基于优化神经网络的沥青混合料力学特性预估

doi:10.3976/j.issn.1002-4026.20230149

摘要/Abstract

摘要：

现有的沥青混合料疲劳寿命预估大多基于传统的疲劳方程拟合得到,但由于路面结构的多向性和材料的复杂性,其预测精度往往不尽人意。为了提高预测精度,在遗传算法的基础上对神经网络架构进行优化,通过室内间接拉伸试验建立了沥青混合料强度及疲劳寿命预估模型,并对预估模型的精度进行了验证。试验结果表明,采用遗传算法优化的神经网络用于预测沥青混合料疲劳力学特性精度误差在4%以内,远优于传统的疲劳预测方程,可以作为获取沥青混合料疲劳特性研究数据的一种有效方法。

关键词: 交通工程, 沥青混合料, 深度学习模型, 强度预测, 疲劳寿命预测

Abstract:

The existing fatigue life prediction of asphalt mixtures is mostly based on traditional fatigue equation fitting; however, due to the multidirectionality of pavement structure and the complexity of materials, the prediction accuracy is often not satisfactory. Therefore, this article establishes an optimized neural network-based model for predicting the strength and fatigue life of asphalt mixtures using indoor indirect tensile tests and verifies the accuracy of the prediction model. The experimental results show that the accuracy of Genetic Algorithm-Back Propagation neural network to predict the fatigue mechanical properties for asphalt mixture is within 4%, which is far superior to traditional fatigue prediction equations and can be used as an effective method to obtain data on the fatigue characteristics of asphalt mixtures.

Key words: traffic engineering, asphalt mixture, deep-learning model, strength prediction, fatigue life prediction

中图分类号:

U411

王晓阳, 万晨光, 王笑风. 基于优化神经网络的沥青混合料力学特性预估[J]. 山东科学, 2024, 37(4): 56-64.

WANG Xiaoyang, WAN Chenguang, WANG Xiaofeng. Prediction of mechanical properties of asphalt mixtures based on optimized neural networks[J]. Shandong Science, 2024, 37(4): 56-64.

图/表 11

图1

表1

表2

表3

图2

表4

表5

图3

图4

表6

表7

参考文献 21

[1]	吕松涛, 刘超超, 屈芳婷, 等. 沥青混合料疲劳性能试验与表征方法综述[J]. 中国公路学报, 2020, 33(10): 67-75. DOI: 10.19721/j.cnki.1001-7372.2020.10.002.
[2]	沈钱超, 吴帮伟, 沈燕, 等. 应力控制模式下沥青混合料疲劳损伤力学性能研究[J]. 吉林建筑大学学报, 2020, 37(1): 38-44.
[3]	DAS B P, BABAR P D, SIDDAGANGAIAH A K. Impact of aging and moisture on fatigue life of asphalt mixture[J]. Advances in Civil Engineering Materials, 2021, 10(1): 20190211. DOI: 10.1520/acem20190211.
[4]	SUN Y, HE D P, LI J. Research on the fatigue life prediction for a new modified asphalt mixture of a support vector machine based on particle swarm optimization[J]. Applied Sciences, 2021, 11(24): 11867. DOI: 10.3390/app112411867.
[5]	王淋, 郭乃胜, 温彦凯, 等. 几种改性沥青疲劳破坏评价指标及性能研究[J]. 土木工程学报, 2020, 53(1): 118-128. DOI: 10.15951/j.tmgcxb.2020.01.012.
[6]	单丽岩, 谭忆秋, 许亚男, 等. 应力、应变控制模式下沥青疲劳损伤演化规律[J]. 中国公路学报, 2016, 29(1): 16-21. DOI: 10.19721/j.cnki.1001-7372.2016.01.002.
[7]	徐骁龙, 叶奋, 宋卿卿, 等. 沥青疲劳评价指标试验研究[J]. 华东交通大学学报, 2014, 31(2): 14-19. DOI: 10.16749/j.cnki.jecjtu.2014.02.008.
[8]	LI Q A, MENG Y P, LI N H, et al. Characterization of fatigue performance of the warm-mix recycled asphalt mixture using different models[J]. Fatigue & Fracture of Engineering Materials & Structures, 2022, 45(3): 770-782. DOI: 10.1111/ffe.13632.
[9]	KAMARUDIN S N N, HAININ M R, WARID M N M, et al. The usage of treated plastic as additive to improve the asphalt mixture’s performance by using dry mix method[J]. Key Engineering Materials, 2021, 879: 126-135. DOI: 10.4028/www.scientific.net/kem.879.126.
[10]	董玲云, 何兆益, 黄刚, 等. 热再生沥青混合料的疲劳特性研究[J]. 重庆交通大学学报(自然科学版), 2013, 32(4): 606-609. DOI: 10.3969/j.issn.1674-0696.2013.04.14.
[11]	吕松涛, 郑健龙. 基于耗散能理论的老化沥青混合料疲劳方程[J]. 中外公路, 2012, 32(3): 284-286. DOI: 10.14048/j.issn.1671-2579.2012.03.016.
[12]	孟勇军, 张肖宁. 基于累计耗散能量比的改性沥青疲劳性能[J]. 华南理工大学学报(自然科学版), 2012, 40(2): 99-103. DOI: 10.3969/j.issn.1000-565X.2012.02.018.
[13]	吴志勇, 张肖宁, 游宏, 等. 基于应变控制的沥青混合料疲劳寿命预测[J]. 华南理工大学学报(自然科学版), 2014, 42(2): 139-144. DOI: 10.3969/j.issn.1000-565X.2014.02.021.
[14]	LV S T, TAN L, PENG X H, et al. Fatigue resistance design of rubberized asphalt mixture pavement under three-dimensional stress state[J]. Construction and Building Materials, 2021, 307: 125138. DOI: 10.1016/j.conbuildmat.2021.125138.
[15]	AHMED T M, GREEN P L, KHALID H A. Predicting fatigue performance of hot mix asphalt using artificial neural networks[J]. Road Materials and Pavement Design, 2017, 18(S2): 141-154.
[16]	JIA H C, CHEN H X, SHENG Y P, et al. Effect of laboratory aging on the stiffness and fatigue cracking of asphalt mixture containing bamboo fiber[J]. Journal of Cleaner Production, 2022, 333: 130120. DOI: 10.1016/j.jclepro.2021.130120.
[17]	王晓阳. 考虑拉压模量差异的沥青混合料疲劳损伤特性研究[D]. 长沙: 长沙理工大学, 2017.
[18]	陈建荣, 叶俊, 吴逢春, 等. 短切玄武岩纤维沥青混合料疲劳性能研究[J]. 公路, 2013, 58(11): 188-191. DOI: 10.3969/j.issn.0451-0712.2013.11.040.
[19]	李亦鹏. 不同应力状态下沥青混合料模量与强度特性研究[D]. 长沙: 长沙理工大学, 2017.
[20]	谢春磊, 张勇, 耿红斌, 等. 基于BP神经网络的沥青混合料疲劳性能预测模型[J]. 重庆交通大学学报(自然科学版), 2018, 37(2): 35-40. DOI: 10.3969/j.issn.1674-0696.2018.02.06.
[21]	闫楚良, 郝云霄, 刘克格. 基于遗传算法优化的BP神经网络的材料疲劳寿命预测[J]. 吉林大学学报(工学版), 2014, 44(6): 1710-1715. DOI: 10.13229/j.cnki.jdxbgxb201406027.

