Process in the application of machine learning in ejector research

doi:10.16085/j.issn.1000-6613.2024-0375

Abstract

Abstract:

The ejector is a widely used mechanical device with advantages such as simple structure, low initial cost, easy maintenance, and reliable operation. It is widely applied in fields such as refrigeration, desalination, chemical engineering, fuel cells, aerospace, etc. The ejector does not directly consume mechanical energy, which allows for energy-saving purposes, making it more important and attractive with the national goals of“carbon peaking and carbon neutrality”in China. Machine learning methods, as a data-driven automated analysis approach, can be used for analyzing the internal flow characteristics of ejectors and optimizing ejector performance. In recent years, a small number of scholars have already applied machine learning methods to the study of ejectors in various applications, aiming at improving the ejector performance and the system performance. But the research in the open literature is currently scattered, and the state of the art is not yet clear. The present work comprehensively reviewed the literature on the application of machine learning methods in the study of ejectors for different applications, analyzed the current research status, summarized the machine learning methods utilized in the open literature, and pointed out that in the future machine learning methods can be applied to the study of internal flow characteristics of ejectors, providing a basis and guidance for improving the efficiency and performance of ejectors. Machine learning methods can be applied to the study of ejector performance under variable operating conditions, constructing a pathway from automated design to real-world application of ejectors. Constructing more suitable algorithms and proposing a series of targeted solutions.

Key words: ejector refrigeration, ejector, machine learning, artificial neural network, prediction, optimization

摘要：

喷射器是一种用途广泛的机械装置，具有结构简单、初投资低、维护容易、运行可靠等优点，广泛应用于制冷、海水淡化、化工、燃料电池、航空航天等领域。喷射器不直接消耗机械能，可实现节能的目的，在我国“双碳”背景下备受关注。机器学习方法作为一种基于数据的自动化分析方法，可以用于喷射器内部流动特性的分析及喷射器性能的优化。近年来，已有部分学者将机器学习方法应用于喷射器的研究中，以提升喷射器性能和系统性能。但是目前文献中的研究方向较为分散，研究现状和研究水平尚不明晰。本文对采用机器学习方法对喷射器进行研究的文献进行了全面的梳理，分析现状、总结方法，并指出未来可以将机器学习方法应用于喷射器内部流动特征的研究中，为提升喷射器效率和性能提供依据和指导；将机器学习方法应用于喷射器变工况性能研究中，构建喷射器从自动化设计到真实应用的通路；构建更加合适的算法，提出一系列具有针对性的解决方案。

关键词: 喷射制冷, 喷射器, 机器学习, 人工神经网络, 预测, 优化

CLC Number:

TB65

DAI Zhengshu, ZUO Yuanhao, CHEN Xiaoluo, ZHANG Li, ZHAO Gen, ZHANG Xuejun, ZHANG Hua. Process in the application of machine learning in ejector research[J]. Chemical Industry and Engineering Progress, 2024, 43(S1): 1-12.

戴征舒, 左元浩, 陈孝罗, 张犁, 赵根, 张学军, 张华. 机器学习在喷射器研究中的应用进展[J]. 化工进展, 2024, 43(S1): 1-12.

