流态地聚物固化土强度特性及其强度预测

易富; 姜珊; 慕德慧; 管茂成

doi:10.16030/j.cnki.issn.1000-3665.202205038

流态地聚物固化土强度特性及其强度预测

doi: 10.16030/j.cnki.issn.1000-3665.202205038

易富^1,,
姜珊^2, ,,
慕德慧³,
管茂成²

1.
辽宁工程技术大学建筑与交通学院，辽宁阜新　123000
2.
辽宁工程技术大学土木工程学院，辽宁阜新　123000
3.
长春高新建设开发有限公司，吉林长春　130015

基金项目: 国家自然科学基金项目（51774163）；辽宁省教育厅青年基金项目（LJKQZ2021153）；辽宁省教育厅科学研究一般项目（LJ2020JCL037）

详细信息

作者简介:
易富（1978-），男，博士，教授，博士生导师，主要从事环境岩土工程研究工作。E-mail：yifu9716@163.com

通讯作者:
姜珊（1997-），女，硕士研究生，从事固化土力学特性研究。E-mail：13188009871@163.com

中图分类号: TU44
计量
- 文章访问数: 281
- HTML全文浏览量: 207
- PDF下载量: 111
- 被引次数: 0
出版历程
- 收稿日期: 2022-05-14
- 录用日期: 2022-09-26
- 修回日期: 2022-08-15
- 网络出版日期: 2022-12-06
- 刊出日期: 2023-01-13

Strength characteristics and strength prediction of fluid geopolymer solidified soil

YI Fu^1
,,
JIANG Shan^{2
, ,},
MU Dehui³,
GUAN Maocheng²

1.
College of Architecture and Transportation, Liaoning Technical University, Fuxin, Liaoning　123000, China
2.
College of Civil Engineering, Liaoning Technical University, Fuxin, Liaoning　123000, China
3.
Changchun High Tech Construction and Development Co. Ltd., Changchun, Jilin　130015, China

摘要

摘要: 地聚物胶凝材料能够替代水泥基胶凝材料作为固化剂应用于狭窄肥槽回填等工程问题中，有效降低水泥生产过程中的污染及能耗，但目前对于流态地聚物固化土胶凝材料的研究较少。采用3种新型绿色胶凝材料联合碱激发剂固化工程渣土形成流态地聚物固化土，通过对比其无侧限抗压强度，探究每种胶凝材料对于固化土强度特性的影响，同时建立强度预测模型，分析不同因素对于强度的影响程度。研究结果表明：固化土的强度随着碱激发剂模数的增加先提高后降低；固化土强度随着高炉矿渣（GGBS）、粉煤灰、稻壳灰掺量的增加均呈上升趋势，随着稻壳灰粒径的增长呈下降趋势；碱激发剂模数增至1.2、GGBS掺量增至10%、粉煤灰掺量增至8%和稻壳灰掺量增至11%时，固化土强度提升最为显著；强度预测模型预测结果的平均相对误差仅为5.57%，预测结果较为精准；预测模型中各层权值的计算结果表明养护龄期对于固化土强度影响最大，稻壳灰粒径影响程度最小。研究结果可以为固化土在实际工程的应用提供理论支持。
- 流态固化土 /
- 地聚物 /
- 无侧限抗压强度 /
- 强度预测模型 /
- 权重分析
Abstract: Geopolymer cementitious materials can replace cement-based cementitious materials as curing agents in engineering problems, such as backfilling of narrow fertilizer troughs, and effectively reduce pollution and energy consumption in the cement production process. There are few studies on cementitious materials. Three new green cementitious materials combined with alkali activators are used to solidify engineering slag and form fluidized geopolymer-solidified soil. The strength prediction model is established to analyze the influence of different factors on the strength. The results show that the strength of the solidified soil increases first and then decreases with the increasing modulus of the alkali activator, increases with the content of GGBS, fly ash and rice husk ash, and decreases with the increasing particle size. When the modulus of alkali activator increases to 1.2, the content of GGBS increases to 10%, the content of fly ash increases to 8%, and the content of rice husk ash increases to 11%, the strength of the solidified soil increases significantly. The average relative error of the prediction results of the strength prediction model is only 5.57%, which is relatively accurate for the solidified soil. The calculation results of the weights of each layer in the prediction model show that the curing age has the greatest impact on the strength of the solidified soil, and the particle size of rice husk ash has the minimal impact. The research results can provide theoretical support for the application of solidified soil in practical engineering.
- fluid solidified soil /
- geopolymer /
- unconfined compressive strength /
- strength prediction model /
- weight analysis

