基于GA-BP神经网络模型的二手载货汽车价值评估方法

doi:10.6040/j.issn.1671-9352.0.2024.243

《山东大学学报(理学版)》 ›› 2025, Vol. 60 ›› Issue (12): 11-20.doi: 10.6040/j.issn.1671-9352.0.2024.243

基于GA-BP神经网络模型的二手载货汽车价值评估方法

闫晟煜¹,刘杨¹,刘继祥²,陈海峰^3*,郑元旺⁴,温福华¹,王洪瑀⁵

1.长安大学汽车学院, 陕西西安 710018;2.中建八局第一建设有限公司, 山东济南 250100;3.中国汽车技术研究中心有限公司, 天津 300300;4. 山东广安车联科技股份有限公司, 山东济宁 272000;5.上汽通用五菱汽车股份有限公司, 重庆 401135

发布日期:2025-12-10
通讯作者: 陈海峰(1980— ),男,高级工程师,研究方向为汽车后市场、产业研究. E-mail:chenhaifeng@catarc.ac.cn
作者简介:闫晟煜(1987— ),男,副教授,博士,研究方向为智慧交通工程、公路运输规划. E-mail:Leo9574@163.com*通信作者:陈海峰(1980— ),男,高级工程师,研究方向为汽车后市场、产业研究. E-mail:chenhaifeng@catarc.ac.cn
基金资助:
国家重点研发计划项目(2023YFB3209803);长安大学中央高校基本科研业务费专项资金资助项目(300102224206);山东省科技型中小企业创新能力提升工程资助项目(2023TSGC0335)

Evaluation method of used truck value based on GA-BP neural network model

YAN Shengyu¹, LIU Yang¹, LIU Jixiang², CHEN Haifeng^3*, ZHENG Yuanwang⁴, WEN Fuhua¹, WANG Hongyu⁵

1. School of Automobile, Changan University, Xian 710018, Shaanxi, China;
2. The First Company of China Eighth Engineering Bureau Ltd., Jinan 250100, Shandong, China;
3. China Automotive Technology and Research Center Co., Ltd., Tianjin 300300, China;
4. Shandong Guangan Connected Vehicle Technology Co., Ltd., Jining 272000, Shandong, China;
5. SAIC-GM-Wuling Automobile Co., Ltd., Chongqing 401135, China

Published:2025-12-10

摘要/Abstract

摘要： 为准确评估二手载货汽车的价值,基于电商平台历史交易数据,提出遗传算法-反向传播神经网络(genetic algorithm-backpropagation neural network, GA-BP)评估方法。结合二手载货汽车的使用特性,选取影响交易价值的11个关键指标并提出指标的量化标准;通过皮尔逊相关系数法对各指标进行一致性检验,验证指标选取的合理性;融合遗传算法和BP神经网络,提出二手载货汽车GA-BP神经网络价值评估模型,利用9 016条交易数据对模型进行训练和验证。相较于未优化过的BP神经网络模型,本文模型的均方误差降低了74.85%。通过平均影响值分析发现,累计行驶里程、累计行驶时间、排放标准、比功率和上装情况5项价值评估指标对二手载货汽车价值评估影响最大,排放标准逐渐成为影响二手载货汽车的重要指标。

关键词: 汽车价值评估, 二手载货汽车, 遗传算法, BP神经网络, 排放标准

Abstract: To accurately evaluate the value of the used trucks, a GA-BP(genetic algorithm-backpropagation)neural network evaluation method is proposed, based on the historical transaction data of e-commerce platform. Considering the use features of the used trucks, 11 key indicators affecting the transaction value are selected and the quantitative standards of the indicators are put forward. The consistency of each index is tested by Pearson correlation coefficient method to verify the rationality of index selection. By compromising Genetic Algorithm and BP neural network, the evaluation model of the used trucks value by GA-BP neural network is proposed, and the model is trained and verified by using 9 016 transaction data. Compared with the unoptimized BP neural network model, the mean square error(MSE)of the model is reduced by 74.85%. Through the analysis of the average impact value, it is found that the cumulative mileage, cumulative driving time, emission standard, specific power and loading situation show the greatest influence on the value evaluation of used trucks, and the emission standard has gradually become an important indicator affecting used trucks.

