模拟人脑工作机制的机械产品概念设计行为原理启发求解方法

doi:10.3901/JME.2023.24.344

机械工程学报 ›› 2023, Vol. 59 ›› Issue (24): 344-358.doi: 10.3901/JME.2023.24.344

扫码分享

模拟人脑工作机制的机械产品概念设计行为原理启发求解方法

娄山河¹, 冯毅雄¹, 郑浩², 胡炳涛¹, 洪兆溪¹, 谭建荣¹

1. 浙江大学流体动力与机电系统国家重点实验室杭州 310027;
2. 北京航空航天大学杭州创新研究院杭州 310051

收稿日期:2023-03-20 修回日期:2023-09-23 出版日期:2023-12-20 发布日期:2024-03-05
通讯作者: 冯毅雄(通信作者),男,1975年出生,博士,教授,博士研究生导师。主要研究方向为现代设计理论与方法。E-mail:fyxtv@zju.edu.cn
作者简介:娄山河,男,1993年出生,博士研究生。主要研究方向为机械产品概念设计、设计认知。E-mail:loushanhe@zju.edu.cn;郑浩,男,1988年出生,博士,博士后。主要研究方向为产品设计理论、智能设计与信息融合技术。E-mail:haozheng@zju.edu.cn;胡炳涛,男,1992年出生,博士,博士后。主要研究方向为产品设计理论与智能制造。E-mail:hbt@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:hzx@zju.edu.cn;谭建荣,男,1954年出生,博士,教授,博士研究生导师,中国工程院院士。主要研究方向为CAX方法学、工程图学、企业信息化。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)资助项目

Brain-like Behavior Effect-inspiring and Principle Structure-combining Method for Mechanical Product Conceptual Design

LOU Shanhe¹, FENG Yixiong¹, ZHENG Hao², HU Bingtao¹, HONG Zhaoxi¹, TAN Jianrong¹

1. State Key Laboratory of Fluid Power and Mechatronic Systems, Zhejiang University, Hangzhou 310027;
2. Hangzhou Innovation Institute, Beihang University, Hangzhou 310051

Received:2023-03-20 Revised:2023-09-23 Online:2023-12-20 Published:2024-03-05

摘要/Abstract

摘要： 机械产品概念设计是现代设计理论研究的热点问题，其核心是功构映射，即根据既定的功能需求，以行为效应作为中间载体，求解满足设计约束的原理构件组合。以概念设计的功构映射理论作为基础，利用设计师大脑认知过程的神经机制作为依据，提出模拟人脑工作机制的机械产品概念设计行为原理启发求解方法。分析前额叶-海马体神经回路中设计经验知识的记忆启发机制，并以此为基础构建行为效应重构的记忆启发强化学习模型，利用多巴胺调控的Q学习在状态转移概率和设计环境回报函数未知的情况下实现行为效应的重构认知传递。同时，分析小脑模块化神经结构的反射调控作用，构建以全连接多层感知机和循环神经网络为模块的约束满足神经网络，形成复杂设计约束下的操作性条件反射判断设计空间中的约束可解性，为原理构件的组合求解提供设计依据。以电梯曳引系统的概念设计为例验证所提出方法的有效性和可行性。

关键词: 概念设计, 功构映射, 设计认知, 强化学习, 约束满足神经网络

Abstract: The research of mechanical product conceptual design is a hot topic in modern design theory. Its core is function-structure mapping, which aims to solve the combination of principle structures satisfying the design constraints according to the functional requirements and using behavior effect as the intermediate carrier. Based on the function-structure mapping theory and the neural mechanism of the designer’s brain cognitive process, a brain-like behavior effect-inspiring and principle structure-combining method is proposed. The memory inspiration mechanism of the frontal lobe-hippocampus neural circuit is revealed. Then a memory-inspired reinforcement learning model for behavior effect refactoring is constructed. The dopamine-regulated Q learning is utilized to realize the process of behavioral effects restructuring and cognitive transmission. Moreover, the reflex regulation mechanism of the modularized cerebellum is revealed, and a constraint satisfaction neural network is constructed based on fully connected multilayer perceptron and recurrent neural networks. This can form the operant conditioning under complex design constraints to judge the constraint solvability, and then provide combinatorial solutions for principle structures. A case study is conducted to illustrate the feasibility of the proposed method.