HTML全文

图 1 试样破坏形式

Figure 1. Failure form of sample

下载: 全尺寸图片幻灯片

图 2 不同碱激发剂模数固化土抗压强度曲线图

Figure 2. Compression strength curves of modulus solidified soil with different alkali activators

下载: 全尺寸图片幻灯片

图 3 不同GGBS掺量固化土抗压强度曲线图

Figure 3. Compressive strength curve of solidified soil with different GGBS content

下载: 全尺寸图片幻灯片

图 4 不同粉煤灰掺量固化土抗压强度曲线图

Figure 4. Compressive strength curve of solidified soil with different fly ash content

下载: 全尺寸图片幻灯片

图 5 不同稻壳灰掺量固化土抗压强度曲线图

Figure 5. Compressive strength curve of solidified soil with different amount of rice husk ash

下载: 全尺寸图片幻灯片

图 6 不同稻壳灰粒径固化土抗压强度曲线图

Figure 6. Compressive strength curve of solidified soil with different rice husk ash particle size

下载: 全尺寸图片幻灯片

图 7 强度预测模型拓扑结构

Figure 7. Topology of strength prediction model

下载: 全尺寸图片幻灯片

图 8 预测值与实际值对比图

Figure 8. Comparison between predicted value and actual value

下载: 全尺寸图片幻灯片

表 1 粉煤灰、GGBS和稻壳灰的化学组成

Table 1. Chemical composition of fly ash, GGBS and rice husk ash

材料	w（SiO₂）/%	w（Al₂O₃）/%	w（Fe₂O₃）/%	w（MgO）/%	w（CaO）/%	w（Na₂O）/%	w（SO₃）/%	w（K₂O）/%
粉煤灰	63.34	27.00	2.00	1.00	3.00	1.11	1.10	1.05
GGBS	35.41	20.24	0.18	8.16	31.64	1.36	1.79	0.29
稻壳灰	84.00	1.35	1.45	—	3.17	—	0.93	—
注：“—”表示不含此成分或含量极低。

下载: 导出CSV

表 2 流态地聚物固化土设计方案

Table 2. Design scheme of fluid geopolymer Solidified Soil

试验编号	GGBS 掺量/%	粉煤灰掺量/%	碱激发剂模数/（mol·L^–1）	稻壳灰掺量/%	稻壳灰粒径/mm
GF1	8	8	1.2	0	—
GF2	10	8	1.2	0	—
GF3	12	8	1.2	0	—
GF4	14	8	1.2	0	—
GF5	10	6	1.2	0	—
GF6	10	10	1.2	0	—
GF7	10	12	1.2	0	—
GF8	10	8	0.6	0	—
GF9	10	8	0.9	0	—
GF10	10	8	1.5	0	—
GFD1	10	8	1.2	5	1.2
GFD2	10	8	1.2	8	1.2
GFD3	10	8	1.2	11	1.2
GFD4	10	8	1.2	14	1.2
GFD5	10	8	1.2	11	0.6
GFD6	10	8	1.2	11	0.3
GFD7	10	8	1.2	11	0.15
GFD8	10	8	1.2	11	0.075

下载: 导出CSV

表 3 不同隐含层层数于节点数的预测模型对比

Table 3. Comparison of prediction models with different hidden layers and nodes

隐含层层数	隐含层节点数	相关系数	均方误差
1	4	0.810 73	0.004 66
1	6	0.838 99	0.004 09
1	8	0.845 78	0.003 73
1	10	0.744 04	0.004 17
1	12	0.765 88	0.004 06
2	8、4	0.965 43	0.000 80
2	8、6	0.997 70	0.000 37
2	8、8	0.999 43	0.000 08
2	8、10	0.997 84	0.000 32

