硅酸盐学报

基于去噪扩散概率模型的混凝土配合比数据增强方法

PDF下载

梁涛 ¹ , 李悦 ¹ , 杨斌 ¹ , 刘江林 ¹ , 李亚强 ² , 梁美洁 ¹

硅酸盐学报 | 第二十八届中国科协年会智能建造关键无机非金属材料–AI驱动水泥基材料创新专题-研究论文 2026,54(10): 3499-3517

收起

硅酸盐学报 |第二十八届中国科协年会智能建造关键无机非金属材料–AI驱动水泥基材料创新专题-研究论文 2026 , 54 (10) : 3499 -3517

基于去噪扩散概率模型的混凝土配合比数据增强方法

全屏

梁涛¹, 李悦¹, 杨斌¹, 刘江林¹, 李亚强², 梁美洁¹

作者信息

^1.北京工业大学城市与工程安全减灾教育部重点实验室，北京　100124

^2.北京林业大学水土保持学院，北京　100083

通讯作者:

李悦（1972-），男，博士，教授。

作者简介:

梁涛（1997-），男，博士研究生。

A Concrete Mix Proportion Data Augmentation Method Based on Denoising Diffusion Probabilistic Models

Tao LIANG¹, Yue LI¹, Bin YANG¹, Jianglin LIU¹, Yaqiang LI², Meijie LIANG¹

Affiliations

^1.Key Laboratory of Urban and Engineering Safety and Disaster Mitigation, Ministry of Education, Beijing University of Technology, Beijing 100124, China

^2.College of Soil and Water Conservation, Beijing Forestry University, Beijing 100083, China

Corresponding Author:

LI Yue (1972-), male, Ph.D., Professor. E-mail: liyue@bjut.edu.cn

LIANG Tao (1997-), male, Doctoral candidate. E-mail: lt971102@163.com

出版时间: 2026-09-02 doi: 10.14062/j.issn.0454-5648.20260312

文章导航

摘要

收起

机器学习模型被广泛应用于混凝土配合比设计和性能预测，数据集的质量与数量对模型的拟合精度有着直接影响。然而，由于试验周期长、测试成本高、人为影响因素大等原因，构建大规模、高质量的混凝土配合比数据集仍面临显著挑战。本工作基于去噪扩散概率模型，构建了一种生成混凝土配合比虚拟数据的深度学习模型--MixDiff。定量分析结果显示，MixDiff生成的虚拟数据集的特征分布、特征相关性与真实数据集具有较高相似性，两者的瓦瑟斯坦距离仅为0.019 5，K-S统计量仅为0.095 2，J-S散度仅为0.120 7，皮尔逊相关矩阵差异的弗罗贝尼乌斯距离仅为0.332 7。使用多种机器学习模型分别基于真实小样本数据集和MixDiff生成的虚拟数据集进行训练，试验表明：通过引入虚拟数据集，RF、SVR、KNN、XGBoost、LightGBM、CatBoost和DNN共7种模型的平均R²相对上升16.72%，平均RMSE相对下降47.96%，平均MAE相对下降46.86%。MixDiff生成的配合比数据在7 d强度、28 d强度、流动度、坍落度、初凝时间与终凝时间上的平均MAPE仅为3.8%。本工作为混凝土配合比的数据扩充提供了一种新方法，可提升机器学习模型在小样本条件下的预测精度。

关键词

混凝土 / 配合比 / 去噪扩散概率模型 / 数据增强 / 深度学习

Abstract

收起

Introduction

Machine learning has been widely applied to various concrete materials tasks, such as macroscopic performance prediction and mix proportion optimization. The performance of machine learning models depends strongly on the quantity and quality of the available datasets. However, constructing large-scale and high-quality concrete mix proportion datasets remains challenging because concrete experiments are time-consuming, costly, and easily affected by material sources, curing conditions, and human operations. At present, small-sample modeling problems are mainly addressed through two strategies: model optimization and data augmentation. Model optimization methods mainly improve the extraction of information from existing small-sample datasets by adjusting model structures or training strategies. However, they do not fundamentally increase the amount of data, and their improvement in prediction accuracy is still limited by the original sample size. Conventional data augmentation methods mainly rely on generative adversarial networks, which learn the real data distribution through adversarial training between a generator and a discriminator, thereby generating virtual samples to expand the original dataset. Nevertheless, when dealing with complex tabular data, generative adversarial networks may encounter unstable training, insufficient sample diversity, and distribution shifts in generated data, resulting in limited consistency between virtual and real data. To address these issues, this study proposes MixDiff, a denoising diffusion probabilistic model (DDPM) for generating virtual concrete mix proportion data under small-sample conditions. This model aims to generate high-quality virtual concrete mix proportion data.