技术指标	技术要求	检验结果
针入度(25 ℃,100 g,5 s)/10^-1 mm	40~60	55
延度(5 cm/min,15 ℃)/cm	≥20	33
软化点(环球法)/℃	≥60	75.5
密度(15 ℃)/(g·cm^-3)	―	1.02

筛孔尺寸/mm	通过率/%	筛孔尺寸/mm	通过率/%	筛孔尺寸/mm	通过率/%
19.000	100.0	4.750	37.0	0.300	7.7
16.000	90.4	2.360	24.2	0.150	6.1
13.200	79.5	1.180	18.0	0.075	4.3
9.500	63.5	0.600	11.4

序号	温度/℃	加载速率/ (MPa·s^-1)	间接拉伸强度/MPa			间接拉伸强度均值/MPa
序号	温度/℃	加载速率/ (MPa·s^-1)	1	2	3	间接拉伸强度均值/MPa
1	10	0.02	0.521	0.762	0.665	0.649
2		0.05	0.844	0.811	0.825	0.827
3		0.10	0.850	1.105	0.875	0.943
4		0.20	1.420	1.126	1.334	1.293
5		0.50	1.785	1.552	1.778	1.705
6		1.00	1.692	1.892	1.624	1.736
7		2.00	2.356	1.994	2.234	2.195
8	15	0.02	0.523	0.496	0.568	0.529
9		0.05	0.682	0.653	0.642	0.659
10		0.10	0.763	0.793	0.924	0.827
11		0.20	1.056	0.905	0.928	0.963
12		0.50	1.214	0.998	1.036	1.083
13		1.00	1.562	1.426	1.386	1.458
14		2.00	1.588	1.695	1.458	1.580
15	20	0.02	0.467	0.436	0.502	0.468
16		0.05	0.599	0.582	0.572	0.584
17		0.10	0.688	0.653	0.699	0.680
18		0.20	0.836	0.721	0.734	0.764
19		0.50	1.052	0.968	0.853	0.958
20		1.00	1.153	1.010	1.005	1.056
21		2.00	1.312	1.201	1.223	1.245

温度/℃	α	β	R²
10	1.838 9	0.256 0	0.967
15	1.353 2	0.233 6	0.974
20	1.066 5	0.206 2	0.998

应力比	应力水平/MPa	疲劳寿命/次				疲劳寿命均值/次
应力比	应力水平/MPa	1	2	3	4	疲劳寿命均值/次
0.2	0.141	218 863	169 532	225 846	159 635	193 469
0.3	0.194	25 843	26 541	27 896	30 262	27 635
0.4	0.244	4 415	4 682	4 632	4 589	18 318
0.5	0.292	889	796	938	902	879