下载: 导出CSV

表 4 测试样本误差分析表

Table 4. Error analysis of test samples

编号	GGBS掺量/%	粉煤灰掺量/%	碱激发模数	稻壳灰掺量/%	稻壳灰粒径/mm	养护龄期/d	预测值/MPa	实际值/MPa	绝对误差/MPa	相对误差/%
1	12	8	1.2	0	0	3	0.380 5	0.36	0.020 4	5.68
2	10	8	1.2	0	0	3	0.296 8	0.27	0.026 7	9.91
3	10	8	0.6	0	0	28	1.268 8	1.26	0.008 8	0.69
4	10	8	1.2	0	0	28	1.474 5	1.57	0.095 5	6.08
5	10	8	1.2	11	0.6	7	1.378 7	1.44	0.061 3	4.25
6	10	8	1.2	14	1	14	2.262 9	1.93	0.332 9	17.25
7	10	8	1.2	11	0.075	7	1.849 0	1.86	0.010 9	0.58
8	10	10	1.2	0	0	14	1.018 5	1.02	0.001 4	0.14

下载: 导出CSV

表 5 预测模型训练与预测样本相对误差分布

Table 5. Relative error distribution of BP neural network training and prediction samples

相对误差分布范围	训练样本	预测样本	样本总数	占比/%
>20%	5	0	5	5.68
(10%, 20%]	6	1	7	7.95
(1%, 10%]	25	4	29	32.96
≤1%	44	3	47	53.41
总计	80	8	88	100

下载: 导出CSV

表 6 预测模型权重贡献与权重贡献率

Table 6. weight contribution rate of prediction model

影响因素	权重贡献	权重贡献率/%（排名）
GGBS掺量	0.501 169	6.92（5）
粉煤灰掺量	2.183 732	30.15（2）
碱激发剂模数	0.586 749	8.10（4）
稻壳灰掺量	1.197 763	16.54（3）
稻壳灰粒径	0.021 158	0.29（6）
养护龄期	2.752 821	38.00（1）
总计	7.243 328	100

下载: 导出CSV

参考文献(26)

