基于QTTI和BPNN的产品造型感性意象预测模型构建

赵项; 闫茂铠; 李正慧; 吕寒露

doi:10.19554/j.cnki.1001-3563.2025.20.016

PDF(5061 KB)

包装工程（设计栏目） ›› 2025, Vol. 46 ›› Issue (20) : 170-183. DOI: 10.19554/j.cnki.1001-3563.2025.20.016

工业设计

基于QTTI和BPNN的产品造型感性意象预测模型构建

赵项¹, 闫茂铠¹, 李正慧¹, 吕寒露^2,*

作者信息 +

Construction of Product Shape Sensory Imagery Prediction Model Based on QTTI and BPNN

ZHAO Xiang¹, YAN Maokai¹, LI Zhenghui¹, LYU Hanlu^2,*

Author information +

文章历史 +

摘要

目的为明确消费者对产品造型传达情感意象期望,及解决传统产品造型意象研究中感性词汇收集范围狭窄、效率低和情感预测模型精度不佳等问题,提出一种产品造型感性意象预测模型,并以电吹风为例展开探讨。方法应用数量化I类（QTTI）和鲸鱼算法及海鸥算法优化的反向传播神经网络（WOA-BPNN;SOA-BPNN）等探索产品造型传达的情感意象,首先基于Python编程语言从电商平台爬取产品样本和消费者评论文本,并采用词频-逆文档频率算法（TF-IDF）和Word2vec从评论文本中提取代表性感性语汇;其次,采用形态分析对样本造型进行解构得出项目和类目设计元素,并基于语义差异量表对消费者感性意象展开评估;最后,基于QTTI、BPNN、WOA-BPNN、SOA-BPNN构建产品造型设计元素与用户感性意象间关联模型,并采用误差法对其进行精度比较,结果表明WOA-BPNN感性预测模型精度最佳。结论该产品造型意象预测方法可为设计师在造型开发阶段指明方向,并为相关领域学者提供借鉴思路。

Abstract

The work aims to propose a predictive model for perceptual imagery of product shapes with a case study on hair dryers so as to fulfill consumers' expectations for product shapes to convey emotional imagery and address issues such as the narrow scope of perceptual vocabulary collection, low efficiency, and suboptimal accuracy of emotional prediction models in traditional research on product shape imagery. The Quantitative Type-Token TheoryⅠ (QTTI), the Whale Optimization Algorithm (WOA), and the Seagull Optimization Algorithm (SOA) were applied to optimize the Backpropagation Neural Network (BPNN; WOA-BPNN; SOA-BPNN) for exploring the emotional imagery conveyed by product shapes. Firstly, product samples and consumer review texts were crawled from e-commerce platforms based on the Python programming language, and representative perceptual vocabularies were extracted from the review texts using the Term Frequency-Inverse Document Frequency (TF-IDF) algorithm and Word2vec. Secondly, morphological analysis was adopted to deconstruct the shapes of the samples to obtain the project and category design elements, and the perceptual imagery of consumers was evaluated based on the Semantic Differential Scale. Finally, a correlation model between the design elements of product shapes and users' perceptual imagery was constructed based on QTTI, BPNN, WOA-BPNN, and SOA-BPNN, and the accuracy was compared using the error method. The results indicated that the WOA-BPNN perceptual prediction model had the best accuracy. In conclusion, this method can provide guidance for product designers in shaping their design directions and offer reference ideas for researchers in related fields.

导出引用1

赵项, 闫茂铠, 李正慧, 吕寒露. 基于QTTI和BPNN的产品造型感性意象预测模型构建[J]. 包装工程. 2025, 46(20): 170-183 https://doi.org/10.19554/j.cnki.1001-3563.2025.20.016

ZHAO Xiang, YAN Maokai, LI Zhenghui, LYU Hanlu. Construction of Product Shape Sensory Imagery Prediction Model Based on QTTI and BPNN[J]. Packaging Engineering. 2025, 46(20): 170-183 https://doi.org/10.19554/j.cnki.1001-3563.2025.20.016

