基于机理修正的小型燃气轮机变工况神经网络模型构建方法研究

叶旖茵; 韩巍; 李济超; 马文静; 韩泽鹏; 宋新阳

doi:10.19805/j.cnki.jcspe.2025.240030

欢迎来到能源电力期刊网！

English Version 读者登录期刊登录

您当前的位置：

首页 >

文章列表页 >

基于机理修正的小型燃气轮机变工况神经网络模型构建方法研究

动力设备与系统 | 更新时间：2026-01-27

- 基于机理修正的小型燃气轮机变工况神经网络模型构建方法研究
- Research on Construction Method of Neural Network Model for Small Gas Turbine Under Variable Working Conditions Based on Mechanism Modification
- 动力工程学报 2025年45卷第4期页码：505-513
- 作者机构：
  
  1. 中国科学院大学,北京,100190
  2. 中国科学院工程热物理研究所,北京,100190
  3. 华北电力大学能源动力与机械工程学院,北京,102206
- 作者简介：
  
  [ "叶旖茵(1998—),女,广西柳州人,硕士研究生,研究方向为综合能源系统" ]
  [ "韩巍(通信作者),男,研究员,博士,E-mail:hanwei@iet.cn" ]
- 基金信息：
  
  国家自然科学基金资助项目(U22A6007)
- DOI：10.19805/j.cnki.jcspe.2025.240030
  中图分类号： TK471;TP183
- 网络出版：2025-04-28，
  
  纸质出版：2025
- 稿件说明：
移动端阅览
叶旖茵,韩巍,李济超,马文静,韩泽鹏,宋新阳. 基于机理修正的小型燃气轮机变工况神经网络模型构建方法研究动力工程学报, 2025, 45(4): 505-513 https://doi.

org/10.19805/j.cnki.jcspe.2025.240030
叶旖茵,韩巍,李济超,马文静,韩泽鹏,宋新阳. 基于机理修正的小型燃气轮机变工况神经网络模型构建方法研究动力工程学报, 2025, 45(4): 505-513 https://doi. DOI： 10.19805/j.cnki.jcspe.2025.240030.

org/10.19805/j.cnki.jcspe.2025.240030 DOI：

摘要

燃气轮机变工况热力性能具有较强的非线性特点

基于物理模型的方法计算耗时长且计算误差较大

而基于数据驱动的方法对样本数据数量依赖强且计算过程中没有物理机理上的约束。为了克服这两种方法存在的问题

根据小型燃气轮机压缩和膨胀过程的物理机理

提炼出变工况运行的内外部影响参数与发电效率、排烟温度及排烟流量之间存在的约束关系

并结合神经网络基本框架

提出了基于机理修正的小型燃气轮机变工况神经网络模型(即改进模型)。具体算例的结果表明:利用已知的燃气轮机部件性能和部分数据样本

即可建立高精度的燃气轮机变工况神经网络模型;改进模型对关键参数的预测相对误差在0.941%以下

预测平均相对误差为0.583%

其预测平均相对误差为使用相同样本的基于纯数据驱动的神经网络预测误差的70.122%

为基于物理模型的方法计算误差的14.916%。

Abstract

The thermal performance of gas turbines under variable working conditions exhibits strong non-linear characteristics. Calculations based on physical models are time-consuming and have larger calculation errors

while data-driven methods depend heavily on the quantity of sample data and lack constraints from physical mechanisms during the calculation process. To overcome the shortcomings of both approaches

the constraints among the internal and external influencing parameters

power generation efficiency

exhaust temperature and exhaust flow rate under variable working conditions were extracted from the physical mechanism of the compression and expansion processes of small gas turbines. Combined with the basic framework of neural network

a neural network model for small gas turbines operated under variable working conditions was proposed based on mechanism modification. The results of specific calculations indicate that high-precision neural network model of gas turbines under variable working conditions can be established using known gas turbine component performance and some data samples. The relative error is less than 0.941% in the prediction of key parameters by the improved model. The average relative error of prediction is 0.583%

which is 70.122% of the prediction error of the purely data-driven neural network using the same samples

and 14.916% of the calculation error of the method based on the physical model.

关键词

Keywords

references

韩俊涛, 韩恺, 王永真, 等. 低碳分布式综合能源系统的能值、经济和环境优化评价[J]. 动力工程学报, 2022, 42(11): 1089-1098. HAN Juntao, HAN Kai, WANG Yongzhen, et al. Emergy, economic and environmental optimization evaluation of low carbon distributed integrated energy system[J]. Journal of Chinese Society of Power Engineering, 2022, 42(11): 1089-1098.