[1]	DU Yanjun,YU Bowei,LIU Kai,et al. Physical,hydraulic,and mechanical properties of clayey soil stabilized by lightweight alkali-activated slag geopolymer[J]. Journal of Materials in Civil Engineering,2017,29(2):04016217. doi: 10.1061/(ASCE)MT.1943-5533.0001743
[2]	PHUMMIPHAN I,HORPIBULSUK S,RACHAN R,et al. High calcium fly ash geopolymer stabilized lateritic soil and granulated blast furnace slag blends as a pavement base material[J]. Journal of Hazardous Materials,2018,341:257 − 267. doi: 10.1016/j.jhazmat.2017.07.067
[3]	俞家人,陈永辉,陈庚,等. 地聚物固化软黏土的力学特征及机理分析[J]. 建筑材料学报,2020,23(2):364 − 371. [YU Jiaren,CHEN Yonghui,CHEN Geng,et al. Mechanical behaviour of geopolymer stabilized clay and its mechanism[J]. Journal of Building Materials,2020,23(2):364 − 371. (in Chinese with English abstract)
[4]	PELISSER F,GUERRINO E L,MENGER M,et al. Micromechanical characterization of metakaolin-based geopolymer[J]. Construction and Building Materials,2013,49:547 − 553. doi: 10.1016/j.conbuildmat.2013.08.081
[5]	THAARRINI J,VENKATASUBRAMANI R. Feasibility studies on compressive strength of ground coal ash geopolymer mortar[J]. Periodica Polytechnica Civil Engineering,2015,59(3):373 − 379. doi: 10.3311/PPci.7696
[6]	王东星, 王宏伟, 王瑞红. 活性MgO–粉煤灰固化淤泥微观机制研究[J]. 岩石力学与工程学报, 2019, 38（增刊2）: 3717 − 3725 WANG Dongxing, WANG Hongwei, WANG Ruihong, Micro-mechanisms of dredged sludge solidified with reactive MgO-fly ash[J]. Chinese Journal of Rock Mechanics and Engineering, 2019（Sup 2）: 3717 − 3725. （in Chinese with English abstract）
[7]	王东星,王宏伟,肖杰,等. 活性MgO-粉煤灰软土固化材料强度与机理研究[J]. 中国矿业大学学报,2018,47(4):879 − 884. [WANG Dongxing,WANG Hongwei,XIAO Jie,et al. Strength and micromechanism of reactive MgO-activated fly ash as an alternative soft soil stabilizer[J]. Journal of China University of Mining & Technology,2018,47(4):879 − 884. (in Chinese with English abstract) doi: 10.13247/j.cnki.jcumt.000899
[8]	王东星,王宏伟,邹维列,等. 碱激发粉煤灰固化淤泥微观机制研究[J]. 岩石力学与工程学报,2019,38(增刊1):3197 − 3205. [WANG Dongxing,WANG Hongwei,ZOU Weilie,et al. Research on micro-mechanisms of dredged sludge solidified with alkali-activated fly ash[J]. Chinese Journal of Rock Mechanics and Engineering,2019,38(Sup1):3197 − 3205. (in Chinese with English abstract)
[9]	王东星, 何福金. CO₂碳化–矿渣/粉煤灰协同固化土效果与机制研究[J]. 岩石力学与工程学报, 2020, 39（7）: 1493 − 1502 WANG Dongxing, HE Fujin. Investigation on performance and mechanism of CO₂ carbonated slag/fly ash solidified soils[J]. Chinese Journal of Rock Mechanics and Engineering. 2020, 39（7）: 1493 − 1502. （in Chinese with English abstract）
[10]	陈忠清,朱泽威,吕越. 粉煤灰基地聚物加固土的强度及抗冻融性能试验研究[J]. 水文地质工程地质,2022,49(4):100 − 108. [CHEN Zhongqing,ZHU Zewei,LYV Yue. Laboratory investigation on the strength and freezing-thawing resistance of fly ash based geopolymer stabilized soil[J]. Hydrogeology & Engineering Geology,2022,49(4):100 − 108. (in Chinese with English abstract) doi: 10.16030/j.cnki.issn.1000-3665.202111045
[11]	贾栋钦,裴向军,张晓超,等. 改性糯米灰浆固化黄土的微观机理试验研究[J]. 水文地质工程地质,2019,46(6):90 − 96. [JIA Dongqin,PEI Xiangjun,ZHANG Xiaochao,et al. A test study of the microscopic mechanism of modified glutinous rice mortar solidified loess[J]. Hydrogeology & Engineering Geology,2019,46(6):90 − 96. (in Chinese with English abstract) doi: 10.16030/j.cnki.issn.1000-3665.2019.06.12
[12]	陈伟,乐绍林,高文波,等. 海相疏浚淤泥流动固化的作用机制和微观结构分析[J]. 岩石力学与工程学报,2020,39(增刊1):3114 − 3122. [CHEN Wei,YUE Shaolin,GAO Wenbo,et al. Solidification mechanism and microstructural investigations on flow-solidified marine dredged sludge[J]. Chinese Journal of Rock Mechanics and Engineering,2020,39(Sup1):3114 − 3122. (in Chinese with English abstract) doi: 10.13722/j.cnki.jrme.2019.0528
[13]	何俊,栗志翔,石小康,等. 侵蚀环境中碱渣-矿渣固化淤泥的力学性质[J]. 水文地质工程地质,2019,46(6):83 − 89. [HE Jun,LI Zhixiang,SHI Xiaokang,et al. Mechanical properties of the soft soil stabilized with soda residue and ground granulated blast furnace slag under the erosion environment[J]. Hydrogeology & Engineering Geology,2019,46(6):83 − 89. (in Chinese with English abstract) doi: 10.16030/j.cnki.issn.1000-3665.2019.06.11
