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): G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06N3/044G06N3/045G06F18/2113G06F18/24323
Inventor 赵计生米路中强保华谢元曹亚伟张艳萍陈锐东
Owner BEIJING HUADIAN TIANREN ELECTRIC POWER CONTROL TECH
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

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com