Key words: conceptual design, function-structure mapping, design cognition, reinforcement learning, constraint satisfaction neural network

中图分类号:

TG156

娄山河, 冯毅雄, 郑浩, 胡炳涛, 洪兆溪, 谭建荣. 模拟人脑工作机制的机械产品概念设计行为原理启发求解方法[J]. 机械工程学报, 2023, 59(24): 344-358.

LOU Shanhe, FENG Yixiong, ZHENG Hao, HU Bingtao, HONG Zhaoxi, TAN Jianrong. Brain-like Behavior Effect-inspiring and Principle Structure-combining Method for Mechanical Product Conceptual Design[J]. Journal of Mechanical Engineering, 2023, 59(24): 344-358.

参考文献

[1] BHASIN D,BEHMER S T,MCADAMS D A. Fostering function-sharing using bioinspired product architecture[J]. Journal of Mechanical Design,2020,143(6):1-35.
[2] LOU S,FENG Y,LI Z W,et al. An edge-based distributed decision-making method for product design scheme evaluation[J]. IEEE Transactions on Industrial Informatics,2020,17(2):1375-1385.
[3] FORMENTINI G,BOUISSIERE F,CUILLER C,et al. Conceptual design for assembly methodology formalization:Systems installation analysis and manufacturing information integration in the design and development of aircraft architectures[J]. Journal of Industrial Information Integration,2022,26:100327.
[4] GOMES M G,OGLIARI A,FERNANDES R B,et al. Evaluation of physical models as creative stimuli in conceptual design of products[J]. Design Studies,2022,81:101119.
[5] 邓家褆. 产品概念设计:理论,方法与技术[M]. 北京:机械工业出版社,2002. DENG Jiati. Product conceptual design:Theory,method and technology[M]. Beijing:China Machine Press,2002.
[6] 刘龙繁,李彦,马金龙,等. 基于知识粒度的TRIZ在创新设计中的应用[J]. 机械工程学报,2016,52(5):22-32. LIU Longfan,LI Yan,MA Jinlong,et al. Application of TRIZ in creative design based on knowledge granularity[J]. Journal of Mechanical Engineering,2016,52(5):22-32.
[7] HASHEMI H,SHAHAROUN A M,SUDIN I. A case-based reasoning approach for design of machining fixture[J]. The International Journal of Advanced Manufacturing Technology,2014,74(1-4):113-124.
[8] RELICH M,PAWLEWSKI P. A case-based reasoning approach to cost estimation of new product development[J]. Neurocomputing,2018,272:40-45.
[9] YUKISH M A,STUMP G M,MILLER S W. Using recurrent neural networks to model spatial grammars for design creation[J]. Journal of Mechanical Design,2020,142:104501.
[10] CAO G,SUN Y,TAN R,et al. A function-oriented biologically analogical approach for constructing the design concept of smart product in Industry 4. 0[J]. Advanced Engineering Informatics,2021,49:101352.
[11] JING L,LI Q,MA J,et al. A conceptual design decision approach by integrating rough Bayesian network and game theory under uncertain behavior selections[J]. Expert Systems with Applications,2022,202:117108.
[12] HU J,AMINZADEH M,WANG Y. Searching feasible design space by solving quantified constraint satisfaction problems[J]. Journal of Mechanical Design,2014,136(3):031002.
[13] WANG L,ZHONG S S,ZHANG Y J. Process configuration based on generative constraint satisfaction problem[J]. Journal of Intelligent Manufacturing,2017,28(4):945-957.
[14] CHAMPASAK P,PANAGANT N,PHOLDEE N,et al. Self-adaptive many-objective meta-heuristic based on decomposition for many-objective conceptual design of a fixed wing unmanned aerial vehicle[J]. Aerospace Science and Technology,2020,100:105783.
[15] SONG W,CAO Z,ZHANG J,et al. Learning variable ordering heuristics for solving constraint satisfaction problems[J]. Engineering Applications of Artificial Intelligence,2022,109:104603.
[16] STERMBERG R J. Views on origins of genius[J]. Treands in Cognitive Science,2000,4(6):50.
[17] GAZZANIGA M,IVRY R,MANGUN G,等. 认知神经科学:关于心智的生物学[M]. 北京:中国轻工业出版社,2011. GAZZANIGA M,IVRY R,MANGUN G,et al. Cognitive neuroscience:Biology of the mind[M]. Beijing:China Light Industry Press,2011.
