Image Data-Driven-Deep Learning in Geosystems: Exploratory Investigation into the Stability of Retaining Walls

SPONSOR: NATIONAL SCIENCE FOUNDATION

PI:  Zhen Liu

 

Project Period:  09.01.17 – 08.31.19

The project is proposed to introduce the most recent breakthroughs in computer vision attributed to deep learning to address a rarely discussed yet urgent issue in most engineering disciplines: how to analyze the explosively increasing image data including images and videos, which can hardly be analyzed with traditional methods? Therefore, the proposed work is very time-sensitive. The core concepts enabling the breakthroughs in image recognition and adopted in AlphaGo, i.e., deep convolutional neural nets, will be used to explore the possibility to accurately assessing the safety of retaining walls with image data. The analysis will be validated against traditional methods including limit analysis and numerical simulation. The research is hypothesis-driven and rationally built on our preliminary study, which clearly supports the hypothesis and shows the high potential of the proposed study. In addition to further proving the power of the proposed concept, this collaborative effort between geotechnical engineering and computer science will understand the data and deep learning in geotech analysis, relate image patterns to physical mechanisms, and investigate two key potential issues associated with the methodology.

The proposed work will be organized into three tasks. Task 1 is to understand how the geotechnical data and analysis results can be related to the input and output of deep learning using convolutional nets. This is a key to the successful application of deep learning in geotech. In Task 2, eight factors in three categories, i.e., geology (soil property), topography (geometry) and boundary conditions, will be evaluated regarding their influence on the safety factors. The relationships between these factors, their roles and significance in deep learning, and correlation between the patterns identified by the CNN and the physical mechanisms will be compared and revealed to connect the deep learning to the existing geotech knowledge pool. In Task 3, two additional key aspects representing the unknowns and possible weaknesses of the machine learning method will be analyzed: model robustness and extrapolation. Testing data prepared in ways different from the input data, i.e., appearance and ranges, will be used to assess and improve these two aspects.

Zhen Liu
Zhen Liu
Shiyan Hu
Shiyan Hu

 

Collaborative Research: Understanding Mechanism of Internal Frost-Induced Damage of Concrete from Microstructure Aspects

Sponsoring Agency: National Science Foundation

PI: Qingli Dai

This project integrates research and education to advance the state of knowledge of the mechanism of frost-induced damage in Portland cement concrete under freeze-thaw cycles. The primary objective of this research project is to combine expertise in microstructure-based computational modeling and innovative sensor technologies to study the fundamental mechanisms of frost damage in concrete. Research will include the experimental characterization of concrete microstructure across different length scales, the development of an innovative Time Domain Reflectometry (TDR) sensor to accurately determine the freeze-thaw status, and the formulation and validation of a frost-induced damage model. This research is expected to result in a model that can clearly and concisely describe the damage that frost can inflict in concrete. This model will provide a valuable tool to assess the potential success of various frost damage prevention strategies and products.

This research will help develop durable concrete and benefit the industries involved with concrete design and construction in cold regions. The durability of concrete plays a central role in the sustainability of the whole infrastructure system on which such regions depend for their development. In this project, research and educational activities will be integrated to promote teaching, training, and learning for the K-12 students and teachers, undergraduate and graduate students in engineering and science, and professional engineers. Additionally, the methodology developed in this project for understanding the frost damage mechanisms of concrete will be applicable for solving other durability issues such as salt scaling and chemical reaction.

Publications: https://nsf.gov/awardsearch/showAward?AWD_ID=0900015

Qingli Dai
Qingli Dai

DRU Collaborative Research: Understanding Mental Models of Expertise in Construction Management using Interactive Adaptive Simulations

Sponsor: National Science Foundation

PI: Amlan Mukherjee

This research uses recent advances in simulations and data analysis techniques to investigate the cognitive and engineering aspects of decision making in complex dynamic construction management scenarios. Expertise plays a crucial role in managing crisis scenarios that call for critical decision making under constraints of time, resource, and rapidly unfolding events. An example of such a crisis scenario is managing complex heavy construction projects. In such scenarios, effective decision making requires knowledge of complex inter-relationships between several simultaneous events and preparing for the uncertainty and risks arising from feedbacks in time and space. Such knowledge is inductively constructed by assimilating and organizing experiential knowledge into patterns of information that are difficult to formalize or analytically perceive. The researchers propose to investigate the dynamics and variation of such cognitive knowledge organization patterns, or mental models, of decision making, specifically among construction managers.