Figures/Tables 5

References 80

1	SADASIVAN Sreeja, ARUMUGAM Senthil Kumar, Jishnu CHANDRAN R, et al. Effects of influencing factors on the performance and morphology of shock waves in ejectors: A review[J]. International Journal of Modern Physics C, 2023, 34(11): 2350142.
2	OMIDVAR Amir, GHAZIKHANI Mohsen, RAZAVI Sayyed Mohammad Reza Modarres. Entropy analysis of a solar-driven variable geometry ejector using computational fluid dynamics[J]. Energy Conversion and Management, 2016, 119: 435-443.
3	BESAGNI Giorgio, MEREU Riccardo, INZOLI Fabio. Ejector refrigeration: A comprehensive review[J]. Renewable and Sustainable Energy Reviews, 2016, 53: 373-407.
4	ZHANG Guojie, Sławomir DYKAS, LI Pan, et al. Accurate condensing steam flow modeling in the ejector of the solar-driven refrigeration system[J]. Energy, 2020, 212: 118690.
5	LIU Bo, GUO Xiangji, XI Xiuzhi, et al. Thermodynamic analyses of ejector refrigeration cycle with zeotropic mixture[J]. Energy, 2023, 263: 125989.
6	DING Hongbing, ZHAO Yafei, WEN Chuang, et al. Energy efficiency and exergy destruction of supersonic steam ejector based on nonequilibrium condensation model[J]. Applied Thermal Engineering, 2021, 189: 116704.
7	ZHAO Li, YANG Xingyang, DENG Shuai, et al. Performance analysis of the ejector-expansion refrigeration cycle using zeotropic mixtures[J]. International Journal of Refrigeration, 2015, 57: 197-207.
8	YAN Gang, BAI Tao, YU Jianlin. Thermodynamic analysis on a modified ejector expansion refrigeration cycle with zeotropic mixture (R290/R600a) for freezers[J]. Energy, 2016, 95: 144-154.
9	ZHANG Guojie, WANG Xiaogang, POURRANJBAR Dariush, et al. The comprehensive analysis of the relationship between the latent heat, entrainment ratio, and ejector performance under different superheating degree conditions considering the non-equilibrium condensation[J]. Applied Thermal Engineering, 2022, 200: 117701.
10	TANG Yongzhi, LIU Zhongliang, LI Yanxia, et al. A combined pressure regulation technology with multi-optimization of the entrainment passage for performance improvement of the steam ejector in MED-TVC desalination system[J]. Energy, 2019, 175: 46-57.
11	GU Wending, WANG Xinli, WANG Lei, et al. Performance investigation of an auto-tuning area ratio ejector for MED-TVC desalination system[J]. Applied Thermal Engineering, 2019, 155: 470-479.
12	LIU Jiapeng, WANG Lei, JIA Lei, et al. Thermodynamic analysis of the steam ejector for desalination applications[J]. Applied Thermal Engineering, 2019, 159: 113883.
13	LI He, WANG Xiaodong, HUANG Hailong, et al. Numerical study on the effect of superheat on the steam ejector internal flow and entropy generation for MED-TVC desalination system[J]. Desalination, 2022, 537: 115874.
14	DING Hongbing, ZHAO Yafei, WEN Chuang, et al. A visual mass transfer study in the ejector considering phase change for multi-effect distillation with thermal vapour compression (MED-TVC) desalination system[J]. Desalination, 2022, 532: 115722.
15	WEN Chuang, DING Hongbing, YANG Yan. Performance of steam ejector with nonequilibrium condensation for multi-effect distillation with thermal vapour compression (MED-TVC) seawater desalination system[J]. Desalination, 2020, 489: 114531.
16	WEN Chuang, GONG Liang, DING Hongbing, et al. Steam ejector performance considering phase transition for multi-effect distillation with thermal vapour compression (MED-TVC) desalination system[J]. Applied Energy, 2020, 279: 115831.
17	CHEN Jikai, SUN Mingbo, LI Peibo, et al. Effects of excess oxidizer coefficient on RBCC engine performance in ejector mode: A theoretical investigation[J]. Energy, 2024, 289: 130070.
18	SONG Yajie, WANG Xinli, WANG Lei, et al. A twin-nozzle ejector for hydrogen recirculation in wide power operation of polymer electrolyte membrane fuel cell system[J]. Applied Energy, 2021, 300: 117442.
19	PEI Pucheng, REN Peng, LI Yuehua, et al. Numerical studies on wide-operating-range ejector based on anodic pressure drop characteristics in proton exchange membrane fuel cell system[J]. Applied Energy, 2019, 235: 729-738.
20	YU Zhongshuai, LIU Fang, LI Chengzhang. Numerical study on effects of hydrogen ejector on PEMFC performances[J]. Energy, 2023, 285: 129481.