[14]	HE Jun, WANG Xiaoqi, SU Ying, et al. Shear strength of stabilized clay treated with soda residue and ground granulated blast furnace slag[J]. Journal of Materials in Civil Engineering, 2019, 31（3）.
[15]	LAKSHMI S M,GEETHA S,SELVAKUMAR M,et al. Application of lime and GGBS to improve the strength of clayey sand[J]. IOP Conference Series Materials Science and Engineering,2020,989:12 − 28.
[16]	刘猛,戚红雨,王荆宁,等. 基于神经网络算法的智能抗干扰系统设计[J]. 计算机测量与控制,2018,26(10):155 − 159. [LIU Meng,QI Hongyu,WANG Jingning,et al. Design of intelligent anti-jamming system based on neural network algorithm[J]. Computer Measurement & Control,2018,26(10):155 − 159. (in Chinese with English abstract) doi: 10.16526/j.cnki.11-4762/tp.2018.10.034
[17]	王玉振. 基于EMD的小波神经网络模型预测大坝变形[J]. 水力发电,2018,44(8):101 − 104. [WANG Yuzhen. Prediction of dam deformation based on EMD neural network model[J]. Water Power,2018,44(8):101 − 104. (in Chinese with English abstract) doi: 10.3969/j.issn.0559-9342.2018.08.027
[18]	江显群,陈武奋. BP神经网络与GA-BP农作物需水量预测模型对比[J]. 排灌机械工程学报,2018,36(8):762 − 766. [JIANG Xianqun,CHEN Wufen. Comparison between BP neural network and GA-BP crop water demand forecasting model[J]. Journal of Drainage and Irrigation Machinery Engineering,2018,36(8):762 − 766. (in Chinese with English abstract)
[19]	王锦力,殷志祥,周明伟. 免振捣粉煤灰混凝土抗压强度的神经网络预测[J]. 辽宁工程技术大学学报,2006,25(增刊 1):131 − 132. [WANG Jinli,YIN Zhixiang,ZHOU Mingwei. Neural network prediction of compressive strength for self-compaction concrete of coal ash[J]. Journal of Liaoning Technical University,2006,25(Sup 1):131 − 132. (in Chinese with English abstract)
[20]	刘婵娟. 神经网络模型对拔出法检测超高强混凝土强度评定研究[D]. 长沙: 湖南大学, 2017 LIU Chanjuan. Strength prediction research on pullout method testing the strength of ultra-high strength concrete by neural network model[D]. Changsha: Hu’nan University, 2017. （in Chinese with English abstract）
[21]	张龙元,杨春峰,王培竹. BP神经网络模型在橡胶集料混凝土冻融循环后力学性能研究中的应用[J]. 住宅与房地产,2017(29):13 − 13. [ZHANG Longyuan,YANG Chunfeng,WANG Peizhu. Application of BP neural network model in the study of mechanical properties of rubber aggregate concrete after freeze-thaw cycle[J]. Housing and Real Estate,2017(29):13 − 13. (in Chinese)
[22]	赵明亮,水中和,周华新,等. 中低强度等级混凝土抗压强度的BP神经网络模型预测研究[J]. 混凝土,2021(3):35 − 38. [ZHAO Mingliang,SHUI Zhonghe,ZHOU Huaxin,et al. Prediction of compressive strength of medium and low strength grade concrete by BP neural network model[J]. Concrete,2021(3):35 − 38. (in Chinese with English abstract) doi: 10.3969/j.issn.1002-3550.2021.03.009
[23]	李扬,王伯昕,陈冬昕,等. 基于BP神经网络预测复合盐侵蚀后混凝土的相对动弹性模量[J]. 混凝土,2018(7):21 − 23. [LI Yang,WANG Boxin,CHEN Dongxin,et al. Prediction of relative dynamic elastic modulus of concrete after erosion based on BP neutral network theory[J]. Concrete,2018(7):21 − 23. (in Chinese with English abstract) doi: 10.3969/j.issn.1002-3550.2018.07.006
[24]	张伟,刘晓强,李顺群,等. 天津临港疏浚土固化特性及强度预测分析[J]. 水利水电技术,2020,51(4):20 − 26. [ZHANG Wei,LIU Xiaoqiang,LI Shunqun,et al. Analysis on prediction of solidifying characteristics and strength of dredged soil in port-vicinity area of Tianjin[J]. Water Resources and Hydropower Engineering,2020,51(4):20 − 26. (in Chinese with English abstract)
[25]	路晓宇. 大连海相软土固化强度试验研究及其神经网络预测[D]. 大连: 大连理工大学, 2021 LU Xiaoyu. Experimental study on solidification strength of Dalian marine soft soil and its neural network prediction[D]. Dalian: Dalian University of Technology, 2021. （in Chinese with English abstract）
[26]	童国庆,张吾渝,高义婷,等. 碱激发粉煤灰地聚物的力学性能及微观机制研究[J]. 材料导报,2022,36(4):129 − 134. [TONG Guoqing,ZHANG Wuyu,GAO Yiting,et al. Mechanical properties and micromechanism of alkali-activated fly ash geopolymer[J]. Materials Reports,2022,36(4):129 − 134. (in Chinese with English abstract)