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

Numerical simulation-based neural network hearth oxygen concentration prediction system and method

A numerical simulation and neural network technology, applied in the field of neural network applications, can solve problems such as insufficient measurement accuracy and low efficiency, and achieve the effects of processing a large amount of information, improving calculation efficiency, and improving combustion efficiency

Pending Publication Date: 2021-01-01
HARBIN BOILER
View PDF0 Cites 6 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] In order to solve the problem of insufficient measurement accuracy and low efficiency of the traditional oxygen concentration measurement method, the present invention proposes a numerical simulation-based neural network prediction furnace oxygen concentration system and method, the specific scheme is as follows:

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
  • Numerical simulation-based neural network hearth oxygen concentration prediction system and method
  • Numerical simulation-based neural network hearth oxygen concentration prediction system and method
  • Numerical simulation-based neural network hearth oxygen concentration prediction system and method

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0044]Specific embodiment one: the neural network prediction furnace oxygen concentration system based on numerical simulation, the system includes a numerical simulation simulation module, a data processing module, an algorithm prediction module and an implementation module, and the above-mentioned modules are connected by a progressive logical relationship, wherein the numerical simulation The simulation module is responsible for establishing the physical model inside the furnace, the data processing module is used to process the numerical simulation results, the algorithm prediction module is responsible for building the double-layer BP neural network model, the MLP neural network model and the DBN neural network model, and the realization module is responsible for selecting the best algorithm In order to realize the prediction of oxygen distribution.

specific Embodiment approach 2

[0045] Specific implementation mode two: the method of predicting the oxygen concentration in the furnace based on the neural network of numerical simulation, for the offline data collected, the physical model of the furnace is established with numerical simulation software, and its grid is divided and the boundary is set in ICEM; using ANSYS FLUENT solves and simulates the divided models to obtain data results; uses data processing methods to process numerical simulation results; uses neural network learning machines to make predictions, and finally realizes the prediction of oxygen distribution inside the furnace, providing theory and practice for the actual operation of boilers basis, including the following steps:

[0046] 1) Numerical simulation: use numerical simulation software to establish a physical model inside the furnace, and perform simulation operations.

[0047] 2) Data processing: processing numerical simulation results.

[0048] 3) Algorithm prediction: use t...

specific Embodiment approach 3

[0067] Specific implementation mode three: In addition to the neural network learning machine described in the first implementation mode, DBN can also be used to predict the oxygen content. The deep belief network (DBN) is composed of multi-layer RBM superposition and the last layer of regression neural network. The data is input through the bottom layer of the model, and passes through the RBM to the hidden layer. The output of the low-level RBM is used as the input of the high-level RBM, such as figure 2 Shown is a DBN model composed of two layers of RBM. The DBN training method firstly uses a bottom-up unsupervised learning method to initialize the parameters of the entire DBN model layer by layer, and then adopts a top-down supervised learning method to fine-tune the network parameters.

[0068] Write a program to integrate the three algorithms of BP neural network, multi-layer perceptron machine MLP, and deep belief neural network DBN into a neural network learning machi...

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 relates to a numerical simulation-based neural network hearth oxygen concentration prediction system and method, in particular to a neural network learning machine-based hearth oxygen concentration prediction system and method, belongs to the field of neural network application, and aims to solve the problems of insufficient measurement precision and low efficiency of a traditional oxygen concentration measurement method. The system comprises a numerical simulation module, a data processing module, an algorithm prediction module and an implementation module. The numerical simulation module is used for establishing a physical model in the hearth by utilizing numerical simulation software, and carrying out simulation operation; the data processing module is used for processinga numerical simulation result, then the algorithm prediction module builds three neural network modules and carries out algorithm prediction, and finally the optimal algorithm is selected through theimplementation module, so that prediction of oxygen amount distribution in a hearth is achieved. The method is large in processing information amount and high in calculation speed, the complex hearthoxygen amount problem is processed in a simple, convenient and rapid mode, the future trend can be accurately predicted, and the calculation efficiency is improved.

Description

technical field [0001] The invention is a numerical simulation-based neural network prediction furnace oxygen concentration system and method, in particular relates to a neural network learning machine to predict furnace oxygen concentration, and belongs to the neural network application field. Background technique [0002] Coal-fired boilers produce CO in motion 2 , SO 2 When the flue gas is not fully burned, a small amount of CO, CH4, etc. will also be produced. If the boiler air distribution is improper, the furnace leaks air and the flue gas contains oxygen, then during the combustion process of the furnace, it will fully contact with the oxygen entering the furnace to achieve full combustion, so it is very necessary to maintain an appropriate excess air to ensure the full combustion of pulverized coal in the furnace . But once the amount of air supplied to combustion increases, there will be some O 2 It is not utilized and is discharged from the chimney as hot flue ...

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): G06F30/28G06N3/04G06N3/08G06F111/10
CPCG06F30/28G06N3/084G06F2111/10G06N3/045
Inventor 杜宪涛崔宇佳朱慧娟夏良伟黄莺马孝纯沈涛姜文婷张超孙晶
Owner HARBIN BOILER
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