21	ZHANG Hailun, SUN Wenxu, XUE Haoyuan, et al. Performance analysis and prediction of ejector based hydrogen recycle system under variable proton exchange membrane fuel cell working conditions[J]. Applied Thermal Engineering, 2021, 197: 117302.
22	DING Hongbing, DONG Yuanyuan, ZHANG Yu, et al. Energy efficiency assessment of hydrogen recirculation ejectors for proton exchange membrane fuel cell (PEMFC) system[J]. Applied Energy, 2023, 346: 121357.
23	YANG Daolong, XING Bangsheng, LI Jianping, et al. Experimental study on the injection performance of the gas-solid injector for large coal particles[J]. Powder Technology, 2020, 364: 879-888.
24	SHAHZAMANIAN B, VARGA S, SOARES J, et al. Performance evaluation of a variable geometry ejector applied in a multi-effect thermal vapor compression desalination system[J]. Applied Thermal Engineering, 2021, 195: 117177.
25	HANAFI A S, MOSTAFA G M, WAHEED A, et al. 1-D mathematical modeling and CFD investigation on supersonic steam ejector in MED-TVC[J]. Energy Procedia, 2015, 75: 3239-3252.
26	陆宏圻, 等. 喷射技术理论及应用[M]. 武汉: 武汉大学出版社, 2004.
	LU Hongqi, et al. Theory and application of jet technology[M]. Wuhan: Wuhan University Press, 2004.
27	邢日. 液气两相射流泵抽取不凝气3D仿真研究[D]. 济南: 山东大学, 2020.
	XING Ri. Three-dimensional simulation of liquid-gas jet pump for evaluating vacuum[D]. Jinan: Shandong University, 2020.
28	周志华. 机器学习[M]. 北京: 清华大学出版社, 2016.
	ZHOU Zhihua. Machine learning[M]. Beijing: Tsinghua University Press, 2016.
29	李胜. 机器学习应用研究[M]. 武汉: 武汉大学出版社, 2022.
	LI Sheng. Machine learning[M]. Wuhan: Wuhan University Press, 2022.
30	胡祝华, 赵瑶池. 计算机视觉与机器学习技术在智慧农业中的应用研究[M]. 哈尔滨: 哈尔滨工业大学出版社, 2019.
	HU Zhuhua, ZHAO Yaochi. Research on the application of computer vision and machine learning technology in smart agriculture[M]. Harbin: Harbin Institute of Technology Press, 2019.
31	郑娅峰. 人工智能视域下机器学习在教育研究中的应用[M]. 北京: 中国经济出版社, 2020.
	ZHENG Yafeng. Applying machine learning in educational research from the perspective of artificial intelligence[M]. Beijing: Economic Press China, 2020.
32	GUPTA Pradeep, RAO Srisha M V, KUMAR Pramod. Performance evaluation of air ejectors using artificial neural network approach[J]. Sādhanā, 2023, 48(2): 45.
33	YANG Yan, KARVOUNIS Nikolas, WALTHER Jens Honore, et al. Effect of area ratio of the primary nozzle on steam ejector performance considering nonequilibrium condensations[J]. Energy, 2021, 237: 121483.
34	ZHANG Guojie, ZHANG Xinzhe, WANG Dingbiao, et al. Performance evaluation and operation optimization of the steam ejector based on modified model[J]. Applied Thermal Engineering, 2019, 163: 114388.
35	YANG Yan, ZHU Xiaowei, YAN Yuying, et al. Performance of supersonic steam ejectors considering the nonequilibrium condensation phenomenon for efficient energy utilisation[J]. Applied Energy, 2019, 242: 157-167.
36	ZHANG Guojie, Sławomir DYKAS, YANG Sichong, et al. Optimization of the primary nozzle based on a modified condensation model in a steam ejector[J]. Applied Thermal Engineering, 2020, 171: 115090.
37	RINGSTAD Knut Emil, ALLOUCHE Yosr, GULLO Paride, et al. A detailed review on CO₂ two-phase ejector flow modeling[J]. Thermal Science and Engineering Progress, 2020, 20: 100647.
38	BENCHARIF Mehdi, CROQUER Sergio, FANG Yu, et al. Prediction of performance and geometrical parameters of single-phase ejectors using artificial neural networks[J]. Thermo, 2023, 3(1): 1-20.
39	BANASIAK Krzysztof, PALACZ Michał, HAFNER Armin, et al. A CFD-based investigation of the energy performance of two-phase R744 ejectors to recover the expansion work in refrigeration systems: An irreversibility analysis[J]. International Journal of Refrigeration, 2014, 40: 328-337.
40	黄亮亮. 基于人工神经网络的喷射器性能预测及优化研究[D]. 上海: 东华大学, 2010.
	HUANG Liangliang. The study of prediction and optimization of ejector performance based on artificial neural network[D]. Shanghai: Donghua University, 2010.
41	AHMED Fareeha, CHEN Weixiong. Investigation of steam ejector parameters under three optimization algorithm using ANN[J]. Applied Thermal Engineering, 2023, 225: 120205.
42	RINGSTAD Knut Emil, BANASIAK Krzysztof, Åsmund ERVIK, et al. Machine learning and CFD for mapping and optimization of CO₂ ejectors[J]. Applied Thermal Engineering, 2021, 199: 117604.
