第
21
卷第
3
期
2010
年
9
月
中国计量学院学报
10umal
of
China
University
of
Metrology
【文章编号】
1004-1540(2010)03-0246-05
模拟退火算法在线热源反问题
数值求解中的应用
郑朋周杭霞於可广
2
O.
中国计量学院信息工程学院，浙江杭州
310018;
2.
浙江水利水电专科学校现代教育技术中心，浙江杭州
310018)
Vo
l.
21
NO.3
Sep. 2010
【摘
要】
提出采用模拟退火算法
(simulated
anneali
吨.
SA)
来数值求解线热源反问题.探讨了如何设计算
法使之适合反问题求解，并给出了算法求解的伪代码;通过线源正问题的模拟数据，使用设计的
SA
算法进行
反问题求解，以此来验证算法求解的准确性和可靠性，并对→组实测数据进行了计算.结果表明，该算法不但
可以实现两个参数同时、快速反演，而且具有求解精度高，对初始条件依赖少，编制容易等优点.
【关键词】
模拟退火算法;线热源反问题;数值求解;参数反演
【中图分类号】
TP18
【文献标识码】
A
Application
of
simulated
annealing
in
solving
the
inverse
problem
of
line
heat
sources
ZHENG
Peng
1
,
ZHOU
Hang-xia
1
,
YU
Ke-guang
2
(1.
College of Information Engineering. China J iliang Univcrsity. Hangzholl :110018. China;
2.
Modern
Ed
ucational
τèchnology
Centre. Zhejiang Water Conservancy
and
Hydropower College. Hangzhou 310018. China)
Abstract:
1n
order
to
solve
the
problcm
of
line
heat
sourc
巳
parameter
compuling
which
mainly
depends
on
approximate
formula
and
the
result
compared
10
the
lrue
value
b
巳
l
口 g
poor.
we
used
the
simulated
annealing
algorithm
to
solve
the
inverse
problem
of
heat
line
sources.
First.
the
simulated
annealing(
SA)
algorithm
was
briefly
introduced;
and
then
the
SA
algorithm
was
discussed
according
to
the
characteristic
of
the
inverse
problem
of
the
heat
line
source.
The
pseudo
code
of
SA
was
then
also
given
to
show
the
detailed
process
of
design
finally.
1n
order
to
v 巳
rify
the
accuracy
and
reliability
of
the
algorithm.
we
simulated
a
serial
data
based
on
the
hcat
linc
source
equation
to
solve
thc
inverse
problem
of
the
heat
line
source.
Also
a
set
of
measurcd
data
were
calculated.
Thc
results
show
that
the
algorithms
can
not
only
rapidly
inverse
two
parameters
at
lhe
same
time.
but
31so wi
th
high
accuracy.
lc
目
dependence
on
initial
conditions
and
casy
implem
巳
nta
11011.
Key
words:
simulated
annealing
(SA);
inverse
prohlem
of
heat
line
source;
numerical
solution;
parameler
ll1
VerSl0n
E
收稿日期】
2010-06-06
【基金项目】
国家
863
计划项目
(No.2008
八
A09Z303)
.浙江省水利厅计划专项
(No.
2008C21 099)
【作者简介】
郑
朋
(198;'
-
).男.i'
1
苏
ifl
，安人.硕
1:
研究
'1
主
~j
研究方向为智能'):):法优化和
l
向性能
ii
如
资源评论
