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

Improved data cleaning method for stack noise reduction auto-encoder

An auto-encoder and data cleaning technology, applied in the data cleaning field of stack noise reduction auto-encoder, can solve the problems of evaluating equipment operating conditions, destroying data continuity and integrity, ignoring correlations, etc.

Inactive Publication Date: 2019-07-05
NORTHEAST DIANLI UNIVERSITY
View PDF3 Cites 17 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

As the data set shows increasingly strong ability to express comprehensive characteristics of equipment, the current data cleaning method only considers the abnormal characteristics of local state quantities, ignores the correlation between the overall attributes, and destroys the continuity and integrity of the data. Facilitate subsequent analysis of data and evaluation of equipment operating conditions

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
  • Improved data cleaning method for stack noise reduction auto-encoder
  • Improved data cleaning method for stack noise reduction auto-encoder
  • Improved data cleaning method for stack noise reduction auto-encoder

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0104] Taking a 330MW boiler in a thermal power plant as an example, select the normal state data of the drum water level, steam pressure and temperature online monitoring data of 900 groups of boilers from June to August 2016 as training samples, and select the data from October 2016 Up to December, 900 sets of abnormal state data of the same state quantity were used as test samples, and the experimental results were compared with the actual operation of the boiler to verify the validity of the model.

[0105] Use the normal state data to train and construct the AS-SDAE model, and obtain the optimal network parameters of the model. The number of nodes in the input layer is 272, and there are 3 hidden layers. The number of nodes is set to 200, 100, and 2, and the number of training rounds is 2500. The noise ratio is 20%, and the learning rate is 0.01. Figure 4 Table 1 and Table 1 are the comparison chart and numerical statistical table of the convergence of SDAE (Adam), SDAE ...

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 improved data cleaning method for a stack noise reduction auto-encoder, which comprises the following steps of firstly, introducing an Adam and SGD hybrid algorithm to continuously adjust the network parameters of a stack noise reduction auto-encoder model; secondly, training the normal state data by using the model, obtaining the hidden characteristics of the data, andobtaining a reconstruction error in a normal state; thirdly, using the model for detecting the abnormal state data, analyzing the influence of various types of data on the model according to reconstruction errors of the abnormal state data, and rapidly classifying, cleaning and repairing the dirty data and the abnormal data reflecting the equipment faults. The AS-SDAE can directly and intelligently analyze the monitoring data, can better mine the high-order characteristics hidden in the data, guarantees the high efficiency of cleaning the dirty data, retains the useful data reflecting the abnormal condition of the equipment, and improves the data analysis efficiency.

Description

technical field [0001] The invention relates to the field of data processing, in particular to an improved data cleaning method of a stack noise reduction autoencoder. Background technique [0002] With the development of electric power enterprises, the demand for national electricity is increasing. As one of the main equipment for electric energy production, the boiler's condition monitoring data increases in series during the production process, which conforms to the characteristics of big data and contains rich potential. Resource value, mining and analysis of these data is helpful to understand the overall operation law of the power system. However, the operating conditions of the power plant are complex and diverse, and the boiler is subject to various external interferences during the actual operation process, such as the environment is changeable, the sensor is abnormal for a short time, and the data transmission chain is blocked, etc., resulting in the monitoring dat...

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): G06K9/62G06F16/215G06N3/04
CPCG06N3/045G06F18/241
Inventor 娄建楼李燕孙博曲朝阳王蕾郭晓利
Owner NORTHEAST DIANLI UNIVERSITY
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