43	SHUKLA Prabhav B, MYDUR Manoj, NAYAK Subhash, et al. Design optimization of ejectors using ANN and GA[J]. Heat Transfer, 2022, 51(5): 4768-4782.
44	LIU Guangdi, ZHAO Hongxia, DENG Jianqiang, et al. Performance improvement of CO₂ two-phase ejector by combining CFD modeling, artificial neural network and genetic algorithm[J]. International Journal of Refrigeration, 2023, 154: 151-167.
45	刘光弟. 基于优化算法的跨临界CO₂两相喷射器结构优化和性能分析[D]. 济南: 山东大学, 2023.
	LIU Guangdi. Structure optimization and performance analysis of transcritical CO₂ two-phase ejector based on optimization algorithm[D]. Jinan: Shandong University, 2023.
46	ZHANG Yu, HE Chao, SUN Lei. Optimization of an ejector to mitigate cavitation phenomena with coupled CFD/BP neural network and particle swarm optimization algorithm[J]. Progress in Nuclear Energy, 2022, 153: 104412.
47	MAGHSOODI A, AFSHARI E, AHMADIKIA H. Optimization of geometric parameters for design a high-performance ejector in the proton exchange membrane fuel cell system using artificial neural network and genetic algorithm[J]. Applied Thermal Engineering, 2014, 71(1): 410-418.
48	Nayem JAHINGIR M, HUQUE Ziaul. Design optimization of rocket-based combined-cycle inlet/ejector system[J]. Journal of Propulsion and Power, 2005, 21(4): 650-655.
49	杨飞飞. 乳胶制品输送用环形喷射器的设计与优化[D]. 南京: 东南大学, 2017.
	YANG Feifei. Design and optimization of annular ejector for conveying latex products[D]. Nanjing: Southeast University, 2017.
50	高孝良. 回油喷射器的设计与优化[D]. 大连: 大连理工大学, 2009.
	GAO Xiaoliang. Design and optimization of an oil return ejector[D]. Dalian: Dalian University of Technology, 2009.
51	RASHIDI M M, AGHAGOLI A, RAOOFI R. Thermodynamic analysis of the ejector refrigeration cycle using the artificial neural network[J]. Energy, 2017, 129: 201-215.
52	SANGESARAKI Alireza Gholizadeh, GHAREHGHANI Ayat, MEHRENJANI Javad Rezazadeh. 4E analysis and machine learning optimization of a geothermal-based system integrated with ejector refrigeration cycle for efficient hydrogen production and liquefaction[J]. International Journal of Hydrogen Energy, 2023, 48(82): 31875-31904.
53	Amir EBRAHIMI-MOGHADAM, Mahmood FARZANEH-GORD, MOGHADAM Ali Jabari, et al. Design and multi-criteria optimisation of a trigeneration district energy system based on gas turbine, Kalina, and ejector cycles: Exergoeconomic and exergoenvironmental evaluation[J]. Energy Conversion and Management, 2021, 227: 113581.
54	Tao HAI, DHAHAD Hayder A, ATTIA El-Awady, et al. Proposal 3E analysis and multi-objective optimization of a new biomass-based energy system based on the organic cycle and ejector for the generation of sustainable power, heat, and cold[J]. Sustainable Energy Technologies and Assessments, 2022, 53: 102551.
55	SOLTANI Mohammad Mohsen, AHMADI Pouria, ASHJAEE Mehdi. Techno-economic optimization of a biomass gasification energy system with supercritical CO₂ cycle for hydrogen fuel and electricity production[J]. Fuel, 2023, 333: 126264.
56	HAO Xinyue, GAO Neng, WANG Xuehui, et al. Thermodynamic investigation on low GWP drop-in alternatives to HFC-245fa in ejector refrigeration cycle for air condition application[J]. Asia-Pacific Journal of Chemical Engineering, 2023, 18(5): e2948.
57	WANG Hao, CAI Wenjian, WANG Youyi. Modeling of a hybrid ejector air conditioning system using artificial neural networks[J]. Energy Conversion and Management, 2016, 127: 11-24.
58	SHEN Yanliang, HU Liangmou, LI Yonglin, et al. Modeling of aircraft fuel pressurization ejector system based on support vector regression[C]//2007 International Conference on Mechatronics and Automation. Harbin, China: IEEE, 2007: 2082-2086.
59	史棋棋. 利用神经网络的新型喷射-压缩制冷系统适用性研究[D]. 杭州: 浙江工业大学, 2021.
	SHI Qiqi. Research on the suitability of novel ejection-compression refrigeration system using neural network[D]. Hangzhou: Zhejiang University of Technology, 2021.
60	Adnan SÖZEN, Erol ARCAKLIOĞLU, Mehmet ÖZALP. A new approach to thermodynamic analysis of ejector-absorption cycle: Artificial neural networks[J]. Applied Thermal Engineering, 2003, 23(8): 937-952.
61	Adnan SÖZEN, Ali AKÇAYOL M. Modelling (using artificial neural-networks) the performance parameters of a solar-driven ejector-absorption cycle[J]. Applied Energy, 2004, 79(3): 309-325.
