Scholarlink

<p>In this study, the shrinkage evolution characteristics of Red Clay under Wetting–Drying Cycles were systematically analyzed through laboratory tests. The results indicate that the shrinkage deformation of Red Clay exhibits an S-shaped nonlinear growth trend characterized by "fast first and then slow". Specifically, it can be divided into three stages: during the initial response stage, samples with high water content and low dry density exhibit larger shrinkage increments; a significant transition occurs during the rapid change stage; and in the stabilization stage, it completely transforms into the opposite correspondence. To address the prediction challenges posed by the small-sample size and high volatility, an Empirical Bayes residual-corrected grey prediction model, EBGM(1,1), was constructed. The results indicate that, across the entire sample, the Mean Absolute Percentage Error (MAPE) for both the linear and volumetric shrinkage rates of the modified model decreased by approximately 50% compared to the traditional GM(1,1) model. Furthermore, all accuracy evaluation metrics for the six groups of typical samples exhibited a substantial reduction, suggesting a notable enhancement in the stability and accuracy of the predictions. Further verification suggests that the model not only exhibits strong generalization capability and engineering applicability, but its adaptive adjustment logic also appears to align well with the physical damage process of red clay. Consequently, this study can provide a reliable theoretical foundation and technical reference for the early warning of geological hazards and the prevention and control of deformation in red clay.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Application of a Modified Grey Model in Predicting the Shrinkage Evolution of Red Clay Under Wetting–Drying Cycles

  • 期刊论文

  • 发表日期：

    2026年5月3日

查看全文
  • Binghui Zhang,
  • Jicheng Xu,
  • Yanhua Xie,
  • Yiming Zhong,
  • Chenxu Wang,
  • Weizhen Zhu

摘要

In this study, the shrinkage evolution characteristics of Red Clay under Wetting–Drying Cycles were systematically analyzed through laboratory tests. The results indicate that the shrinkage deformation of Red Clay exhibits an S-shaped nonlinear growth trend characterized by "fast first and then slow". Specifically, it can be divided into three stages: during the initial response stage, samples with high water content and low dry density exhibit larger shrinkage increments; a significant transition occurs during the rapid change stage; and in the stabilization stage, it completely transforms into the opposite correspondence. To address the prediction challenges posed by the small-sample size and high volatility, an Empirical Bayes residual-corrected grey prediction model, EBGM(1,1), was constructed. The results indicate that, across the entire sample, the Mean Absolute Percentage Error (MAPE) for both the linear and volumetric shrinkage rates of the modified model decreased by approximately 50% compared to the traditional GM(1,1) model. Furthermore, all accuracy evaluation metrics for the six groups of typical samples exhibited a substantial reduction, suggesting a notable enhancement in the stability and accuracy of the predictions. Further verification suggests that the model not only exhibits strong generalization capability and engineering applicability, but its adaptive adjustment logic also appears to align well with the physical damage process of red clay. Consequently, this study can provide a reliable theoretical foundation and technical reference for the early warning of geological hazards and the prevention and control of deformation in red clay.