A Surrogate Model for a CAES Radial Inflow Turbine with Test Data-Based MLP Neural Network Algorithm

WANG Xing, ZHU Yangli, LI Wen, ZUO Zhitao, CHEN Haisheng

热科学学报 ›› 2023, Vol. 32 ›› Issue (6) : 2081-2092.

PDF(4074 KB)
PDF(4074 KB)
热科学学报 ›› 2023, Vol. 32 ›› Issue (6) : 2081-2092. DOI: 10.1007/s11630-023-1846-3  CSTR: 32141.14.JTS-023-1846-3

A Surrogate Model for a CAES Radial Inflow Turbine with Test Data-Based MLP Neural Network Algorithm

  • WANG Xing1, ZHU Yangli2,3, LI Wen2,3, ZUO Zhitao1,2,3, CHEN Haisheng1,2,3*
作者信息 +

A Surrogate Model for a CAES Radial Inflow Turbine with Test Data-Based MLP Neural Network Algorithm

  • WANG Xing1, ZHU Yangli2,3, LI Wen2,3, ZUO Zhitao1,2,3, CHEN Haisheng1,2,3*
Author information +
文章历史 +

摘要

现有研究常对向心涡轮进行全尺寸三维流动分析以找到压缩空气储能(CAES)系统效率提升方法。然而,求解时间长、计算资源消耗大成为开展该类分析的主要阻碍。因此,本研究提出了一种基于测试数据的多层感知器(MLP)神经网络代理模型来克服这一困难。该模型不是采用流场求解过程,而是通过“学习测量结果”在较短求解时间内提供可靠的涡轮气动性能和流场分布特性。研究结果表明,该模型对等熵效率、折合流量和折合功率预测的最大相对误差分别为0.03%、0.22%和0.26%。集气室、轮盖空腔和叶轮出口区的流动参数分布也与实验结果基本一致。在进气室中,当总压比降低时,在275°的圆周位置处观察到压力驻点;在轮盖空腔中,尽管存在迷宫式密封,但在导流罩出口附近发现了明显的压力变化。在转子出口处,0.0~0.4和0.6~0.8相对叶高范围内有明显的速度和压力变化。同时,在叶片高度0.0~0.4和0.6~0.8范围内,速度和压力变化明显,这是由于上通道涡流、下通道涡流和端壁二次流的影响。本研究可为CAES径向流入式水轮机的动态性能评估提供进一步的参考。

Abstract

It is usually to conduct a full-scale three-dimensional flow analysis for a radial turbine to find a way to increase the efficiency of a Compressed Air Energy Storage (CAES) system. However, long solving time and huge consumption of computing resources become a major obstacle to the analysis. Therefore, in present study, a surrogate model with test data-based multi-layer perceptron (MLP) Neural Network is proposed to overcome the difficulty. Instead of complex flow field solving process, it provides reliable turbine aerodynamic performance and flow field distribution characteristics in a short solution time by “learning the measurement results”. The validation results illustrated that the predicted maximum relative errors of isentropic efficiency, corrected mass flow rate and corrected power are only 0.03%, 0.22% and 0.26% respectively. The predicted flow distribution parameters in chamber, shroud cavity and outlet region of rotor are also basically consistent with the experimental results. In the chamber, it can be found that a pressure stagnation point is observed at circumferential angle of 270° when total pressure ratio is decreased. In the shroud cavity, obvious pressure variation is found near outlet of shroud cavity which although labyrinth seals exist. At outlet of rotor, obvious variations of velocity and pressure are found in the 0.0–0.4 and 0.6–0.8 of blade height. At the same time, obvious variations of velocity and pressure are found in the 0.0–0.4 and 0.6–0.8 of blade height and this is because the influence of upper passage vortex, lower passage vortex and end wall secondary flow. The present study can provide further reference for the dynamic performance evaluation of CAES radial inflow turbine.

