Combustion Optimization Based on Computational Intelligence

verfasst von: Hao Zhou, Prof. Kefa Cen

Verlag: Springer Singapore

Buchreihe : Advanced Topics in Science and Technology in China

Enthalten in: Springer Professional "Wirtschaft+Technik" , Springer Professional "Technik" , Springer Professional "Wirtschaft"

Einloggen, um Zugang zu erhalten

Über dieses Buch

This book presents the latest findings on the subject of combustion optimization based on computational intelligence. It covers a broad range of topics, including the modeling of coal combustion characteristics based on artificial neural networks and support vector machines. It also describes the optimization of combustion parameters using genetic algorithms or ant colony algorithms, an online coal optimization system, etc. Accordingly, the book offers a unique guide for researchers in the areas of combustion optimization, NOx emission control, energy and power engineering, and chemical engineering.

Inhaltsverzeichnis

Frontmatter

Chapter 1. Introduction

Abstract

In this chapter, the background and the key problems are presented, like the emission of nitrogen oxides (NO_x) and the level of unburned carbon. Selective Catalytic Reduction (SCR) and Selective Non-Catalytic Reduction (SNCR) are conducted to reduce the NO_x emissions. Artificial intelligence methods are used to solve the complexity of boiler system. The characteristic of coal combustion and the parameter of unburned carbon content are discussed in this chapter. Later, coal combustion optimization is proposed. The outline of the book is recommended at last.

Hao Zhou, Kefa Cen

Chapter 2. The Influence of Combustion Parameters on NO x Emissions and Carbon Burnout

Abstract

In this chapter, the influence of combustion parameters on NO_x emissions and carbon burnout is discussed briefly. On the one hand, the effects of coal type, chemical equivalent and residence time, temperature, moisture and ash content, air dynamic field and flame species, particle size, boiler load, and OFA nozzle on NO_x emissions are investigated. On the other hand, the influence of combustion parameters, such as the operational conditions of the boiler and the coal properties, OFA nozzle, primary air, inner secondary air and outer secondary air and swirling intensity on carbon burnout are studied.

Hao Zhou, Kefa Cen

Chapter 3. Modeling Methods for Combustion Characteristics

Abstract

In this chapter, through the study of the complex combustion process, different influence factors are exposed. Experimental method and CFD methods are compared, respectively, with economic and accuracy. The former focuses on the detailed and accurate information about coal combustion, including ash content, fusion temperature, flame temperature, and flue gas. Many CFD methods, such as turbulence model and radiative heat transfer model, are introduced to understand their appropriate operating condition. Due to the convenience to analyze large and complex data, CFD methods are widely applied in combustion simulation. Computational intelligence method based on combustion studying is also proposed in this section.

Hao Zhou, Kefa Cen

Chapter 4. Neural Network Modeling of Combustion Characteristics

Abstract

In this chapter, artificial neural network (ANN) model is discussed with the basic idea and principle. It has been applied in many AI areas and industrial fields including pattern recognition and speech recognition. Then, two typical kinds of ANN algorithm (BPNN and GRNN) are introduced in this section. Through the comparative analysis of the principle and the conditions of the two methods of BPNN and GRNN, a new method combining the two methods is proposed.

Hao Zhou, Kefa Cen

Chapter 5. Classification of the Combustion Characteristics based on Support Vector Machine Modeling

Abstract

In this chapter, support vector machine (SVM) based on the structural risk minimization principle is exposed. As a vital computational method with outstanding generalization ability, it has been widely used in the field of machine learning and data mining. Support vector classification (SVC) and support vector regression (SVR) are the main parts of SVM [1]. The principle of them is introduced in this section. In the field of coal-fired power generation, there are lots of applications of SVC and SVR in the simulation. They are conducted to the modeling of coal identification, prediction of NO_x emission, prediction of unburned carbon in fly ash and others.

Hao Zhou, Kefa Cen

Chapter 6. Combining Neural Network or Support Vector Machine with Optimization Algorithms to Optimize the Combustion

Abstract

In this chapter, optimization algorithms of GA, ACO, and PSO are introduced. They are combined with SVM and ANN and show convenience in experiments. Also multi-objective optimization is introduced, like MOCell, AbYSS, OMOPSO, and SPEA2. They are compared in many aspects. Among them, OMOPSO and MOCell are the proposed algorithms for online multi-object optimization of coal-fired boilers.

Hao Zhou, Kefa Cen

Chapter 7. Online Combustion Optimization System

Abstract

In this chapter, the online combustion optimization system is exposed. First, the demand of the software system for the practical combustion optimization and the need of the local optimization are summarized, such as data detection requirements, quickness and accuracy requirements, requirements of different optimization goal, requirements online self-learning, parameter optimization limit requirements, fault tolerance requirements, alarm requirements, compatibility of off-line data processing and optimizing and so on. Second the instruments or sensors for online combustion optimization system are introduced. Then, the online SVM algorithm is presented, mainly about derivation of the Incremental Relations, AOSVR Bookkeeping Procedure, efficiently updating the R Matrix, initialization of the Incremental Algorithm and Decremental Algorithm. In addition, there are three main functions with different modules of online combustion optimization system. They are, respectively, online monitoring and alarm function, online optimization and self-learning function, off-line modeling, and optimization function. Finally the application of online combustion optimization system is discussed.

Hao Zhou, Kefa Cen

Chapter 8. Combustion Optimization Based on Computational Intelligence Applications: Future Prospect

Abstract

In this section, the prospect of combustion optimization based on computational intelligence is forecasted. Combustion optimization is considered to be very significant for both the power plant de-NO_x and carbon burnout of fly ash. Then, three unclear factors are discussed and prospected: coal characteristics for model, dealing with large data set, and feature selection. Combustion optimization based on computational intelligence is a very valuable and practical technology to approve the industrial change.

Hao Zhou, Kefa Cen

Backmatter

Titel: Combustion Optimization Based on Computational Intelligence
verfasst von: Hao Zhou
Prof. Kefa Cen
Verlag: Springer Singapore
Electronic ISBN: 978-981-10-7875-0
Print ISBN: 978-981-10-7873-6
DOI: https://doi.org/10.1007/978-981-10-7875-0