The goal of this research effort will be to use an interdisciplinary approach to understand how expert and novice construction managers differ in their knowledge organization, information processing, risk assessment, and decision making in construction management crisis scenarios. Interactive, adaptive simulations of critical construction scenarios will be developed in collaboration with construction management firms, and expert and novice construction managers will be tested in them to capture human-subject interaction data that will be analyzed to develop mental models of expertise. In addition, an instructional interface will be integrated into the simulation using pedagogical agents, and it will be deployed in the construction management curriculum to test its effectiveness as a training environment for novice decision makers. This will also allow the researchers to investigate how novices construct knowledge in simulated training environments.

Amlan Mukherjee
Amlan Mukherjee

A Microstructure-Based Modeling Approach to Characterize Asphalt Materials

SPONSOR:  NATIONAL SCIENCE FOUNDATION

PI: Zhanping You

The short-term goals of this integrated research and education activities within the project period include: 1) development of a microstructure-based discrete element modeling approach to characterize asphalt materials; 2) implementation of the model to evaluate asphalt material response and performance to improve pavement structural design; 3) integration of the proposed research activities into the educational programs for high school students, K-12 educators, and undergraduate and graduate students, and; 4) dissemination of the research results through publications, conferences, and professional development for practicing professionals. The long-term goals are to: 1) establish a multi-user research and education center for asphalt material and virtual testing by integrating the proposed advanced modeling approach, and; 2) implement advanced technologies into pavement materials, locally through the Michigan Department of Transportation, and broadly through research collaborators and industry partners. This project will: 1) advance the understanding of asphalt pavement materials and pavement structures; 2) increase collaboration among researchers at many leading institutions and industries; 3) enhance scientific and technological understanding of micromechanical aspects of pavement infrastructure, and; 4) significantly reduce the cost of pavement infrastructure construction and maintenance. The development and application of the microstructure-based discrete element model in asphalt materials will enhance understanding by correlating material behavior to pavement performance.

The research will translate directly into improved asphalt mixture design and pavement thickness design. This in turn should create substantial cost savings. A one-percent decrease in asphalt concrete life-cycle cost would amount to approximately $500 million in U.S. Federal government savings alone. Activities are planned to advance discovery and understanding of asphalt pavement infrastructure materials while promoting teaching, training, and learning through specific activities for K-12 students and teachers, undergraduate and graduate engineering and science majors, and practicing engineers.

Zhanping You
Zhanping You

US-Malaysia Planning Visit: Collaborative Research on the Micromechanics of Cubic Stone Materials for Pavements

Sponsor: National Science Foundation

PI: Zhanping You

This award supports the participation of an American researcher, graduate student and undergraduate students in the planning visit which will take place in Malaysia. The visit will enable Professor Zhanping You in the Civil Engineering Department at Michigan Technology University to meet with Professor Meor Othman Hamzah in the School of Civil Engineering at the University Sains Malaysia (USM) in Penang. Their proposed project will involve: 1) increasing multidisciplinary collaboration among researchers in Michigan and Malaysia at their leading institutions; and 2) discovering the mechanism of rutting and fatigue distresses by using advanced micromechanics based discrete element models through the collaborative effort. The team will visit USM?s Asphalt Laboratory to work with the Malaysian professors and students to study and evaluate the feasibility of using cubical aggregated in pavement to reduce rutting potential. The students will also have an opportunity to participate in the testing of samples with/without cubic-stone materials on dynamic modulus and resilient modulus testing. The U.S. students will receive the testing results in order to use the data for discrete element modeling. The discrete element model will be further refined to study the various material phases (aggregates and mastic/asphalt) of pavement materials in order to determine the rutting and fatigue performance of asphalt pavements.