关键词

CAES / surrogate model / radial inflow turbine / MLP neural network

Key words

CAES / surrogate model / radial inflow turbine / MLP neural network

引用本文

导出引用
WANG Xing, ZHU Yangli, LI Wen, ZUO Zhitao, CHEN Haisheng. A Surrogate Model for a CAES Radial Inflow Turbine with Test Data-Based MLP Neural Network Algorithm[J]. 热科学学报, 2023, 32(6): 2081-2092 https://doi.org/10.1007/s11630-023-1846-3
A Surrogate Model for a CAES Radial Inflow Turbine with Test Data-Based MLP Neural Network Algorithm[J]. Journal of Thermal Science, 2023, 32(6): 2081-2092 https://doi.org/10.1007/s11630-023-1846-3

参考文献

[1] Guo H., Xu Y.J., Chen H.S., Guo C., Qin W., Thermodynamic analytical solution and exergy analysis for supercritical compressed air energy storage system. Applied Energy, 2017, 199: 96–106.
[2] Al Juboria A.M., Jawadb Q.A., Investigation on performance improvement of small scale compressed-air energy storage system based on efficient radial-inflow expander configuration. Energy Conversion and Management, 2019, 182: 224–239.
[3] Shao Z., Li W., Wang X., Zhang X., Chen H., Analysis of shroud cavity leakage in a radial turbine for optimal operation in compressed air energy storage system. ASME Journal of Engineering for Gas Turbines and Power, 2020, 142(7): 071005.
[4] Wang X., Zhu Y., Li W., Zhang X., Chen H., Flow characteristics of an axial turbine with chamber and diffuser adopted in compressed air energy storage system. Energy Report, 2020, 6(7): 45–57.
[5] Wang X., Li W., Zhang X., et al., Flow characteristic of a multistage radial turbine for supercritical compressed air energy storage system. Proceedings of the Institution of Mechanical Engineers, Part A: Journal of Power and Energy, 2018, 232(6): 622–640.
[6] Wang X., Li W., Zhang X., Zhu Y., Zuo Z., Chen H., Efficiency improvement of a CAES low aspect ratio radial inflow turbine by NACA blade profile. Renewable Energy, 2019, 138: 1214–1231.
[7] Wang X., Li W., Zhang X., Zhu Y., Chen H., Flow analysis and performance improvement of a radial inflow turbine with back cavity under variable operation condition of compressed air energy storage. International Journal of Energy Research, 2019, 43(12): 6396–6408.
[8] Li W., Wang X., Zhang X., Zhang X., Zhu Y., Chen H., Experimental and numerical investigation of closed radial inflow turbine with labyrinth seals. ASME Journal of Engineering for Gas Turbines and Power, 2018, 140(10): 102502.
[9] Cheng Z., Tong S., Tong Z., Bi-directional nozzle control of multistage radial-inflow turbine for optima part-load operation of compressed air energy storage. Energy Conversion and Management, 2019, 181: 485–500.
[10] Guo Q., Li S., Song X., Liu W., Three-dimensional transient simulation method for transonic compressor based on dynamic boundaries. Journal of Propulsion Technology, 2019, 40(6): 1231–1238.
[11] Copeland C., Newton P., Martinez-Botas R., et al., A comparison of pulsed flow timescales within a turbine stage. IMECHE Turbochargers and Turbocharging, London, 2012, 4: 15–16.
[12] Cerdoun M, Ghenaiet A., Unsteady behavior of a twin entry radial turbine under engine like inlet flow conditions. Applied Thermal Engineering, 2018, 130(5): 93–111.
[13] Lee S.P., Jupp M.L., Barrans S.M., Nickson A.K., Analysis of leading edge flow characteristics in a mixed flow turbine under pulsating flows. Proceedings of the Institution of Mechanical Engineers, Part A: Journal of Power and Energy, 2019, 233(1): 78–95.
[14] Roclawski H., Gugau M., Bohle M., Computational fluid dynamics analysis of a radial turbine during load step operation of an automotive turbocharger. ASME Journal of Fluids Engineering, 2018, 140(2): 021102.
[15] Zhu Z., Guo X., Cui B., Li Y., External characteristics and internal flow features of a centrifugal pump during rapid startup period. Chinese Journal of Mechanical Engineering, 2011, 24(5): 798–804.
[16] Li Z., Bi H., Wang Z., Yao Z., Three-dimensional simulation of unsteady flows in a pump-turbine during start-up transient up to speed no-load condition in generating mode. Proceedings of the Institution of Mechanical Engineers, Part A: Journal of Power and Energy, 2016, 230(6): 570–585.
[17] Xiao Y., Xiao R., Transient simulation of a pump-turbine with misaligned guide vanes during turbine model start-up. Acta Mechanica Sinica, 2014, 30: 646–655.