Key words: automobile value evaluation, used truck, genetic algorithm, BP neural network, emission standard

中图分类号:

TP392

闫晟煜,刘杨,刘继祥,陈海峰,郑元旺,温福华,王洪瑀. 基于GA-BP神经网络模型的二手载货汽车价值评估方法[J]. 《山东大学学报(理学版)》, 2025, 60(12): 11-20.

YAN Shengyu, LIU Yang, LIU Jixiang, CHEN Haifeng, ZHENG Yuanwang, WEN Fuhua, WANG Hongyu. Evaluation method of used truck value based on GA-BP neural network model[J]. JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE), 2025, 60(12): 11-20.

参考文献

[1] 中华人民共和国商务部. 二手车鉴定评估技术规范:GB/T 30323—2013[S]. 北京: 中国标准出版社, 2013:1. Ministry of Commerce of the Peoples Republic of China. Technical specification for appraisal of used cars: GB/T 30323—2013[S]. Beijing: China standard Press, 2013:1.
[2] 王乾,冯强. 2019年中国二手车电商行业研究报告[J]. 互联网经济,2019,50(6):62-67. WAN Qian, FENG Qiang. China used car e-commerce industry research report in 2019[J]. Digital Economy, 2019, 50(6):62-67.
[3] 胡诣文,张天佑,张旭,等. 基于机器学习的二手车价格预测算法研究[J]. 信息技术与信息化,2022,271(10):52-55. HU Yiwen, ZHANG Tianyou, ZHANG Xu, et al. Research on used car price forecasting algorithm based on machine learning[J]. Information Technology and Informatization, 2022, 271(10):52-55.
[4] 王传杏,郑艳,沈易晨. 基于特征价格理论的二手车价值估计模型研究[J]. 时代汽车,2020,345(21):170-172. WANG Chuanxing, ZHENG Yan, SHEN Yichen. Research on value estimation model of used cars based on feature price theory[J]. Auto Time, 2020, 345(21):170-172.
[5] 代金辉,仲璇,王梦恩. 基于LightGBM和随机森林算法的二手车估价[J]. 高师理科学刊,2022,42(12):15-22. DAI Jinhui, ZHONG Xuan, WANG Mengen. Second-hand car valuation based on LightGBM and random forest algorithm[J]. Journal of Science of Teachers College and University, 2022, 42(12):15-22.
[6] PRIETO M, CAEMMERER B, BALTAS G. Using a hedonic price model to test prospect theory assertions: the asymmetrical and nonlinear effect of reliability on used car prices[J]. Journal of Retailing and Consumer Services, 2015, 22(1):206-212.
[7] SYAHPUTRA I, ZARLIS M, SUTARMAN. Analysis of the application of fuzzy logic and Levenbergmarquardt in the calculation of used car prices[J]. Journal of Physics Conference Series, 2020, 1566(1):012106.
[8] ALEXSTAN A S J, MONESH K M, POONKODI M, et al. Used car price prediction using machine learning[J]. Advances in Science and Technology, 2023, 6630(3):512-517.
[9] 杨致远. 基于PCA-DNN和LightGBM的二手车价格预测[J]. 信息与电脑,2022,34(21):73-75. YANG Zhiyuan. Prediction of the transaction price of used car based on PCA-DNN and LightGBM[J]. Information & Computer, 2022, 34(21):73-75.
[10] 李富强,彭海丽,杨熙,等. 基于深度学习的二手车价格预测模型及影响分析[J]. 汽车工程学报,2021,11(5):379-385. LI Fuqiang, PENG Haili, YANG Xi, et al. Prediction modelling of second-hand car price based on deep learning and influence factors analysis[J]. Chinese Journal of Automotive Engineering, 2021, 11(5):379-385.
[11] DEDY S, ALFIAN T, DONNY B. A data mining approach to predicting the inventory day of used cars[J].International Journal of Knowledge Engineering and Data Mining, 2021, 7(1-2):127-144.
[12] 张远森. 基于神经网络的二手车价格评估模型[D]. 天津:天津大学,2019:46-47. ZHANG Yuansen. A used cars price forecasting model based on artificial neural network[D]. Tianjin: Tianjin University, 2019.
[13] FU Yuan, CHEN Xiang, LIU Yu, et al. Multi-sensor GA-BP Algorithm based gearbox fault diagnosis[J]. Applied Sciences, 2022, 12(15):7535.
[14] LIU J, ASHRAF M A. Face recognition method based on GA-BP neural network algorithm [J]. Open Physics, 2018, 16(1):1056-1065.
[15] ZOU Meng, XUE Long, GAI Hongjian, et al. Identification of the shear parameters for lunar regolith based on a GA-BP neural network[J]. Journal of Terramechanics, 2020, 89(2):21-29.
[16] 曹镓玺,王鑫,雷光春. 基于遗传算法优化BP神经网络的青藏高原海北高寒湿地CO₂通量模拟及其影响因子[J]. 山东大学学报(理学版),2021,56(5):33-41. CAO Jiaxi, WANG Xin, LEI Guangchun. Simulation of alpine wetlands CO₂ flux and its influencing factors based on BP neural network optimized by genetic algorithm in Qinghai-Tibet Plateau[J]. Journal of Shandong University(Natural Science), 2021, 56(5):33-41.
[17] LIU Junbiao, JIN Xinyu, DONG Fang, et al. Fading channel modelling using single-hidden layer feedforward neural networks[J]. Multidimensional Systems and Signal Processing, 2016, 28(1):885-903.
[18] 王静红,吴芝冰,黄鹏,等. 基于元路径属性融合的异质网络表示学习[J]. 山东大学学报(理学版),2024,59(3):1-13. WANG Jinghong, WU Zhibing, HUANG Peng, et al. Heterogeneous network representation learning based on metapath attribute fusion[J]. Journal of Shandong University(Natural Science), 2024, 59(3):1-13.