Methods

A small-sample dataset containing 100 sets of concrete mix proportions was first established. The input features included water, cement, fly ash, mineral fines, river sand, manufactured sand, coarse aggregate, water-reducing agent, total binder content, water-to-binder ratio, total fine aggregate content, and sand ratio. The target variables included 7 d compressive strength, 28 d compressive strength, slump, expansion degree, initial setting time, and final setting time. MixDiff was developed using a fully connected neural network and a denoising diffusion probabilistic model. The model consisted of a sinusoidal timestep embedding module and four stacked residual fully connected blocks. Each residual block contained fully connected layers, batch normalization layers, and SiLU activation functions. In the forward diffusion process, Gaussian noise was progressively added to the normalized data according to a cosine variance schedule. In the reverse generation process, a neural network was used to predict the noise and progressively reconstruct concrete mix proportion samples with realistic distribution characteristics from standard Gaussian noise. Subsequently, the quality of virtual data generated by MixDiff was quantitatively evaluated from two perspectives: marginal distribution and feature correlation. Under five-fold cross-validation and grid search, several mainstream machine learning models were further used to compare the prediction performance of the baseline mode and the augmentation mode. Finally, 15 mix proportions were selected from the generated virtual dataset for experimental validation to evaluate the engineering plausibility of the generated data.

Results and discussion

The quantitative evaluation results showed that the virtual dataset generated by MixDiff showed high similarity to the real dataset. The average Wasserstein distance, K-S statistic, and J-S divergence between the two datasets were 0.019 5, 0.095 2, and 0.120 7, respectively. The Frobenius distance between their Pearson correlation matrices was 0.332 7, indicating that MixDiff effectively preserved the marginal distribution characteristics of the real data and the correlation structure among variables. The machine learning comparison results showed that the prediction performance of all models improved significantly after the introduction of MixDiff-generated data. In the baseline mode, the average R² of the seven models was 0.817 1, and increased to 0.953 7 in the data augmentation mode. Meanwhile, the average RMSE decreased from 14.20 to 7.39, and the average MAE decreased from 8.77 to 4.66. Correspondingly, R² increased by 16.72%, while RMSE and MAE decreased by 47.96% and 46.86%, respectively. Among the machine learning models, the deep neural network achieved the best performance after data augmentation, with R², RMSE, and MAE values of 0.959 4, 7.00, and 4.52, respectively. In addition, the standard deviations of the five-fold cross-validation results were significantly reduced in the data augmentation mode. This indicates that MixDiff-generated data not only improved the prediction accuracy of machine learning models but also enhanced model stability and generalization ability. Experimental validation of 15 virtual mix proportions showed that the data generated by MixDiff had high plausibility in six key performance indicators, including 7 d compressive strength, 28 d compressive strength, slump, expansion degree, initial setting time, and final setting time. The average MAPE of each performance indicator was lower than 5%, and the overall average MAPE was 3.8%.