62	Adnan SÖZEN, Ali AKÇAYOL M. Reformulation, as a function of only working temperatures, of performance parameters of a solar driven ejector-absorption cycle using artificial neural networks[J]. Energy Sources, 2005, 27(12): 1133-1149.
63	Adnan SÖZEN, Erol ARCAKLIOĞLU, Mehmet ÖZALP. Performance analysis of ejector absorption heat pump using ozone safe fluid couple through artificial neural networks[J]. Energy Conversion and Management, 2004, 45(13/14): 2233-2253.
64	Adnan SÖZEN, Erol ARCAKLIOĞLU, Mehmet ÖZALP, et al. Performance parameters of an ejector-absorption heat transformer[J]. Applied Energy, 2005, 80(3): 273-289.
65	Adnan SÖZEN, Erol ARCAKLIOĞLU. Exergy analysis of an ejector-absorption heat transformer using artificial neural network approach[J]. Applied Thermal Engineering, 2007, 27(2/3): 481-491.
66	CHEN Haoran, CAI Wenjian. Artificial neural network modeling for variable area ratio ejector[C]//2014 9th IEEE Conference on Industrial Electronics and Applications. Hangzhou, China: IEEE, 2014: 220-225.
67	BENCHARIF Mehdi, CROQUER Sergio, PONCET Sébastien, et al. Prediction of performance for an ejector refrigeration cycle working with R245fa using artificial neural network[C]//Proceedings of the International Conference on Fluid Flow and Thermal Science (ICFFTS'20). Avestia Publishing, 2020: 1-6.
68	GUPTA Pradeep, KUMAR Pramod, RAO Srisha M V. Artificial neural network model for single-phase real gas ejectors[J]. Applied Thermal Engineering, 2022, 201: 117615.
69	ABBADY Karim, Nawaf AL-MUTAWA, ALMUTAIRI Abdulrahman. The performance analysis of a variable geometry ejector utilizing CFD and artificial neural network[J]. Energy Conversion and Management, 2023, 291: 117318.
70	GUPTA Pradeep, KUMAR Pramod, RAO Srisha M V. Artificial neural network based shape optimization of supersonic ejectors in the critical flow regime[J]. Applied Thermal Engineering, 2022, 216: 119046.
71	PETROVIC Andrija, JOVANOVIC Milos Z, GENIC Srbislav, et al. Evaluating performances of 1-D models to predict variable area supersonic gas ejector performances[J]. Energy, 2018, 163: 270-289.
72	黄亮亮, 曹家枞, 汤小亮. 基于SA-BP算法的小波神经网络在喷射器性能预测中应用[J]. 建筑热能通风空调, 2009, 28(6): 1-4.
	HUANG Liangliang, CAO Jiacong, TANG Xiaoliang. Application of wavelet neural networks based on SA-BP algorithm in forecasting the performance of ejectors[J]. Building Energy & Environment, 2009, 28(6): 1-4.
73	黄亮亮, 曹家枞. 基于连续优化蚁群算法的神经网络在喷射器性能预测中的应用[J]. 东华大学学报(自然科学版), 2010, 36(2): 185-190, 201.
	HUANG Liangliang, CAO Jiacong. Application of neural networks based on continuous-ACO algorithm in predicting the performance of ejectors[J]. Journal of Donghua University (Natural Science), 2010, 36(2): 185-190, 201.
74	ZHU Hanzeng, LIU Jiapeng, YU Jinpeng, et al. Artificial neural network-based predictive model for supersonic ejector in refrigeration system[J]. Case Studies in Thermal Engineering, 2023, 49: 103313.
75	ZHANG Kun, ZHANG Zhen, HAN Yuning, et al. Artificial neural network modeling for steam ejector design[J]. Applied Thermal Engineering, 2022, 204: 117939.
76	谷迎港. 非能动应急喷射冷却系统数值模拟设计[D]. 大连: 大连理工大学, 2021.
	GU Yinggang. Numerical simulation design of passive emergency injection cooling system[D]. Dalian: Dalian University of Technology, 2021.
77	XU Keda, CHEN Li, DONG Zuomin, et al. Data-driven prediction of ejector performance for PEMFC hydrogen fuel recirculation system[C]//2023 IEEE 6th International Conference on Big Data and Artificial Intelligence (BDAI). Jiaxing, China: IEEE, 2023: 35-39.
78	ANTÓNIO C C, CUNHA D A, VARGA S. A model for the optimized control of an innovative air conditioning system with variable geometry[C]//Proceedings of the 7th Luso-Mozambican Engineering Conference and the 4th Mozambique Engineering Conference. Porto: INEGI, 2014.
79	JEON Yongseok, LEE Dongchan, CHO Honghyun. Optimization of motive nozzle position in a modified two-phase ejector expansion household refrigeration cycle using an artificial neural network[J]. Energy Reports, 2022, 8: 1114-1123.
80	JAMALI Arash, AHMADI Pouria, JAAFAR Mohammad Nazri Mohd. Optimization of a novel carbon dioxide cogeneration system using artificial neural network and multi-objective genetic algorithm[J]. Applied Thermal Engineering, 2014, 64(1/2): 293-306.