[18] Nicolle J., Morissette J., Giroux A., Transient CFD simulation of a Francis turbine startup. IOP Conference Series: Earth and Environmental Science, 2012, 15(6): 062014.
[19] Sun K., Li Y., Zhao L., et al., Transient starting performance analysis of vertical axis tidal turbine. Journal of Huazhong University of Science and Technology (Natural Science Edition), 2017, 45(04): 51–56.
[20] Liu Y., Nie G.K., Li Q.F., Han W., Tan H.Y., Inner flow field analysis on the process of synchronous initiation of pump turbine. Journal of Gansu Science, 2016, 28(04): 95–99.
[21] Li H., Li W., Liu D., Zhang X., Chen H., Effect of adjustable guide vanes on the characteristic of multistage reheating radial inflow turbine. Chinese Journal of Electrical Engineering, 2016, 36(22): 6180–6186.
[22] Li H., Li W., Zhang X., Zhu Y., Chen H., Characteristic of a multistage reheating radial inflow in supercritical compressed air energy storage with variable operating parameters. Proceeding of Institution of Mechanical of Engineers Part A: Journal of Power and Energy, 2019, 233(3): 397–412.
[23] Liu D., Li W., Li H., Zhang X., Zhu Y., Chen H., Characteristic analysis of combined regulation of adjustable guide vanes of multistage radial inflow turbines. Energy Storage Science and Technology, 2017, 6(6): 1286–1294.
[24] Lucia D., Beran P.S., Silva W.A., Reduced-order modeling: new approaches for computational physics. Progress in Aerospace Sciences, 2004, 40(1–2): 51–117.
[25] Ding P., Wu X., He Y., Tao W., A fast and efficient method for predicting fluid flow and heat transfer problems. ASME Journal of Heat Transfer, 2008, 130(3): 1–17.
[26] Yang M., A reduced order model for aerodynamic response of blades excited by back flow. North University for China, Taiyuan, China, 2018.
[27] Kosowski K., Karol T., Kosowski A., Application of artificial neural networks in investigations of steam turbine cascades. ASME Journal of Turbomachinery, 2010, 132(1): 014501.
[28] Hui X., Bai J., Wang H., Zhang Y., Fast pressure distribution prediction of airfoils using deep learning. Aerospace Science and Technology, 2020, 105: 105949.
[29] Wang Q., Yang L., Rao Y., Establishment of a generalizable model on a small-scale dataset to predict the surface pressure distribution of gas turbine blades. Energy, 2021, 214: 118878.
[30] Li Z.B., Wen F.B., Tang X.L., Su L.J., Wang S.T., Prediction of single-row hole film cooling performance based on deep learning. Acta Aeronautica et Astronautica Sinica, 2021, 42(4): 124331.
[31] Dai W., Yang L., Rao Y., Modeling method of film cooling of turbine endwall based on generative adversarial networks. Journal of Engineering Thermophysics, 2020, 41(10): 2420–2424.
[32] Li D., Qiu L., Tao K., Zhu J., Artificial intelligence aided design of film cooling scheme on turbine guide vane. Propulsion and Power Research, 2020, 9(4): 344–354.
[33] Li Y., Wei D., Yu R., Chyu M.K., Optimization of the hole distribution of an effusively cooled surface facing non-uniform incoming temperature using deep learning. International Journal of Heat and Mass Transfer, 2019, 145: 118749.
[34] Li Y., Wei D., Yu R, Chyu M.K., A machine learning approach to quantify the film cooling superposition effect for effusion cooling structures. International Journal of Thermal Science, 2021, 162: 106774.
[35] Pedregosa F, Varoquaux G, Gramfort A Michel V., Thiron Bertrand., Grisel O., Blondel M, Müller A., Northman J., Louppe G., Prettenhofer P., Weiss R., Dubourg V., Vanderplas J., Scikit-learn: Machine Learning in Python. Journal of Machine Learning Research, 2011, 12: 2825–2830.
[36] Zhang N., Cai R., Analytical solutions and typical characteristics of part-load performances of single shaft gas turbine and its cogeneration. Energy Conversion and Management, 2002, 43: 1323–1337.

基金

The work is supported by Strategic Priority Research Program of the Chinses Academy of Sciences (51925604); National Natural Science Foundation of China (51806211); The Science and Technology Foundation of Guizhou Province (No. [2019]1285). The authors would like to thank the above organizations for their financial support.

版权

Science Press, Institute of Engineering Thermophysics, CAS and Springer-Verlag GmbH Germany, part of Springer Nature 2023
PDF(4074 KB)

56

Accesses

0

Citation

Detail

段落导航
相关文章

/