[18] HU W L,REID T. The effects of designers' contextual experience on the ideation process and design outcomes[J]. Journal of Mechanical Design,2018,140(10):101101.
[19] GOUCHER-LAMBERT K,MOSS J,CAGAN J. A neuroimaging investigation of design ideation with and without inspirational stimuli-understanding the meaning of near and far stimuli[J]. Design Studies,2019,60:1-38.
[20] MILOVANOVIC J,HU M,SHEALY T,et al. Characterization of concept generation for engineering design through temporal brain network analysis[J]. Design Studies,2021,76:101044.
[21] VIEIRA S,BENEDEK M,GERO J,et al. Design spaces and EEG frequency band power in constrained and open design[J]. International Journal of Design Creativity and Innovation,2022:1-28.
[22] HAY L,DUFFY A H B,GILBERT S J,et al. Functional magnetic resonance imaging (fMRI) in design studies:Methodological considerations,challenges,and recommendations[J]. Design Studies,2022,78:101078.
[23] BADDELEY A. Working memory[J]. Science,1992,255(5044):556-559.
[24] MINXHA J,ADOLPHS R,FUSI S,et al. Flexible recruitment of memory-based choice representations by the human medial frontal cortex[J]. Science,2020,368(6498):eaba3313.
[25] 库逸轩. 工作记忆的认知神经机制[J]. 生理学报,2019,71(1):173-185. KU Yixuan. Cognitive and neural mechanisms underlying working memory[J]. Acta Physiologica Sinica,2019,71(1):173-185.
[26] KRAGEL J E,SCHUELE S,VAN HAERENTS S,et al. Rapid coordination of effective learning by the human hippocampus[J]. Science Advances,2021,7(25):eabf7144.
[27] GERO J S,KANNENGIESSER U. The situated function- behavior-structure framework[J]. Design Studies,2004,25:373-391.
[28] NEFTCI E O,AVERBECK B B. Reinforcement learning in artificial and biological systems[J]. Nature Machine Intelligence,2019,1(3):133-143.
[29] ESTESO A,PEIDRO D,MULA J,et al. Reinforcement learning applied to production planning and control[J]. International Journal of Production Research,2022:1-18.
[30] SCHULTZ W,DAYAN P,MONTAGUE P R. A neural substrate of prediction and reward[J]. Science,1997,275:1593-1599.
[31] MNIH V,KAVUKCUOGLU K,SILVER D,et al. Human-level control through deep reinforcement learning[J]. Nature,2015,518(7540):529-533.
[32] 曹国忠,檀润华,孙建广. 基于扩展效应模型的功能设计过程及实现[J]. 机械工程学报,2009,45(7):157-167. CAO Guozhong,TAN Runhua,SUN Jianguang. Process and realization of functional design based on extended-effect model[J]. Journal of Mechanical Engineering,2009,45(7):157-167.
[33] 王占松,田凌. 基于功能的三维模型检索系统[J]. 计算机辅助设计与图形学学报,2013,25(12):1877-1885. WANG Zhansong,TIAN Ling. Function-based 3D model retrieval system[J]. Journal of Computer Aided-Design & Graphics,2013,25(12):1877-1885.
[34] STADDON J E R,CERUTTI D T. Operant conditioning[J]. Annual Review of Psychology,2003,54:115.
[35] MOMOHARA Y,NEVEU C L,CHEN H M,et al. Specific plasticity loci and their synergism mediate operant conditioning[J]. Journal of Neuroscience,2022,42(7):1211-1223.
[36] DACRE J,COLLIGAN M,CLARKE T,et al. A cerebellar-thalamocortical pathway drives behavioral context-dependent movement initiation[J]. Neuron,2021,109(14):2326-2338.
[37] MITTMANN W,HÄUSSER M. Linking synaptic plasticity and spike output at excitatory and inhibitory synapses onto cerebellar Purkinje cells[J]. Journal of Neuroscience,2007,27(21):5559-5570.
[38] BOSTAN A C,STRICK P L. The basal ganglia and the cerebellum:Nodes in an integrated network[J]. Nature Reviews Neuroscience,2018,19(3):1-11.
[39] SELSAM D,LAMM M,BUNZ B,et al. Learning a SAT solver from single-bit supervision[C]//ICLR 2019,arXiv:1802. 03685.