作者	工质	模型/网络结构	训练算法/优化器	优化方法	数据集
Ahmed和Chen^[41]	水蒸气	4-102-2	Adam，跟随正则化领导者法（follow the regularized leader, Ftrl）自适应学习率法（adaptive delta，Adadelta）	—	实验216
Ringstad等^[42]	CO₂	GPR	—	—	模拟200
Shukla等^[43]	R141b	5-(5-5-5-5)-1	SCG	遗传算法	模拟1008
Liu等^[44-45]	CO₂	5-10-1	LM	遗传算法	模拟350
Zhang等^[46]	水蒸气	4-9-1	—	粒子群优化算法	模拟97
Maghsoodi等^[47]	工作流体为纯氢气，引射流体为氢气和水蒸气	4-4-1	—	遗传算法	模拟167
Jahingir和Huque^[48]	工作流体为氧气与氢气，引射流体为空气	—	—	满意度函数优化	模拟27
杨飞飞^[49]	工作流体为空气，引射流体为乳胶制品	4-(6-1)-1	动量梯度下降算法（gradient descent with momentum，GDM）	控制步长方法	模拟16
高孝良^[50]	一氯二氟甲烷（R22）	2-(10-20)-1	LM	控制步长方法	—

作者	工质	模型/网络结构	训练算法/优化器	优化方法	数据集
Ahmed和Chen^[41]	水蒸气	4-102-2	Adam，跟随正则化领导者法（follow the regularized leader, Ftrl）自适应学习率法（adaptive delta，Adadelta）	—	实验216
Ringstad等^[42]	CO₂	GPR	—	—	模拟200
Shukla等^[43]	R141b	5-(5-5-5-5)-1	SCG	遗传算法	模拟1008
Liu等^[44-45]	CO₂	5-10-1	LM	遗传算法	模拟350
Zhang等^[46]	水蒸气	4-9-1	—	粒子群优化算法	模拟97
Maghsoodi等^[47]	工作流体为纯氢气，引射流体为氢气和水蒸气	4-4-1	—	遗传算法	模拟167
Jahingir和Huque^[48]	工作流体为氧气与氢气，引射流体为空气	—	—	满意度函数优化	模拟27
杨飞飞^[49]	工作流体为空气，引射流体为乳胶制品	4-(6-1)-1	动量梯度下降算法（gradient descent with momentum，GDM）	控制步长方法	模拟16
高孝良^[50]	一氯二氟甲烷（R22）	2-(10-20)-1	LM	控制步长方法	—