ZHANG Na, CAI Ruixian. Analytical solutions and typical characteristics of part-load performances of single shaft gas turbine and its cogeneration[J]. Energy Conversion and Management, 2002, 43(9/10/11/12): 1323-1337.

LI Zhouzheng, MIAO Dongyan, FENG Kun. Determining dynamic thresholds for gas turbine engine condition monitoring[J]. IEEE Access, 2022, 10: 87404-87414.

WU Biying. Dynamic performance simulation analysis method of split shaft gas turbine based on RBF neural network[J]. Energy Reports, 2021, 7(Sup7): 947-958.

黄玉柱, 蒋洪德. 一种重型燃气轮机建模及其参数估计的方法[J]. 动力工程学报, 2016, 36(8): 600-607. HUANG Yuzhu, JIANG Hongde. A modeling method for heavy duty gas turbines and the parameters estimation[J]. Journal of Chinese Society of Power Engineering, 2016, 36(8): 600-607.

董威, 尹家录, 郑培英, 等. 航空发动机及燃气轮机整机性能仿真综述[J]. 航空发动机, 2023, 49(5): 8-21. DONG Wei, YIN Jialu, ZHENG Peiying, et al. Review: engine-level performance simulation of aeroengine and gas turbines[J]. Aeroengine, 2023, 49(5): 8-21.

MALINOWSKI L, LEWANDOWSKA M. Analytical model-based energy and exergy analysis of a gas microturbine at part-load operation[J]. Applied Thermal Engineering, 2013, 57(1/2): 125-132.

KANG D W, KIM T S. Model-based performance diagnostics of heavy-duty gas turbines using compressor map adaptation[J]. Applied Energy, 2018, 212: 1345-1359.

张兆宇, 刘尚明. 基于数据的燃气轮机建模与控制技术概述[J]. 热力透平, 2019, 48(2): 89-95. ZHANG Zhaoyu, LIU Shangming. A review of gas turbine modeling and control technology based on operational data[J]. Thermal Turbine, 2019, 48(2): 89-95.

王文阳. 基于神经网络的燃气轮机燃烧稳定性预测及分析[J]. 动力工程学报, 2023, 43(7): 842-849. WANG Wenyang. Prediction and analysis of combustion stability of gas turbine based on neural network[J]. Journal of Chinese Society of Power Engineering, 2023, 43(7): 842-849.

郑培英, 刘家兴, 钟易成. 基于流体网络拓扑的航空发动机及燃气轮机整机性能通用仿真[J]. 航空发动机, 2023, 49(5): 1-7. ZHENG Peiying, LIU Jiaxing, ZHONG Yicheng. General simulation technology for aeroengine and gas turbine performance based on fluid network topology[J]. Aeroengine, 2023, 49(5): 1-7.

杨钧, 秦朝葵. 基于Modelica的天然气分布式能源系统仿真评估[J]. 城市燃气, 2019(9): 12-17. YANG Jun, QIN Chaokui. The simulation and evaluation of natural gas distributed energy system based on Modelica[J]. ChengShi RanQi, 2019(9): 12-17.

陈家伦, 卞韶帅, 黄新. 基于BP神经网络的9F燃气轮机压气机离线水洗周期优化[J]. 燃气轮机技术, 2020, 33(1): 47-53. CHEN Jialun, BIAN Shaoshuai, HUANG Xin. Optimization of offline washing cycle of 9F gas turbine compressor based on BP neural network[J]. Gas Turbine Technology, 2020, 33(1): 47-53.

刘尚明, 何皑, 郭森闯. 基于神经网络的燃气轮机控制和建模技术综述[J]. 热力透平, 2022, 51(3): 155-160, 169. LIU Shangming, HE Ai, GUO Senchuang. Review of gas turbine control and modeling technology based on neural network[J]. Thermal Turbine, 2022, 51(3): 155-160, 169.

李景轩, 周登极, 肖旺, 等. 燃气轮机机理-数据混合建模方法研究[J]. 热能动力工程, 2019, 34(12): 33-39. LI Jingxuan, ZHOU Dengji, XIAO Wang, et al. Hybrid modeling of gas turbine based on neural network[J]. Journal of Engineering for Thermal Energy and Power, 2019, 34(12): 33-39.

ZHANG Jincheng, ZHAO Xiaowei. Spatiotemporal wind field prediction based on physics-informed deep learning and LIDAR measurements[J]. Applied Energy, 2021, 288: 116641.