中图分类号： TB472

参考文献

[1] 陆蔚华, 姜冠岳, 刘雨婷, 等. 公务机驾驶舱内饰布局设计感性评价[J]. 计算机集成制造系统, 2024, 30(1): 28-41.
LU W H, JIANG G Y, LIU Y T, et al.Kansei Evaluation of Interior Layout Design of Business Jet Cockpit[J]. Computer Integrated Manufacturing Systems, 2024, 30(1): 28-41.
[2] KANG X H.Biologically Inspired Product Design Combining Kansei Engineering and Association Creative Thinking Method[J]. Advanced Engineering Informatics, 2024, 62: 102615.
[3] TANG W Y, XIANG Z R, DING T C, et al.Research on Multi-Objective Optimization of Product Form Design Based on Kansei Engineering[J]. Journal of Engineering Design, 2024, 35(8): 1023-1048.
[4] YUAN B K, WU K, WU X Y, et al.Form Generative Approach for Front Face Design of Electric Vehicle under Female Aesthetic Preferences[J]. Advanced Engineering Informatics, 2024, 62: 102571.
[5] YANG C X, XU T F, YE J N.Applying TRIZ and Kansei Engineering to the Eco-Innovative Product Design towards Waste Recycling with Latent Dirichlet Allocation Topic Model Analysis[J]. Engineering Applications of Artificial Intelligence, 2024, 133: 107962.
[6] ZABOTTO C N, SERGIO LUIS DA S, AMARAL D C, et al. Automatic Digital Mood Boards to Connect Users and Designers with Kansei Engineering[J]. International Journal of Industrial Ergonomics, 2019, 74: 102829.
[7] LI Y F, SHIEH M D, YANG C C.A Posterior Preference Articulation Approach to Kansei Engineering System for Product Form Design[J]. Research in Engineering Design, 2019, 30(1): 3-19.
[8] LIU M, BEN L L.Research on Demand Forecasting Method of Multi-User Group Based on Big Data[C]// Human Interface and the Management of Information: Applications in Complex Technological Environments. Cham: Springer International Publishing, 2022: 45-64.
[9] YI J S, OH Y K.The Informational Value of Multi-Attribute Online Consumer Reviews: A Text Mining Approach[J]. Journal of Retailing and Consumer Services, 2022, 65: 102519.
[10] JOUNG J, KIM H.Interpretable Machine Learning- Based Approach for Customer Segmentation for New Product Development from Online Product Reviews[J]. International Journal of Information Management, 2023, 70: 102641.
[11] LAI X J, ZHANG S, MAO N, et al.Kansei Engineering for New Energy Vehicle Exterior Design: An Internet Big Data Mining Approach[J]. Computers & Industrial Engineering, 2022, 165: 107913.
[12] LIU Z H, WU J F, CHEN Q P, et al.An Improved Kansei Engineering Method Based on the Mining of Online Product Reviews[J]. Alexandria Engineering Journal, 2023, 65: 797-808.
[13] 张国方, 寇姣姣, 陈令华. 网络评论文本驱动的汽车设计规划方法[J]. 机械设计, 2021, 38(2): 139-144.
ZHANG G F, KOU J J, CHEN L H.Method of Automobile Design Planning Driven by Web Review Text[J]. Journal of Machine Design, 2021, 38(2): 139-144.
[14] 李志贤. 用户意象驱动的卡车前脸智能化造型设计研究[D]. 济南: 齐鲁工业大学, 2023.
LI Z X.Research on Intelligent Modeling Design of Truck Front Face Driven by User Image[D]. Jinan: Qilu University of Technology, 2023.
[15] 孔浩. 基于深度学习的汽车进气格栅造型设计与研究[D]. 天津: 天津科技大学, 2023.
KONG H.Modeling Design and Research of Automobile Air Intake Grille Based on Deep Learning[D]. Tianjin: Tianjin University of Science & Technology, 2023.
[16] 黄雪芹. 基于深度学习的服务机器人造型设计方法研究与应用[D]. 天津: 河北工业大学, 2022.
HUANG X Q.Research and Application of Modeling Design Method for Service Robot Based on Deep Learning[D]. Tianjin: Hebei University of Technology, 2022.
[17] 赵项, 李正慧, 吴正仲, 等. 基于文献计量学的感性工学研究进展可视化分析[J]. 包装工程, 2023, 44(16): 168-179.
ZHAO X, LI Z H, WU Z Z, et al.Visual Analysis of Research Progress in Kansei Engineering Based on Bibliometrics[J]. Packaging Engineering, 2023, 44(16): 168-179.
[18] 丁满, 丁婷婷, 宋美佳, 等. 基于内隐测量和BP神经网络的产品色彩情感化设计[J]. 计算机集成制造系统, 2023, 29(2): 616-627.
DING M, DING T T, SONG M J, et al.Product Color Emotional Design Method Based on Implicit Measurement and BP Neural Network[J]. Computer Integrated Manufacturing Systems, 2023, 29(2): 616-627.
[19] SHAO P, TAN R H, PENG Q J, et al.An Integrated Method to Acquire Technological Evolution Potential to Stimulate Innovative Product Design[J]. Mathematics, 2023, 11(3): 619.
[20] 王年文, 王劲松, 毕翼飞, 等. 人工智能在感性工学研究中的应用与趋势[J]. 包装工程, 2023, 44(16): 32-40.