作者	工质	模型/网络结构	训练算法/优化器	优化方法	数据集
Rashidi等^[51]	R600	2-3-5	—	PSO和ACO_R	模拟1000
Sangesaraki 等^[52]	R600a	6-(9-9)-1	—	GA	模拟1000
Ebrahimi-Moghadam等^[53]	氨	6-(4-4)-1	LM	NSGA-Ⅱ	模拟32140
Hai等^[54]	—	6隐藏层	—	NSGA-Ⅱ和TOPSIS	模拟1500
Soltani等^[55]	溴化锂-水	5-15-2	—	GA	模拟1300
史棋棋^[59]	R141b	2-(3隐藏层)-1	Adam，RMSprop，SGD，AdaGrad	—	模拟1640
Hao等^[56]	HFC-245fa HCFO-1233zd(E) HCFO-1224yd(Z) R600 HFO-1336mzz(Z)	SVR，KRR，RFR	—	—	—
Wang等^[57]	R134a	—	—	—	—
Sözen等^[60]	甲醇-溴化锂	3-(3-8)-1	SCG，CGP，LM	—	实验174
Sözen等^[60]	甲醇-溴化锂	4-8-3	SCG	—	实验（大小未知）
Sözen和Akçayol ^[61-62]	氨-水	4-20-3	GDR	—	模拟（大小未知）
Sözen等^[63]	甲醇-氯化锂	3-(3-8)-1	SCG，CGP，LM	—	实验174
Sözen等^[63]	甲醇-氯化锂	4-6-5	LM	—	实验（大小未知）
Sözen等^[64]	氨-水	4-7-3	SCG，LM	—	模拟（大小未知）
Sözen和Arcaklioğlu^[65]	氨-水	4-(5,6,7)-4	SCG，LM	—	模拟（大小未知）
Shen等^[58]	燃油	SVR	—	—	实验15