Conclusions

This study proposed MixDiff, a denoising diffusion probabilistic model for concrete mix proportion data generation. The results show that the virtual data generated by MixDiff were highly consistent with the real data in terms of marginal distribution and feature correlation. The Frobenius distance between the Pearson correlation matrices was only 0.332 7. After introducing MixDiff-generated data, the prediction performance of multiple machine learning models, including RF, SVR, KNN, XGBoost, LightGBM, CatBoost, and DNN, was significantly improved, with R² values of all models increasing above 0.94. Based on the average performance of the seven machine learning models for the six output performance indicators, R² increased by 16.72%, while RMSE and MAE decreased by 47.96% and 46.86%, respectively. Among these models, DNN achieved the highest R² and the lowest RMSE in the data augmentation mode. Experimental validation of 15 virtual mix proportions showed that the generated results were highly consistent with the measured results in terms of 7 d compressive strength, 28 d compressive strength, slump, expansion degree, initial setting time, and final setting time. The MAPE of each performance indicator was below 5%, and the overall average MAPE across the six performance indicators was 3.8%. These results indicate that MixDiff can learn the underlying relationships between concrete mix proportions and macroscopic properties, and it can serve as an effective method for concrete mix proportion data augmentation under small-sample conditions.

Key words

concrete / mixture proportion / denoising diffusion probabilistic model / data augmentation / deep learning

引用本文

梁涛, 李悦, 杨斌, 刘江林, 李亚强, 梁美洁. 基于去噪扩散概率模型的混凝土配合比数据增强方法. 硅酸盐学报, 2026 , 54 (10) : 3499 -3517 . DOI: 10.14062/j.issn.0454-5648.20260312

Tao LIANG, Yue LI, Bin YANG, Jianglin LIU, Yaqiang LI, Meijie LIANG. A Concrete Mix Proportion Data Augmentation Method Based on Denoising Diffusion Probabilistic Models[J]. Journal of the Chinese Ceramic Society, 2026 , 54 (10) : 3499 -3517 . DOI: 10.14062/j.issn.0454-5648.20260312

基金

收起

国家自然科学基金重点项目(U2368207)
北京市自然科学基金(24JL003)
国家自然科学基金(52508244)
中国博士后科学基金(GZC20252135)

参考文献

收起

文献

收起

参考文献引证文献

排序方式：

[1]

张剑. 面向多目标优化的超高性能混凝土抗压强度预测与配合比设计研究[J]. 土木工程与绿色建筑, 2026, 2(2): 84-89.

ZHANG Jian. Civ Eng Green Build, 2026, 2(2): 84-89.

[2]

元强, 夏瑞, 刘易, 等. 基于多目标寻优的低碳混凝土智能设计[J]. 硅酸盐学报, 2026, 54(3): 878-893.

YUAN Qiang, XIA Rui, LIU Yi, et al. J Chin Ceram Soc, 2026, 54(3): 878-893.

[3]

潘磊, 傅宇方, 王元丰. 粉煤灰混凝土冻融性能智能预测方法[J]. 硅酸盐学报, 2026, 54(2): 793-810.

PAN Lei, FU Yufang, WANG Yuanfeng. J Chin Ceram Soc, 2026, 54(2): 793-810.

[4]

孟繁星, 许增光, 付涵, 等. 高海拔强紫外环境下花岗岩沥青混凝土老化性能研究[J]. 水力发电学报, 2026, 45(5): 95-108.

MENG Fanxing, XU Zengguang, FU Han, et al. J Hydroelectr Eng, 2026, 45(5): 95-108.

[5]

马宗方, 戴溢阳, 贺静, 等. 基于机器学习的混凝土3D打印沉积线状态优化[J]. 工程设计学报, 2025, 32(4): 488-498.

MA Zongfang, DAI Yiyang, HE Jing, et al. Chin J Eng Des, 2025, 32(4): 488-498.

[6]

ZHANG Y, YUAN X H, ZHANG X P, et al. Prediction of high-performance concrete compressive strength using Decision Tree-Guided Artificial Neural Network Pretraining approach[J]. Eng Appl Artif Intell, 2025, 156: 110828.

[7]

WOO B H, CHOI D H, PARK C H, et al. Gaussian process-based data augmentable deep Bayesian neural network: Application to estimate the stress transfer length in pretensioned concrete beams[J]. Eng Struct, 2026, 348: 121809.

[8]

TIAN J C, WANG X T, LI Y L, et al. Small-sample modelling of early-stage operation of high concrete dams in cold regions based on deep transfer learning[J]. Appl Soft Comput, 2026, 188: 114492.

[9]

QU H L, HU T Y, QU J T, et al. A machine learning model based on GAN-ANN data augmentation for predicting the bond strength of FRP-reinforced concrete under high-temperature conditions[J]. Compos Struct, 2025, 369: 119321.