WANG N W, WANG J S, BI Y F, et al.Application and Trend of Artificial Intelligence in Kansei Engineering Research[J]. Packaging Engineering, 2023, 44(16): 32-40.
[21] 林哲辉, 吴正仲, 罗峰, 等. 基于感性工学与人工神经网络的电动剃须刀多感官意象设计方法[J]. 机械设计, 2023, 40(2): 149-156.
LIN Z H, WU Z Z, LUO F, et al.Multi-Sensory Design Method of Electric Shavers Based on Kansei Engineering and Artificial Neural Networks[J]. Journal of Machine Design, 2023, 40(2): 149-156.
[22] 钟健平, 费韬. 基于WOA-BPNN的锂电池极片涂布缺陷检测识别[J]. 储能科学与技术, 2022, 11(8): 2537-2545.
ZHONG J P, FEI T.Defects Detection and Recognition of Lithium Battery Electrode Plate Coating Based on WOA-BPNN[J]. Energy Storage Science and Technology, 2022, 11(8): 2537-2545.
[23] ZHAO X, AZIM SHARUDIN S, LV H L.A Novel Product Shape Design Method Integrating Kansei Engineering and Whale Optimization Algorithm[J]. Advanced Engineering Informatics, 2024, 62: 102847.
[24] JENG-CHUNG WOO F L, FENG LUO Z L, et al. Research on the Sensory Feeling of Product Design for Electric Toothbrush Based on Kansei Engineering and back Propagation Neural Network[J]. Journal of Internet Technology, 2022, 23(4), 863-871.
[25] CHEN L L.An extended TF-IDF method for improving keyword extraction in traditional corpus-based research: An example of a climate change corpus[J]. Data & Knowledge Engineering, 2024, 153: 102322.
[26] 苏建宁, 李鹤岐. 基于感性意象的产品造型设计方法研究[J]. 机械工程学报, 2004, 40(4): 164-167.
SU J N, LI H Q.Method of Product Form Design Based on Perceptual Image[J]. Chinese Journal of Mechanical Engineering, 2004, 40(4): 164-167.
[27] LIN Z H, WOO J C, LUO F, et al.Multisensory Design of Electric Shavers Based on Kansei Engineering and Artificial Neural Networks[J]. Mathematical Problems in Engineering, 2023, 2023(1): 1188537.
[28] WANG T, XU M, YANG L, ZHOU M, et al.Constructing a MOEA approach for product form Kansei design based on text mining and BPNN[J]. Journal of Intelligent & Fuzzy Systems, 2024, 46(4):8865-8885.
[29] HU Y, YAN K.Convolutional neural network models combined with kansei engineering in product design[J]. Computational Intelligence and Neuroscience, 2023, 73(1):2572071.
[30] CAI W, WANG Z, WANG Y, et al.Research on wheelchair form design based on Kansei engineering and GWO-BP neural network[J]. Scientific Reports, 2025, 15(1):10258-10267.
[31] SUN B, HU Z B, GUO T.A Multi-GA-BPNN Fusion Algorithm and Full-Scale Experimental Verification for Fire Warning in the Underground Pipe Gallery[J]. Fire Safety Journal, 2024, 144: 104103.
[32] FENG Q, XIE X Y, WANG P H, et al.Prediction of Durability of Reinforced Concrete Based on Hybrid-Bp Neural Network[J]. Construction and Building Materials, 2024, 425: 136091.
[33] LIU J J, WANG J Z, NIU Y B, et al.A Point-Interval Wind Speed Forecasting System Based on Fuzzy Theory and Neural Networks Architecture Searching Strategy[J]. Engineering Applications of Artificial Intelligence, 2024, 132: 107906.
[34] 王正阳. 基于WOA-BPNN的电力企业审计风险评估模型[J]. 中国新技术新产品, 2023(20): 137-139.
WANG Z Y.Audit Risk Assessment Model of Electric Power Enterprises Based on WOA-BPNN[J]. New Technology & New Products of China, 2023(20): 137-139.
[35] LIU Z L, NING D Y, HOU J Y.A Novel Elman Neural Network Based on Gaussian Kernel and Improved SOA and Its Applications[J]. Expert Systems with Applications, 2024, 249: 123453.
[36] 陈默, 黄辰, 苏盼盼. 基于可供性理论的产品设计方法研究[J]. 包装工程, 2022, 43(14): 29-36.
CHEN M, HUANG C, SU P P.Product Design Method Based on Affordance[J]. Packaging Engineering, 2022, 43(14): 29-36.
[37] 倪永波. 面向顾客需求演化的产品弹性评估与优化设计方法[D]. 徐州: 中国矿业大学, 2023.
NI Y B.Product Elasticity Evaluation and Optimization Design Method for Customer Demand Evolution[D]. Xuzhou: China University of Mining and Technology, 2023.
[38] 邓玮丹. 认知差异在设计草图-产品意象造型中的研究[D]. 天津: 天津科技大学, 2017.
DENG W D.Research on Cognitive Differences in Design Sketch-Product Image Modeling[D]. Tianjin: Tianjin University of Science & Technology, 2017.
[39] QAFFAS A A.Optimized back Propagation Neural Network Using Quasi-Oppositional Learning-Based African Vulture Optimization Algorithm for Data Fusion in Wireless Sensor Networks[J]. Sensors, 2023, 23(14): 6261.