混合增强智能驱动的核电重要厂用水系统设计需求分析研究

doi:10.3901/JME.2025.09.383

机械工程学报 ›› 2025, Vol. 61 ›› Issue (9): 383-401.doi: 10.3901/JME.2025.09.383

• 数字化设计与制造 • 上一篇

扫码分享

混合增强智能驱动的核电重要厂用水系统设计需求分析研究

吴轩宇¹, 娄山河^1,2, 洪兆溪¹, 司恒远³, 张志峰¹, 冯毅雄^1,4, 谭建荣¹

1. 浙江大学流体动力基础件与机电系统全国重点实验室杭州 310027;
2. 南洋理工大学机械与宇航工程学院新加坡 639798 新加坡;
3. 中广核工程设计有限公司深圳 518172;
4. 贵州大学省部共建公共大数据国家重点实验室贵阳 550025

收稿日期:2024-05-08 修回日期:2024-11-20 发布日期:2025-06-12
通讯作者: 冯毅雄,男,1975年出生,博士,长江学者特聘教授。主要研究方向为现代机械设计理论与方法。E-mail:fyxtv@zju.edu.cn E-mail:fyxtv@zju.edu.cn
作者简介:吴轩宇,男,1998年出生,博士研究生。主要研究方向为产品数字化设计。E-mail:xuanyuwu@zju.edu.cn;娄山河,男,1993年出生,博士,博士后。主要研究方向为机械产品概念设计。E-mail:loushanhe@zju.edu.cn;洪兆溪,女,1991年出生,博士,博士后。主要研究方向为产品设计方法学。E-mail:hzhx@zju.edu.cn;司恒远,男,1982年出生,正高级工程师。主要研究方向为核电厂型号研发。E-mail:sihengyuan@cgnpc.com.cn;张志峰,男,1991年出生,博士研究生。主要研究方向为绿色设计理论与智能调度。E-mail:zhzhfeng@zju.edu.cn;谭建荣,男,1954年出生,博士,教授,中国工程院院士。主要研究方向为CAX方法学、工程图学。E-mail:egi@zju.edu.cn
基金资助:
国家自然科学基金(52105281，52130501)和浙江省重点研发计划(2023C01214，2024C01207)资助项目。

Research on Design Requirements Analysis for Essential Service Water System of Nuclear Power Driven by Hybrid-augmented Intelligence

WU Xuanyu¹, LOU Shanhe^1,2, HONG Zhaoxi¹, SI Hengyuan³, ZHANG Zhifeng¹, FENG Yixiong^1,4, TAN Jianrong¹

1. State Key Laboratory of Fluid Power and Mechatronic Systems, Zhejiang University, Hangzhou 310027;
2. School of Mechanical and Aerospace Engineering, Nanyang Technological University, Singapore 639798 Singapore;
3. China General Nuclear Power Design Co., Ltd., Shenzhen 518172;
4. State Key Laboratory of Public Big Data, Guizhou University, Guiyang 550025

Received:2024-05-08 Revised:2024-11-20 Published:2025-06-12

摘要/Abstract

摘要： 重要厂用水系统是与核安全紧密相关的系统，负责在各种运行工况条件下导出核岛的热量至最终热阱。由于核电厂选址和机型的差异、海洋水文、地理特征、气象等因素对设计方案的影响，重要厂用水系统具备按单设计型工程特性，合理与直观的设计需求分析是后续配置设计的重要基础。针对如何准确地辨识设计需求重要度以消除认知不确定性、规范地建立设计需求模型以确保需求的完备性与可追溯性等问题，提出混合增强智能驱动的重要厂用水系统设计需求分析方法，旨在利用人类高层次的认知能力弥补计算机处理设计需求信息的不足，同时利用计算机的运算能力突破人脑认知过程的推理能力限制。首先，结合考虑人类语言特性的语言激励双向编码器表示预训练模型、模拟大脑前额叶-海马体神经回路的双向长短时记忆网络以及多头注意力机制对设计需求文本语义进行深度挖掘，并将语义特征输入模拟大脑多巴胺奖励系统的奖励调制脉冲神经网络进行设计需求分类。其次，考虑核电专家语义表达的模糊性、群体权重的不确定性，构建基于犹豫模糊集的非线性规划模型识别设计需求重要度并做进一步修正。最后，利用基于模型的系统工程方法建立形式化、结构化的系统模型对设计需求进行统一完备表达。