[10]

LI H R, YANG H F. Predictive model of bond strength between recycled aggregate concrete and rebar after high-temperature based on conditional table generation adversarial networks data augmentation and gene expression programming algorithm[J]. Eng Appl Artif Intell, 2025, 162: 112786.

[11]

ARJOSKY M, CHINTALA S, BOTTOU L. Wasserstein generative adversarial networks[C]//Proceedings of the 34th International Conference on Machine Learning (ICML 2017), Sydney, NSW, Australia. PMLR, 2017, 70: 214-223.

[12]

XU L, SKOULARIDOU M, CUESTA-INFANTE A, et al. Modeling tabular data using conditional GAN[C]//Advances in Neural Information Processing Systems 32 (NeurIPS 2019). Red Hook, NY: Curran Associates, 2019: 6594-6604.

[13]

HO J, JAIN A, ABBEEL P. Denoising diffusion probabilistic models[C]//Advances in Neural Information Processing Systems 33 (NeurIPS 2020). Red Hook, NY: Curran Associates, 2020: 6840-6851.

[14]

SONG J, MENG C, ERMON S. Denoising diffusion implicit models[C]//International Conference on Learning Representations (ICLR 2021), Vienna, Austria (virtual), 2021.

[15]

SONG Y, SOHL-DICKSTEIN J, KINGMA D P, et al. Score-based generative modeling through stochastic differential equations[C]//Advances in Neural Information Processing Systems 33 (NeurIPS 2020). Red Hook, NY: Curran Associates, 2020: 18483-18494.

[16]

ROMBACH R, BLATTMANN A, LORENZ D, et al. High-resolution image synthesis with latent diffusion models[C]//2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). New Orleans, LA, USA. IEEE, 2022: 10674-10685.

[17]

HORNIK K, STINCHCOMBE M, WHITE H. Multilayer feedforward networks are universal approximators[J]. Neural Netw, 1989, 2(5): 359-366.

[18]

HE K M, ZHANG X Y, REN S Q, et al. Deep residual learning for image recognition[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Las Vegas, NV, USA. IEEE, 2016: 770-778.

[19]

PEARSON K. VII. Note on regression and inheritance in the case of two parents[J]. Proc R Soc Lond, 1895, 58(347-352): 240-242.

[20]

SOHL-DICKSTEIN J, WEISS E A, MAHESWARANATHAN N, et al. Deep unsupervised learning using nonequilibrium thermodynamics[C]//Proceedings of the 32nd International Conference on Machine Learning-Volume 37. ACM, 2015: 2256-2265.

[21]

NICHOL A Q, DHARIWAL P. Improved denoising diffusion probabilistic models[C]//Proceedings of the 38th International Conference on Machine Learning (ICML 2021). PMLR, 2021, 139: 8162-8171.

[22]

VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[C]//Advances in Neural Information Processing Systems 30 (NIPS 2017), Long Beach, California, USA. Red Hook, NY: Curran Associates Inc., 2017: 5998-6008.

[23]

ELFWING S, UCHIBE E, DOYA K. Sigmoid-weighted linear units for neural network function approximation in reinforcement learning[J]. Neural Netw, 2018, 107: 3-11.

[24]

IOFFE S, SZEGEDY C. Batch normalization: accelerating deep network training by reducing internal covariate shift[C]//Proceedings of the 32nd International Conference on Machine Learning (ICML 2015). PMLR, 2015, 37: 448-456.

[25]

刘悦, 邹欣欣, 杨正伟, 等. 材料领域知识嵌入的机器学习[J]. 硅酸盐学报, 2022, 50(3): 863-876.

LIU Yue, ZOU Xinxin, YANG Zhengwei, et al. J Chin Ceram Soc, 2022, 50(3): 863-876.

[26]

杨斌, 李悦, 林辉, 等. 荷载-离子-温度共同作用下地下混凝土损伤程度分析与预测[J]. 硅酸盐学报, 2026, 54(2): 653-665.

YANG Bin, LI Yue, LIN Hui, et al. J Chin Ceram Soc, 2026, 54(2): 653-665.