傅瑞华, 李凡, 王俊峰. 基于损失函数的单元测试用例自动化生成算法研究与实现[J]. 四川大学学报(自然科学版), 2022, 59(3): 032002. FU Ruihua, LI Fan, WANG Junfeng. Research and implementation of an algorithm for automatic generation of unit test cases based on loss function[J]. Journal of Sichuan University (Natural Science Edition), 2022, 59(3): 032002.

PARAND A, SERAJI M, DASHTI H. A modified multi-level cross-entropy algorithm for optimization of problems with discrete variables[J]. Engineering with Computers, 2022, 38(3): 2683-2698.

MA Jun, CHEN Jianan, NG M, et al. Loss Odyssey in medical image segmentation[J]. Medical Image Analysis, 2021, 71: 102035.

KRISTIANI E, TSAN Y T, LIU Poyu, et al. Binary and multi-class assessment of face mask classification on edge AI using CNN and transfer learning[J]. Human-centric Computing and Information Sciences, 2022, 12: 16.

顾云鹏, 马超, 臧绍飞, 等. 融合部分卷积和ECA机制的轴承滚子外观缺陷U-Net分割模型[J/OL]. 轴承, [2023-11-17].http://kns.cnki.net/kcms/detail/41.1148.th.20231116.1132.002.html. GU Yunpeng, MA Chao, ZANG Shaofei, et al. A U-Net model for segmentation of bearing roller appearance defects incorporating partial convolution and ECA mechanisms[J]. Bearing, [2023-11-17]. http://kns.cnki.net/kcms/detail/41.1148.th.20231116.1132.002.html.

刘金品, 王秉中, 陈传升, 等. 基于深度物理启发神经网络的微波波导器件逆设计方法[J]. 物理学报, 2023, 72(8): 080201. LIU Jinpin, WANG Bingzhong, CHEN Chuansheng, et al. Inverse design of microwave waveguide devices based on deep physics-informed neural networks[J]. Acta Physica Sinica, 2023, 72(8): 080201.

LEI Xingyu, YANG Zhifang, YU Juan, et al. Data-driven optimal power flow: a physics-informed machine learning approach[J]. IEEE Transactions on Power Systems, 2021, 36(1): 346-354.

ZOBEIRY N, HUMFELD K D. A physics-informed machine learning approach for solving heat transfer equation in advanced manufacturing and engineering applications[J]. Engineering Applications of Artificial Intelligence, 2021, 101: 104232.

ISHITSUKA K, LIN Weiren. Physics-informed neural network for inverse modeling of natural-state geothermal systems[J]. Applied Energy, 2023, 337: 120855.

杨珂, 王鑫, 凌佳杰, 等. 基于物理信息神经网络的同步发电机建模[J]. 中国电机工程学报, 2024, 44(12): 4924-4932. YANG Ke, WANG Xin, LING Jiajie, et al. Modeling of synchronous generator based on physics-informed neural network[J]. Proceedings of the CSEE, 2024, 44(12): 4924-4932.

CHEN Yang, XU Yongfu, WANG Lei, et al. Modeling water flow in unsaturated soils through physics-informed neural network with principled loss function[J]. Computers and Geotechnics, 2023, 161: 105546.

JIANG Zongxi, ZHANG Luliang, JI Tianyao. NSDAR: a neural network-based model for similar day screening and electric load forecasting[J]. Applied Energy, 2023, 349: 121647.

王志涛, 李健, 李淑英, 等. 基于改进牛顿法的燃-燃联合动力装置建模方法研究[J]. 热能动力工程, 2021, 36(10): 204-211. WANG Zhitao, LI Jian, LI Shuying, et al. Research on modeling method of the combined gas turbine and gas turbine device based on improved Newton-Raphson algorithm[J]. Journal of Engineering for Thermal Energy and Power, 2021, 36(10): 204-211.

闫斌斌, 冯坤, 江志农. 基于改进广义牛顿法的燃气轮机性能数字孪生体构建方法[J]. 热能动力工程, 2022, 37(1): 110-116. YAN Binbin, FENG Kun, JIANG Zhinong. Construction method of gas turbine performance digital twin based on enhanced generalized Newton method[J]. Journal of Engineering for Thermal Energy and Power, 2022, 37(1): 110-116.

浏览量

123

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

燃气轮机-超临界CO₂联合循环布局优化与动态特性研究

基于改进LSTM与注意力机制的SCR入口NO_x生成浓度预测

基于IGRA和CNN-LSTM的垃圾焚烧炉主蒸汽温度预测

不同波段下燃气轮机燃烧室辐射测温误差分析

Prototype of multi-filament electron curtain accelerator