Machine learning-based correlation for economic evaluation of HTSE-nuclear cogeneration plant

被引：0

作者：

Sadeghi, Khashayar ^{[1
,2
]}

Ghazaie, Seyed Hadi ^{[1
,2
]}

Sokolova, Ekaterina ^{[1
,2
]}

Sergeev, Vitaly ^{[1
]}

Ksenia, Naypak ^{[1
]}

Yang, Luopeng ^{[1
,2
]}

机构：

[1] Peter the Great St Petersburg Polytech Univ, Dept Nucl & Heat Power Engn, St Petersburg 195251, Russia

[2] Shandong Jianzhu Univ, Sch Thermal Engn, Jinan 250101, Shandong, Peoples R China

来源：

INTERNATIONAL JOURNAL OF HYDROGEN ENERGY | 2025年 / 114卷

基金：

俄罗斯科学基金会;

关键词：

Nuclear cogeneration; Machine learning; Gene expression programming; High-temperature steam electrolysis; Hydrogen economy; 1ST;

D O I：

10.1016/j.ijhydene.2025.02.423

中图分类号：

O64 [物理化学（理论化学）、化学物理学];

学科分类号：

070304 ; 081704 ;

摘要：

The integration of high-temperature steam electrolysis system (HTSE) into light-water nuclear reactors for low- emission hydrogen production is a newly introduced application of nuclear co-generation plants, which recently has attracted considerable interest due to its high compatibility with the net zero emissions goal. The economic evaluation of the HTSE-nuclear cogeneration plant has a crucial role in effectively persuading decision-makers to use this technology on an industrial scale. Machine learning algorithms trained by well-structured datasets can be used for facilitating the prediction of levelized hydrogen cost based on the most basic input parameters. In this study, a systematic procedure for developing some applicable correlations to predict the levelized cost of the HTSE-nuclear cogeneration plant by means of gene expression programming (GEP) as a powerful tool in evolutionary machine learning approaches is introduced. An economic model based on technical and economic parameters is used for generating datasets to train GEP. The one-way sensitivity analysis has been performed to purify the dataset and remove non-influential parameters from the dataset. The propagation of uncertainties in the dataset is done based on the statistical Wilk's method by taking 59 random samples. GEP is trained by 80% of the non-dimensional generated dataset and three correlations for prediction of levelized capital cost, levelized operation and maintenance cost, and levelized decommission cost are developed. The calculated MSE and R2 errors show that GEP successfully could make proper correlations between input parameters and cost components. The operating and maintenance correlation with the R2 > 0.98 could present the highest accuracy among other correlations. In addition, the obtained relative error for predicted values of total hydrogen cost is less than 2%, which indicates the high accuracy achieved from the correlations derived by GEP.

引用

页码：337 / 351

页数：15

共 47 条

[1] Frick K., Et al., Technoeconomic assessment of hydrogen cogeneration via high temperature steam electrolysis with a light-water reactor, Appl Energy, 306, (2022)
[2] Hadi Ghazaie S., Et al., On the use of advanced nuclear cogeneration plant integrated into latent heat storage for district heating, Sustain Energy Technol Assessments, 50, (2022)
[3] Sadeghi K., Et al., Comprehensive techno-economic analysis of integrated nuclear power plant equipped with various hybrid desalination systems, Desalination, 493, (2020)
[4] Khawaja M.K., Et al., Strategies to promote nuclear energy utilization in hydrogen production, Int J Hydrogen Energy, 82, pp. 36-46, (2024)
[5] Sadeghi K., Et al., Towards net-zero emissions through the hybrid SMR-solar cogeneration plant equipped with modular PCM storage system for seawater desalination, Desalination, 524, (2022)
[6] Jia B., Et al., Review on droplet falling film evaporation in desalination: a focus on heat transfer enhancement of micro-nano structured surfaces, Desalination, 598, (2025)
[7] Marzouk O.A., Ambitions for hydrogen, clean hydrogen, and green hydrogen, Engineering Proceedings, 56, 1, (2025)
[8] Abbasian Hamedani E., Alenabi S.A., Talebi S., Hydrogen as an energy source: a review of production technologies and challenges of fuel cell vehicles, Energy Rep, 12, pp. 3778-3794, (2024)
[9] Sebbahi S., Et al., A comprehensive review of recent advances in alkaline water electrolysis for hydrogen production, Int J Hydrogen Energy, 82, pp. 583-599, (2024)
[10] Smolinka T., Et al., Chapter 4 - the history of water electrolysis from its beginnings to the present, Electrochemical power sources: fundamentals, systems, and applications, pp. 83-164, (2022)

← 1 2 3 4 5 →