[27]

张文生, 曹傅荔, 郅晓, 等. 机器学习方法用于水泥基材料的研究进展[J]. 硅酸盐学报, 2024, 52(11): 3617-3630.

ZHANG Wensheng, CAO Fuli, ZHI Xiao, et al. J Chin Ceram Soc, 2024, 52(11): 3617-3630.

[28]

COVER T, HART P. Nearest neighbor pattern classification[J]. IEEE Trans Inform Theory, 1967, 13(1): 21-27.

[29]

李九阳, 陈立, 吴纪曙, 等. 基于可解释性机器学习算法的FRP筋UHPC粘结强度预测[J]. 辽宁工程技术大学学报(自然科学版), 2026, 45(2): 185-195.

LI Jiuyang, CHEN Li, WU Jishu, et al. J Liaoning Tech Univ Nat Sci Ed, 2026, 45(2): 185-195.

[30]

BREIMAN L. Random forests[J]. Mach Learn, 2001, 45(1): 5-32.

[31]

李华, 古金本, 陶毅. 基于机器学习预测纤维增强聚合物(FRP)筋-混凝土黏结强度的研究[J]. 工业建筑, 2025, 55(10): 102-114.

LI Hua, GU Jinben, TAO Yi. Ind Constr, 2025, 55(10): 102-114.

[32]

郭云杰, 刘俊乐, 孙虎, 等. 基于LightGBM算法的高性能混凝土抗压强度预测研究[J]. 建筑机械, 2025(3): 168-173.

GUO Yunjie, LIU Junle, SUN Hu, et al. Constr Mach, 2025(3): 168-173.

[33]

PROKHORENKOVA L, GUSEV G, VOROBEV A, et al. CatBoost: unbiased boosting with categorical features[C]//Advances in Neural Information Processing Systems 31 (NeurIPS 2018). Red Hook, NY: Curran Associates, 2018: 6638-6648.

[34]

LECUN Y, BENGIO Y, HINTON G. Deep learning[J]. Nature, 2015, 521(7553): 436-444.

[35]

YANG B, LI Y, LIN H, et al. Multi-objective design method of concrete mix ratio based on machine learning: Forward prediction and reverse design[J]. Arch Civ Mech Eng, 2026, 26(2): 57.

2026年第54卷第10期

PDF下载

引用本文

BibTeX

文章信息

doi: 10.14062/j.issn.0454-5648.20260312

接收时间：2026-05-09
首发时间：2026-10-22
出版时间：2026-09-02

补充材料

相关文章

文章信息

作者

出版历史

收稿日期：2026-05-09
修回日期：2026-06-05

基金

国家自然科学基金重点项目(U2368207)

北京市自然科学基金(24JL003)

国家自然科学基金(52508244)

中国博士后科学基金(GZC20252135)

作者信息

^1.北京工业大学城市与工程安全减灾教育部重点实验室，北京　100124

^2.北京林业大学水土保持学院，北京　100083

通讯作者:

李悦（1972-），男，博士，教授。

参考文献

分享链接

https://castjournals.cast.org.cn/joweb/gsyxb/CN/10.14062/j.issn.0454-5648.20260312

分享至

全文二维码

扫描看全文

引用本文

BibTeX

本文的引用情况

2种不同金属材料的力学参数

科 Family	属数 Number of genus	种数 Number of species	占总种数比例 Percentage of total species (%)	属 Genus	种数 Number of species	占总种数比例 Percentage of total species (%)
鹅膏菌科Amanitaceae	2	11	5.26	鹅膏菌属 Amanita	10	4.78
小菇科 Mycenaceae	2	12	5.74	丝盖伞属 Inocybe	5	2.39
多孔菌科 Polyporaceae	8	14	6.70	蜡蘑属 Laccaria	5	2.39
红菇科 Russulaceae	3	23	11.00	小皮伞属 Marasmius	6	2.87
				小菇属 Mycena	11	5.26
				光柄菇属 Pluteus	5	2.39
				红菇属 Russula	17	8.13
				栓菌属 Trametes	5	2.39

关闭全屏

BibTeX
EndNote
RefWorks
TxT