Multi-objective Optimization of Coal-fired Boiler
Combustion Based on NSGA-II
Tingfang Yu, Hongzhen Zhu
School of Mechanical and Electronic Engineering, Nanchang University, Nanchang, Jiangxi Province, China;
Email: wtu_tingfy@163.com, zhangkeat2008@163.com
Chunhua Peng
School of Electrical & Electronics Engineering, East China Jiaotong University, Nanchang, Jiangxi Province, China
Email: chinapch@163.com
Abstract—NOx emission characteristics and overall heat loss
model for a 300MW coal-fired boiler were established by
Back Propagation (BP) neural network, by which the the
functional relationship between outputs (NOx emissions &
overall heat loss of the boiler) and inputs (operational
parameters of the boiler) of a coal-fired boiler can be
predicted. A number of field test data from a full-scale
operating 300MWe boiler were used to train and verify the
BP model. The NOx emissions & heat loss predicted by the
BP neural network model showed good agreement with the
measured. Then, BP model and the non-dominated sorting
genetic algorithm II (NSGA-II) were combined to gain the
optimal operating parameters which lead to lower NOx
emissions and overall heat loss boiler. The optimization
results showed that hybrid algorithm by combining BP
neural network with NSGA-II can be a good tool to solve the
problem of multi-objective optimization of a coal-fired
combustion, which can reduce NOx emissions and overall
heat loss effectively for the coal-fired boiler.
Index Terms—Coal-fired boiler combustion, Multi-objective
optimization, BP neural network, NSGA-II
I. INTRODUCTION
In China, the requirements for environmental
protection are increasingly strict, especially for coal-fired
utility boiler. Coal remains the primary energy resource
in China, and one of the major concerns associated with
coal-fired power plants is the emission of pollutants,
especially for NO
2
and NO (collectively referred to as
NOx). Today, NOx emission is regulated and has become
an important consideration in the design and modification
of coal-fired utility boiler
[1-2]. However, many old-
designed utility boilers in China emit the NOx pollutants
above the limit and have posed terrible threat to the
surrounding environment, coal-fired power plants face
important challenges concerning the methods and
technologies to meet these new environmental
requirements. In addition to the developments in the plant
construction and flue gas cleaners such as Selective
Catalytic Reduction (SCR) reactor, NOx control
techniques based on combustion modification are of
considerable interest[3-5], because they avoid or
postpone large capital expenditures while meeting
environmental compliance requirements compared with
the relatively expensive flue gas NOx reduction
technologies. The control of the boiler operating
conditions through combustion optimization is an
important and cost-effective way to affect NOx emissions
[3-5].
In recently years, many scholars applied artificial
intelligent methods to optimization of coal-fired boiler
combustion, H. Zhou [6] established the boiler fly ash
carbon content model to predict the relationship between
unburned carbon in fly ash and operation parameters of
boiler by using BP neural network , Fuzzy neural-
network was proposed to model NOx emissions by
Ikonen [7]. P. H. Wang [8] established artificial neural-
networks model of a boiler NOx emissions and efficiency,
based on the above model, genetic algorithm (GA) was
introduced to get the optimal operation parameters of a
coal-fired boiler. C. XU [9] set up the boilers efficiency
and NOx emissions model by minimal resource allocating
networks (MRAN), simulation studies on global
optimization on efficiency and low NOx emissions object
were carried out based on MRAN and GA method. A
sharing LSSVM model was established by F. Gao [10] to
predict the NOx emission and thermal efficiency of a
1000MW boiler, and then adopted an improved PSO
algorithm to optimize the operational conditions of the
boiler. C. L. Wang [11] using support vector machine
(SVM) combined with the ant colony algorithm to
optimize the NOx emissions for a coal-fired boiler.
However, coal-fired boiler is a very complex system,
the level of unburned carbon in fly ash is an important
factor affecting the efficiency of pulverized coal fired
boiler, and especially those equipped with low NOx
burners. Due to the reduced mixing intensity and the
formation of fuel rich zones under low NOx combustion
conditions, the residence time of the coal particles in the
oxygen rich environment decreases, resulting in an
increase of the amount of unburned carbon in fly ash.
Pollution formation and carbon burnout in pulverized
coal combustion are dominated by the fuel properties
(reactivity, volatiles, nitrogen content, etc), fuel
preparation (coal fineness) and combustion conditions
(mixing). The emission of NOx from boiler can be
JOURNAL OF NETWORKS, VOL. 8, NO. 6, JUNE 2013
doi:10.4304/jnw.8.6.1300-1306