关键词: 设计需求分析, 混合增强智能, 脉冲神经网络, 犹豫模糊集, 基于模型的系统工程

Abstract: The essential service water system is closely related to nuclear safety and are responsible for exporting heat from the nuclear island to the final heat sink under various operating conditions. Due to the impact of different siting and types of nuclear power plants, ocean hydrology, geographical features, meteorological and other factors on the design scheme, the essential service water system has the design-to-order engineering feature, and reasonable and intuitive design requirement analysis is an important basis for subsequent configuration design. Aiming at identifying the importance of design requirements to eliminate cognitive uncertainty accurately and establishing design requirements models to ensure their completeness and traceability normally, a hybrid-augmented intelligence-driven design requirements analysis method of the essential service water system is proposed. The purpose is to make use of human high-level cognitive ability to make up for the lack of computer processing design requirements information, and to use computer calculating power to break through the reasoning ability of human brain cognitive process. Firstly, combining the pre-training model of linguistically-motivated bi-directional encoder representation from Transformer that considers the human language characteristics, the bi-directional long short-term memory network that simulates the brain's prefrontal and hippocampus neural circuit, and the multi-head attention mechanism, the text semantics of design requirements are deeply mined. The semantic features are input into the reward-modulated spiking neural network that simulates the brain's dopamine reward system to classify design requirements. Secondly, considering the ambiguity of the semantic expression of nuclear power experts and the uncertainty of group weights, a nonlinear programming model based on hesitancy fuzzy sets is constructed to identify the importance of design requirements and make further corrections. Finally, formal and structured system models are established based on the model-based systems engineering method to express design requirements in a unified and complete manner.

Key words: design requirements analysis, hybrid-augmented intelligence, spiking neural network, hesitant fuzzy set, model-based systems engineering

中图分类号:

TG156

吴轩宇, 娄山河, 洪兆溪, 司恒远, 张志峰, 冯毅雄, 谭建荣. 混合增强智能驱动的核电重要厂用水系统设计需求分析研究[J]. 机械工程学报, 2025, 61(9): 383-401.

WU Xuanyu, LOU Shanhe, HONG Zhaoxi, SI Hengyuan, ZHANG Zhifeng, FENG Yixiong, TAN Jianrong. Research on Design Requirements Analysis for Essential Service Water System of Nuclear Power Driven by Hybrid-augmented Intelligence[J]. Journal of Mechanical Engineering, 2025, 61(9): 383-401.

