Hybrid deep learning-based online identification method for key parameters of gas turbine dynamic process

Shoutai SUN; Yali XUE; Mingchun WANG; Li SUN

doi:10.19693/j.issn.1673-3185.02914

Chinese Journal of Ship Research, Volume. 18, Issue 3, 222(2023)

Hybrid deep learning-based online identification method for key parameters of gas turbine dynamic process

Shoutai SUN¹, Yali XUE², Mingchun WANG¹, and Li SUN¹

¹Key Laboratory of Energy Thermal Conversion and Control of Ministry of Education, School of Energy and Environment, Southeast University, Nanjing 210018, China

²State Key Laboratory of Electric Power Systems, Department of Energy and Power Engineering, Tsinghua University, Beijing 100084, China

show less

Abstract Get PDF(in Chinese)

Figures & Tables(11)

Fig. 1. [in Chinese]

Fig. 2. [in Chinese]

Fig. 3. [in Chinese]

Fig. 4. [in Chinese]

Fig. 5. [in Chinese]

Fig. 6. [in Chinese]

Fig. 7. [in Chinese]

Table 1. Nomenclature and meaning of subscripts for gas turbine model

View table

View in Article

Table 1. Nomenclature and meaning of subscripts for gas turbine model

符号	含义及单位	下标	含义
P	压力/kPa	0	标称值
V	室容积/m3	1	压气机进口
T	温度/K	2	燃烧室进口
R	气体常数/(J·kg−1·K−1)	3	透平进口
$ \pi $	压比	4	透平出口
$ \sigma $	总压保持系数	c	压气机
L	扭矩/(N·m)	t	透平
I	轴惯性/(kg·m2)	v	定容
$ \gamma $	绝热指数	f	燃料
C	比热容/(kJ·kg−1·K−1)	I	进气管
n	转速/(r·min−1)	N	喷嘴，管口
$ \rho $	密度/(kg·m−3)	p	定压
$ \eta $	效率/%	comb	燃烧室
v	流量/(kg·s−1)		电力矩
$Q$	燃料热值/(J·kg−1)	l	低位

Table 2. Gas turbine specifications
View table
View in Article
Table 2. Gas turbine specifications
参数数值
额定功率/kW 100
额定转速/（r·min⁻¹） 15000

Table 3. Partial parameters of gas turbine model

View table

View in Article

Table 3. Partial parameters of gas turbine model

参数	数值	参数	数值	参数	数值
${\eta _{\rm{t}}}$/%	0.85677	$ {T_0} $/K	288.15	$ {P_0} $/kPa	101.325
${\sigma _{\rm{I}}}$	0.98	${\eta _{\rm{c}}}$/%	0.675	${C_{\rm{p} } }$/(kJ·kg−1·K−1)	1004
${C_{\rm{v} } }$/(kJ·kg−1·K−1)	717	${\sigma _{\rm{N}}}$	0.96	I/(kg·m2)	30
$ \gamma $	1.4	R/(J·kg−1·K−1)	287	$Q_{\rm{l} }$/(J·kg−1)	$4.28 \times {10^7}$
${\sigma _{{\rm{comb}}} }$	0.95

Table 4. Error analysis of LSTM- and LSTM-GPR-based identification results

View table

View in Article

Table 4. Error analysis of LSTM- and LSTM-GPR-based identification results

动态运行参数	整体平均误差/%		收敛后误差/%		真实值在辨识结果95%置信区间内的比例/%
动态运行参数	LSTM	LSTM-GPR	LSTM	LSTM-GPR	LSTM-GPR
燃料的低位热值$Q_{\rm{l} }$	1.060	0.070	1.830	0.002	85.6
压气机效率${\eta _{\rm{c}}}$	1.020	0.872	1.270	0.428	99.0
电力矩${L_{\rm{e}}}$	14.910	1.530	11.480	0.120	91.2

Tools

Get Citation

Copy Citation Text

Shoutai SUN, Yali XUE, Mingchun WANG, Li SUN. Hybrid deep learning-based online identification method for key parameters of gas turbine dynamic process[J]. Chinese Journal of Ship Research, 2023, 18(3): 222

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites