Eureka AIR delivers breakthrough ideas for toughest innovation challenges, trusted by R&D personnel around the world.

A Nox Emission Prediction Method for Thermal Power Plants Based on Generalized Cross-Entropy Autoencoder

An autoencoder and emission technology, which is applied in the field of NOx emission prediction of thermal power plants based on generalized mutual entropy autoencoder, can solve problems such as poor robustness of loss function, and achieve improved robustness and high practical engineering application value. , predict accurate and reliable results

Active Publication Date: 2022-06-21
TAIYUAN UNIV OF TECH
View PDF9 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The present invention aims at the problem that the stacked autoencoder does not obtain information related to the target value in the training phase, and only uses the feature representation in the last hidden layer for final prediction and the loss function based on mean square error under non-Gaussian noise is poor in robustness , proposed a generalized cross-entropy autoencoder based NO x Emissions Forecast Method

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
  • A Nox Emission Prediction Method for Thermal Power Plants Based on Generalized Cross-Entropy Autoencoder
  • A Nox Emission Prediction Method for Thermal Power Plants Based on Generalized Cross-Entropy Autoencoder
  • A Nox Emission Prediction Method for Thermal Power Plants Based on Generalized Cross-Entropy Autoencoder

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0026] The invention takes the real data collected from the PI real-time database system of a power station under the Guodian Group as the experimental object, and trains the gated stacking target correlation autoencoder based on generalized mutual entropy.

[0027] A NO in Thermal Power Plant Based on Generalized Interentropy Autoencoder xThe emission forecast method, the steps are as follows:

[0028] Step 1: According to the actual measurement point data of the power station, select the boiler load (x 1 ), flue gas oxygen content (x 2 ~x 6 ), primary wind speed (x 7 ~x 12 ), secondary air volume (x 13 ~x 39 ), exhaust gas temperature (x 40 ) and coal quality data (x 41 ~x 43 ), the amount of powder fed by the pulverizer (x 44 ~x 49 ), the opening parameter of the exhaust air baffle (x 50 ~x 57 ) a total of 57-dimensional parameters are used as the input of the model, while NO x The emissions are taken as the true output y of the model.

[0029] Step 2: Prepro...

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 is a thermal power plant NO based on a generalized cross-entropy autoencoder x Emissions prediction method, specifically using the gated stacked target correlation autoencoder based on generalized cross-entropy to predict the NO x emissions. The steps included in the method are as follows: collecting thermal power plant data, including NO x Emissions and related influencing factors, preprocessing all data, will affect NO x The relevant factors of emissions are used as input to perform nonlinear dimension reduction and feature extraction through the self-encoder, and the NO x Emission impact factor and NO x Model between emissions. NO x The emission prediction is accurate and reliable, and has high practical engineering application value.

Description

technical field [0001] The present invention relates to NO x Prediction method of emissions, specifically a generalized cross-entropy autoencoder based NO for thermal power plants x Emissions Prediction Methods. Background technique [0002] Facing the increasingly severe environmental problems in China, there is an urgent need for a method that can reduce the energy consumption and pollutant discharge of power plant boilers. Boiler combustion optimization technology has always been an effective means to improve boiler efficiency and reduce pollutant emissions. It is very important for the power station to establish an accurate combustion model of the power station boiler and optimize the input parameters of the model. However, how to establish a boiler combustion model in a power plant to achieve efficient, automatic, fast and accurate prediction of data is particularly important. [0003] Generally speaking, there are three kinds of soft-sensor modeling methods, namely...

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 Patents(China)
IPC IPC(8): G06Q10/04
CPCG06N3/084G06Q10/04G06Q50/06G06N3/044G06N3/045
Inventor 任密蜂齐慧月巩明月方茜茜马建飞
Owner TAIYUAN UNIV OF 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
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