作者	工质	模型/网络结构	训练算法/优化器	优化方法	数据集
Rashidi等^[51]	R600	2-3-5	—	PSO和ACO_R	模拟1000
Sangesaraki 等^[52]	R600a	6-(9-9)-1	—	GA	模拟1000
Ebrahimi-Moghadam等^[53]	氨	6-(4-4)-1	LM	NSGA-Ⅱ	模拟32140
Hai等^[54]	—	6隐藏层	—	NSGA-Ⅱ和TOPSIS	模拟1500
Soltani等^[55]	溴化锂-水	5-15-2	—	GA	模拟1300
史棋棋^[59]	R141b	2-(3隐藏层)-1	Adam，RMSprop，SGD，AdaGrad	—	模拟1640
Hao等^[56]	HFC-245fa HCFO-1233zd(E) HCFO-1224yd(Z) R600 HFO-1336mzz(Z)	SVR，KRR，RFR	—	—	—
Wang等^[57]	R134a	—	—	—	—
Sözen等^[60]	甲醇-溴化锂	3-(3-8)-1	SCG，CGP，LM	—	实验174
Sözen等^[60]	甲醇-溴化锂	4-8-3	SCG	—	实验（大小未知）
Sözen和Akçayol ^[61-62]	氨-水	4-20-3	GDR	—	模拟（大小未知）
Sözen等^[63]	甲醇-氯化锂	3-(3-8)-1	SCG，CGP，LM	—	实验174
Sözen等^[63]	甲醇-氯化锂	4-6-5	LM	—	实验（大小未知）
Sözen等^[64]	氨-水	4-7-3	SCG，LM	—	模拟（大小未知）
Sözen和Arcaklioğlu^[65]	氨-水	4-(5,6,7)-4	SCG，LM	—	模拟（大小未知）
Shen等^[58]	燃油	SVR	—	—	实验15

作者	工质	模型/网络结构	训练算法/优化器	优化方法	数据集
Chen和Cai^[66]	R134a	7-10-1	BP	—	实验396
Bencharif等^[67]	HFC-245fa	12-10-2	LM	—	实验233
Bencharif等^[38]	R134a HFC-245fa R141b R1234ze(E) R1233zd(E)	12-10-2	LM	—	实验959
Bencharif等^[38]	R134a HFC-245fa R141b R1234ze(E) R1233zd(E)	7-10-4	LM	—	实验959
Gupta等^[68]	R113，R141b R134a，R718 HFC-245fa，R600a R290,R152a R123，R717 R1234ze R744，空气	7-84-2	Adam	—	实验(大小未知)
Gupta等^[32]	空气	5-17-2	Adam	—	实验543
Abbady等^[69]	R1234yf	8-20-1	LM	—	模拟1170
黄亮亮等^[72]	—	5-10-1	SA-BP	—	实验1348
黄亮亮和曹家枞^[73]	—	5-12-1	CACS-BP, ACO_R-BP	—	实验1131
Zhu等^[74]	水，R134a，R141b	6-10-1	PSO-BPNN和动态误差补偿方法	—	实验107
Gupta等^[70]	空气	4-17-1	Adam	—	实验541
Gupta等^[70]	空气	5-18-2	Adam	—	实验541
Zhang等^[75]	水	7-18-1	LM，RBP，SCG	—	实验489
谷迎港^[76]	水	7-18-1	LM，BR，SCG	—	实验516
Xu等^[77]	工作流体为H₂，引射流体为H₂、水和N₂的混合物	LASSO回归和弹性网络回归	—	—	模拟343
Petrovic等^[71]	工作流体为天然气或R2气体，引射流体为工业中产生的废气	混合专家模型	—	—	实验12（天然气）实验22（R2气体）
António等^[78]	—	4-10-2	LM	—	模拟5000
Jeon等^[79]	R600a	4-(3隐藏层)-1	LM	—	实验(大小未知)
Jamali等^[80]	CO₂	11-12-2	LM	多目标遗传算法	模拟>600