模拟人脑工作机制的机械产品概念设计行为原理启发求解方法

Brain-like Behavior Effect-inspiring and Principle Structure-combining Method for Mechanical Product Conceptual Design

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	王无印, 黄子钊, 庄子龙, 方怀瑾, 秦威. 基于深度强化学习的自动化码头堆场场桥调度方法[J]. 机械工程学报, 2024, 60(6): 44-57.
[2]	赵阔, 王皂琦, 潘臻信, 潘扬华, 张中飞, 屈挺. 大数据驱动的快消品终端拜访“云-边”联动决策与优化[J]. 机械工程学报, 2024, 60(6): 58-68.
[3]	郭景华, 李文昌, 王班, 王靖瑶. 基于深度强化学习的网联混合动力汽车队列控制[J]. 机械工程学报, 2024, 60(2): 262-271.
[4]	娄山河, 冯毅雄, 胡炳涛, 洪兆溪, 谭建荣. 人机认知协同的复杂装备概念设计：挑战、进展和展望[J]. 机械工程学报, 2024, 60(11): 2-19.
[5]	李文礼, 张祎楠, 石晓辉, 王梦昕. 基于博弈论的右转无信号交叉口行人行为模拟[J]. 机械工程学报, 2024, 60(10): 86-101.
[6]	曾迪, 郑玲, 李以农, 杨显通. 自动驾驶奖励函数贝叶斯逆强化学习方法[J]. 机械工程学报, 2024, 60(10): 245-260.
[7]	杨硕, 李时珍, 赵中原, 黄小鹏, 黄岩军. 基于时序差分学习模型预测控制的一体化自动驾驶换道策略[J]. 机械工程学报, 2024, 60(10): 329-338.
[8]	张志勇, 黄大洋, 黄彩霞, 胡林, 杜荣华. TD3算法改进与自动驾驶汽车并道策略学习[J]. 机械工程学报, 2023, 59(8): 224-234.
[9]	郭洪飞, 陆鑫宇, 任亚平, 张超勇, 李建庆. 基于强化学习的群体进化算法求解双边多目标同步并行拆解线平衡问题[J]. 机械工程学报, 2023, 59(7): 355-366.
[10]	郑湃, 李成熙, 殷悦, 张荣, 鲍劲松, 王柏村, 谢海波, 王力翚. 增强现实辅助的互认知人机安全交互系统[J]. 机械工程学报, 2023, 59(6): 173-184.
[11]	景立挺, 陈盈臣, 谢静, 谭博文, 姜少飞. 考虑用户权重不一致性和复杂交互影响的产品可持续设计需求评估研究[J]. 机械工程学报, 2023, 59(3): 218-231.
[12]	马丽莹, 魏云冰. 基于DDPG算法的发电企业报价策略研究[J]. 电气工程学报, 2023, 18(2): 192-200.
[13]	林歆悠, 叶卓明, 周斌豪. 基于DQN强化学习的自动驾驶转向控制策略[J]. 机械工程学报, 2023, 59(16): 315-324.
[14]	钟沛成, 骆德渊, 庞明君. 基于深度强化学习的四足机器人跟随策略研究及系统实现[J]. 机械工程学报, 2023, 59(13): 79-88.
[15]	顾文斌, 李育鑫, 刘斯麒, 苑明海, 裴凤雀. 数据驱动的智慧车间实时调度方法研究[J]. 机械工程学报, 2023, 59(12): 47-61.