参考文献

[1] MALLAPATY S. How China could be carbon neutral by mid-century[J]. Nature,2020,586(7830):482-483.
[2] ZHAO X,MA X W,CHEN B Y,et al. Challenges toward carbon neutrality in China:Strategies and countermeasure[J]. Resources Conservation and Recycling,2022,176:105959.
[3] 胡炳涛,冯毅雄,密尚华,等. 面向核电装备的全生命周期价值链协同模式研究[J]. 机械工程学报,2022,58(13):213-227. HU Bingtao,FENG Yixiong,MI Shanghua,et al. Novel collaborative mode of lifecycle value chain for nuclear power equipment[J]. Journal of Mechanical Engineering,2022,58(13):213-227.
[4] 冯毅雄,赵泽田,胡炳涛,等. 工业大数据驱动的核电装备智能设计-制造-服务技术与应用[J]. 中国科学:技术科学,2023,53(7):1043-1061. FENG Yixiong,ZHAO Zetian,HU Bingtao,et al. Technology and application of industrial big data-driven intelligent design manufacturing and service for nuclear power equipment[J]. Scientia Sinica(Technologica),2023,53(7):1043-1061.
[5] 吴轩宇,洪兆溪,刘继红,等. 复杂定制产品智能化设计与验证协同模式[J]. 计算机集成制造系统,2022,28(9):2700-2717. WU Xuanyu,HONG Zhaoxi,LIU Jihong,et al. Collaborative pattern of intelligent design and verification for complex customized product[J]. Computer Integrated Manufacturing Systems,2022,28(9):2700-2717.
[6] WANG B C,PENG T,WANG X V,et al. Human-centric smart manufacturing[J]. Journal of Manufacturing Systems,2023,69:18-19.
[7] LENG J W,SHA W N,WANG B C,et al. Industry 5.0:Prospect and retrospect[J]. Journal of Manufacturing Systems,2022,65:279-295.
[8] LI M D,LOU S H,GAO Y C,et al. A cerebellar operant conditioning-inspired constraint satisfaction approach for product design concept generation[J]. International Journal of Production Research,2023,61(17):5822-5841.
[9] 王建,黎荣,武浩远,等. 按单设计型复杂产品的配置设计技术综述[J]. 计算机集成制造系统,2020,26(02):410-425. WANG Jian,LI Rong,WU Haoyuan,et al. Review of configuration design for complicated engineer-to-order products[J]. Computer Integrated Manufacturing Systems,2020,26(02):410-425.
[10] ZHANG S Y,XU J H,GUO H W,et al. A research review on the key technologies of intelligent design for customized products[J]. Engineering,2017,3(5):631-640.
[11] CHENG F M,YU S H,CHU J J,et al. Customer satisfaction-oriented product configuration approach based on online product reviews[J]. Multimedia Tools and Applications,2022,81(3):4413-4433.
[12] YUKSEL N,BORKLU H,SEZER H,et al. Review of artificial intelligence applications in engineering design perspective[J]. Engineering Applications of Artificial Intelligence,2023,118:105697.
[13] LI X Y,NASSEHI A,WANG B C,et al. Human-centric manufacturing for human-system coevolution in Industry 5.0[J]. CIPR Annals-Manufacturing Technology,2023,72(1):393-396.
[14] YANG M L,JIANG P Y,ZANG T S,et al. Data-driven intelligent computational design for products:Method,techniques and applications[J]. Journal of Computational Design and Engineering,2023,10(4):1561-1578.
[15] LOU S H,FENG Y X,LI Z W,et al. An edge-based distributed decision-making method for product design scheme evaluation[J]. IEEE Transactions on Industrial Informatics,2021,17(2):1375-1385.
[16] MINH D,WANG H X,LI Y F,et al. Explainable artificial intelligence:A comprehensive review[J]. Artificial Intelligence Review,2022,55(5):3503-3568.
[17] TAVANAEI A,GHODRATI M,KHERADPISHEH S R,et al. Deep learning in spiking neural networks[J]. Neural Networks,2019,111:47-63.
[18] WANG X W,LIN X H,DANG X C. Supervised learning in spiking neural networks:A review of algorithms and evaluations[J]. Neural Networks,2020,125:258-280.
[19] YAMAZAKI K,VO-HO V K,BULSARA O,et al. Spiking neural networks and their applications:A review[J]. Brain Sciences,2022,12(7):863.
[20] WU X Y,FENG Y X,LOU S H,et al. Improving Neucube spiking neural networks for EEG-based pattern recognition using transfer learning[J]. Neurocomputing,2023,529:222-235.
[21] LEWIS R G,FLORIO E,PUNZO D,et al. The brain's reward system in health and disease[J]. Circadian Clock in Brain Health and Disease,2021,1344:57-69.
[22] KLEIN M O,BATTAGELLO D S,CARDOSO A R,et al. Dopamine:Functions,signaling,and association with neurological diseases[J]. Cellular and Molecular Neurobiology,2019,39(1):31-59.