多维度评价

Viewed

Full text

Abstract

Cited

Shared

Discussed

基于GA-BP神经网络模型的二手载货汽车价值评估方法

Evaluation method of used truck value based on GA-BP neural network model

PDF (PC)

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

多维度评价

本文评价

推荐阅读 10

[1]	李侦瑷,韦慧,陈馨. 基于双目标的MNSGA-Ⅱ算法求解非线性方程组[J]. 《山东大学学报(理学版)》, 2024, 59(10): 22-29.
[2]	王晓,刘重阳,胡电中,刘刚. 1,3-丙二醇间歇发酵中的时滞最优控制[J]. 《山东大学学报(理学版)》, 2024, 59(1): 124-131, 138.
[3]	曹镓玺,王鑫,雷光春. 基于遗传算法优化BP神经网络的青藏高原海北高寒湿地CO₂通量模拟及其影响因子[J]. 《山东大学学报(理学版)》, 2021, 56(5): 33-41.
[4]	张晶, 肖智斌, 容会, 崔毅. 改进型遗传算法在网络蜘蛛上的应用[J]. 山东大学学报（理学版）, 2015, 50(05): 1-6.
[5]	杜晓军,林柏钢,林志远,李应. 安全软件模糊测试中多种群遗传算法的研究[J]. J4, 2013, 48(7): 79-84.
[6]	孙飞,汪鹏君*,俞海珍,汪迪生. 基于遗传算法的三值FPRM电路面积优化[J]. J4, 2013, 48(05): 51-56.
[7]	马宇红1,2,孙淑芬2. 一个带中转和直销的多产品运输问题及其遗传算法[J]. J4, 2012, 47(7): 121-126.
[8]	吴大华,何振峰*. 对基于聚类和遗传算法的时间序列分割算法的改进[J]. J4, 2010, 45(7): 45-49.
[9]	许民利孙彩群. 基于等待时间限制的服务备件多点转运库存模型研究[J]. J4, 2010, 45(3): 61-65.
[10]	丁然李歧强梁涛. 具有分解结构的多目的批处理过程短期调度模型[J]. J4, 2010, 45(1): 73-79.
[11]	刘冰陆玮洁杨国生. 遗传算法在烷基硝基苯酚类化合物的QSRR中的应用[J]. J4, 2009, 44(9): 8-11.
[12]	. 基于遗传算法的带时间窗邮政车辆路径问题研究[J]. J4, 2009, 44(6): 46-50.
[13]	. 基于移相法的三维面型测量系统优化算法研究[J]. J4, 2009, 44(6): 40-45.
[14]	何爱香,张勇 . 基于遗传算法和决策树的肿瘤分类规则挖掘[J]. J4, 2007, 42(9): 91-95 .
[15]	张晓光,李琰,王海洋 . 一种基于遗传算法QoS敏感的Web服务组合方法[J]. J4, 2007, 42(9): 56-61 .