Ming Huang

Prof. Ming Huang

orcid

Research Focus & Metrics

Academic Career Length

18Years (2009-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
16 January 2018 in Natural Hazards
Verified
Journal Article
Closed access
PDF file
ZLZhuguang Lan
Zhuguang Lan
MHMing Huang
Ming Huang

1 December 2009 in Journal of Performance of Constructed Facilities
Verified
Journal Article
Closed access
MHMing Huang
Ming Huang
JLJun LIU
Jun LIU
Local government built the monitoring project to know seawall performance and ensure its safety in Shanghai Pudong New Area. Up to now, monitoring data of some important monitoring items including tidewater level, seepage pressure, rainfall, wave, deformation, etc., have been collected. The monitoring system is introduced here. Furthermore, the data analysis methods have been studied particularly because the seawall works in a complex special environment. The rules and relationship within seepage pressure, tidewater, and rainfall are presented, and the time distance between tidewater and seepage pressure curves are confirmed by statistical distance analysis. To do quantificational analysis of seawall performance, two methods based on statistical theory and general regression neural network are used to establish monitoring models considering the main effect factors. Instances show the models can achieve good fitting and forecasting results of seawall seepage pressure. Application experiences are also presented.

6 June 2024 in KSCE Journal of Civil Engineering
Verified
Journal Article
Open access
PDF file
MHMing Huang
Ming Huang
JJJunxi Jia
Junxi Jia
Kang LiuKang Liu
Kang Liu
53.5% of China’s territory is seasonally frozen. Along with China’s fast social and economic development, more and more infrastructure facilities, e.g., railways, highways and hydraulic engineering projects, are being constructed in frozen soil areas. Since foundations are the underpinnings of these engineered structures, the performance of frozen soil foundations is critical to the safe and stable operation of these infrastructures. Frozen soil exhibits strong heterogeneity, which influences the prediction of the engineering performance of frozen soil foundations; however, previous studies have ignored this effect. In this paper, the coupling effect of thermal and hydraulic behavior and soil parameter heterogeneity of frozen soils is considered. A rectangular foundation is taken as an illustrative example, and COMSOL Multiphysics is applied to predict the coupled thermal and hydraulic behavior of frozen soil. Karhunen-Loève expansion is selected to capture the frozen soil parameters’ heterogeneity. Based on the analysis, the proposed framework can successfully calculate the stochastic frozen depth of the foundation. Due to the thermal conductivity’s heterogeneity of soil grains, the depth to which the foundation is frozen changes in a range rather than a single value in the deterministic analysis.

7 March 2019 in Arabian Journal for Science and Engineering
Verified
Journal Article
Closed access
PDF file
ZLZhuguang Lan
Zhuguang Lan
MHMing Huang
Ming Huang

1 October 2026 in AIP Advances
Verified
Journal Article
Open access
MHMing Huang
Ming Huang
YCYunhao Chu
Yunhao Chu
Kang LiuKang Liu
Kang Liu
Fan YangFan Yang
Fan Yang
The three-dimensional (3D) ecological mat is a novel ecological slope protection technology. To investigate the anti-scour performance of this slope protection technology, scour experimental studies were conducted on the 3D ecological mat for slope protection specimens. A self-designed composite scour experimental flume and specimens were used. Scour depths were measured. A corresponding numerical model was established according to the experimental model. Parameter combination samples were constructed by using the orthogonal experimental design method, and the scour depth corresponding to each parameter sample was obtained via numerical simulation. Taking the calculated scour depth as input and the corresponding parameter combination as output, a training set of samples was built and a nonlinear mapping model was established using the radial basis function neural network. Based on the experimental study results, soil parameters were intelligently inverted and their accuracy verified. Furthermore, based on the inverted parameters, in-depth calculation and analysis were performed on the slope flow velocity field, scour morphology, and scour depth under different slope scour conditions. The results show that the 3D ecological mat for slope protection can effectively reduce the flow velocity; the scour depth accounts for 82% from the beginning to 1 h of scour, and the scour depth basically stabilizes after 2 h of scour.

1 December 2024 in AIP Advances
Verified
Journal Article
Open access
MHMing Huang
Ming Huang
HYHougang Yang
Hougang Yang
Fan YangFan Yang
Fan Yang
Displacement prediction is crucial to landslide engineering monitoring and early warning. An Optuna-NeuralProphet model is proposed based on the Optuna framework and the NeuralProphet model to address the challenge of predicting step-like landslide displacement. The NeuralProphet model, with its capabilities for time series decomposition and combination prediction, is introduced to predict step-like landslide displacement. The various modules of the NeuralProphet model, such as the trend, periodicity, and auto-regression modules, effectively capture the complex characteristics of landslide monitoring data. The Optuna framework is utilized to optimize the model’s hyperparameters, enhancing its applicability and prediction accuracy. The Baijiabao landslide displacement prediction model is constructed by selecting appropriate modules of the NeuralProphet model based on the monitoring data characteristics. Subsequently, the model’s hyperparameters are optimized to facilitate the training and prediction of landslide displacement data. Finally, the efficacy of the Optuna-NeuralProphet model is validated through comparative analysis with multiple models. The results indicate that the Optuna-NeuralProphet model achieves higher accuracy and accurately predicts landslide deformation, better fulfilling the practical requirements of step-like landslide monitoring and early warning.