[23] ZHOU Q,LI X H. A bio-inspired hierarchical spiking neural network with reward-modulated STDP learning rule for AER object recognition[J]. IEEE Sensors Journal,2022,22(16):16323-16338.
[24] SUN H,GUO W,WANG L,et al. An analysis method of dynamic requirement change in product design[J]. Computers & Industrial Engineering,2022,171:108477.
[25] BOCK C,GALEY C. Integrating four-dimensional ontology and systems requirements modelling[J]. Journal of Engineering Design,2019,30(10-12):477-522.
[26] CAMPO K X,TEPER T,EATON C E,et al. Model-based systems engineering:Evaluating perceived value,metrics,and evidence through literature[J]. Systems Engineering,2023,26(01):104-129.
[27] ZHANG M,SUN L,WANG G A,et al. Using neutral sentiment reviews to improve customer requirement identification and product design strategies[J]. International Journal of Production Economics,2022,254:108641.
[28] JOUNG J,KIM H M. Explainable neural networks-based approach to Kano categorization of product features from online reviews[J]. International Journal of Production Research,2022,60(23):7053-7073.
[29] ULCAR M,ROBNIK-SIKONJA M. Cross-lingual alignments of ELMo contextual embeddings[J]. Neural Computing & Applications,2022,34(15):13043-13061.
[30] NATH S,MARIE A,ELLERSHAW S,et al. New meaning for NLP:The trials and tribulations of natural language processing with GPT-3 in ophthalmology[J]. British Journal of Ophthalmology,2022,106(7):889-892.
[31] CUI Y M,CHE W X,LIU T,et al. Pre-training with whole word masking for Chinese BERT[J]. IEEE-ACM Transactions on Audio Speech and Language Processing,2021,29:3504-3514.
[32] WANG H F,LI J W,WU H,et al. Pre-trained language models and their applications[J]. Engineering,2023,25:51-65.
[33] CHE W X,LI Z H,LIU T,et al. LTP:A Chinese language technology platform[C]//Proceedings of the 23rd International Conference on Computational Linguistics. 2010:13–16.
[34] WU J H,WANG Z C,HU Y,et al. Runoff forecasting using convolutional neural networks and optimized bi-directional long short-term memory[J]. Water Resources Management,2023,37(2):937-953.
[35] FORTUNA L,BUSCARINO A. Spiking neuron mathematical models:A compact overview[J]. Bioengineering- Basel,2023,10(02):174.
[36] SBOEV A,SERENKO A,RYBKA R,et al. Solving a classification task by spiking neural network with STDP based on rate and temporal input encoding[J]. Mathematical Methods in the Applied Sciences,2020,43(13):7802- 7814.
[37] DIEHL P U,COOK M. Unsupervised learning of digit recognition using spike-timing-dependent plasticity[J]. Frontiers in Computational Neuroscience,2015,9:99.
[38] VLASOV D,MINNEKHANOV A,RYBKA R,et al. Memristor-based spiking neural network with online reinforcement learning[J]. Neural Networks,2023,166:512-523.
[39] SBOEV A,VLASOV D,RYBKA R, et al. Modeling the dynamics of spiking networks with memristor-based STDP to solve classification tasks[J]. Mathematics,2022,9(24):3237.
[40] NOBUKAWA S,NISHIMURA H,YAMANISHI T. Pattern classification by spiking neural networks combining self-organized and reward-related spike-timing dependent plasticity[J]. Journal of Artificial Intelligence and Soft Computing Research,2019,9(4):283-291.
[41] XU Z S,ZHANG S. An overview on the applications of the hesitant fuzzy sets in group decision-making:Theory,support and methods[J]. Frontiers of Engineering Management,2019,6(2):163-182.
[42] YU B C,ZHAO X J,ZHENG M F,et al. Entropy on intuitionistic fuzzy sets and hesitant fuzzy sets[J]. Journal of Mathematics,2022,2022:1585079.
[43] GAO J,XU Z S,LIANG Z L,et al. Two integral models and applications of hesitant fuzzy information fusion[J]. IEEE Transactions on Fuzzy Systems,2023,31(1):25-39.
[44] ZENG W Y,MA R,LI D Q,et al. Distance measure of hesitant fuzzy sets and its application in image segmentation[J]. International Journal of Fuzzy Systems,2022,24(7):3134-3143.

混合增强智能驱动的核电重要厂用水系统设计需求分析研究

Research on Design Requirements Analysis for Essential Service Water System of Nuclear Power Driven by Hybrid-augmented Intelligence

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 1

编辑推荐

Metrics

本文评价