There is sufficient overlap of interests between researchers at the two universities to indicate that the researchers can successfully pursue the activities proposed, and the interaction will benefit both sides. This collaboration will advance discovery and understanding of cubic-stone materials micromechanics, while promoting teaching, training, and learning through the specific activities planned for the students. It is anticipated that the inclusion of the students in this visit will provide them unique training and educational opportunities by providing them a global research experience. These early collaborations between the scientists and students from each country will likely lead to long-term collaborations that will benefit both institutions.

Zhanping You
Zhanping You

EAGER: Using Nonmetals Separated from E-Waste and Waste Plastic Bags in Improving the Mechanical Properties of Asphalt Materials

Sponsor: National Science Foundation

PI: Zhanping You

The objective of this EAGER project is to investigate the possibility of improving the mechanical properties of asphalt materials with the use of nonmetals separated from E-waste (e.g., computers, monitors, keyboard, cameras, TVs, etc.) and waste plastic bags (e.g., grocery bags). E-waste and waste plastic bags are recycled materials that have potential to be used in asphalt materials. The research work will include shredding of the non-metallic separations and waste plastic bags into powders and particles, mixing the powder-like polymers and particles to modify selected asphalt materials in the laboratory, and evaluation of the mechanical properties of the modified asphalt.

Through its integrated research and educational plan, this project will advance discovery and understanding of infrastructural materials, while promoting teaching, training, and learning – impacting underrepresented high school students and teachers, undergraduate and graduate students, and professionals from industry and the government. This project will directly benefit society through improved transportation systems and lower infrastructure costs. The project also leads to collaborative efforts with a historically black Carnegie doctoral/research intensive public institution, Jackson State University.

Zhanping You
Zhanping You

I-Corps: Decision Support Systems for Managers of Civil Infrastructure Systems

Sponsor: National Science Foundation

PI: Amlan Mukherjee

The proposed technology is a methodology to assess alternative infrastructure management strategies based on project cost, system performance and estimates of greenhouse emissions. The supporting methods include stochastic analysis, life cycle assessment and Monte Carlo simulation based approaches. The technology is designed to address the problems of reducing life cycle emissions of civil infrastructure systems, helping agencies provide a consistent level of service, while optimally using available resources for construction, maintenance and rehabilitation. This is particularly significant given the twin challenges of climate change, and ongoing shortfalls in state and federal budget appropriations for public works. Finally, the underlying theory and methods are mathematically sophisticated and data intensive, and not easy for decision-makers to implement without significant training. The proposed technology promises efficient implementation by providing an innovative product/service that is reliable and intuitive, and has a friendly and easy to use interface.

Civil infrastructure systems are critical to socioeconomic success. Services such as access to clean drinking water, efficient sewer and waste management, easy mobility and access to multiple modes of transportation provide the backbone for multiple supply chains, besides supporting a healthy standard of living. Challenges due to climate change, aging infrastructure, and the impact of the economic crisis on local and state budgets are hurting the efficient delivery of these services. By providing support to decision-makers the proposed technology is likely to have a significant impact on maintaining and managing infrastructure sustainably.

Amlan Mukherjee
Amlan Mukherjee

Improving Spatial Observability of Dynamic Traffic Systems through Active Mobile Sensor Networks and Crowdsourced Data

SPONSOR:  NATIONAL SCIENCE FOUNDATION (NSF)

PI:  Kuilin Zhang

To provide effective traffic congestion mitigation strategies, transportation agencies need to effectively design sensor networks to reliably estimate and predict traffic conditions across large transportation networks. The next generation traffic sensor network will offer large, diverse data streams not only from fixed traffic detectors, but also from many emerging active mobile traffic sensors such as Unmanned Aerial Vehicles, self-driving cars, and crowdsourced data sources from social sensors and transportation network companies. This new generation of agile sensors can provide a much richer but also increasingly complex traffic data environment. Moreover, crowdsourced data is generally uncontrolled, inaccurate and unreliable. This research focuses on new sensor design/control applications to transform the interconnection between travelers, sensors, data and transportation management systems.

The objective of this research is to develop rigorous mathematical foundations and innovative algorithms to accurately quantify spatial observability of dynamic traffic states, optimize active mobile sensor locations, and mine information from crowdsourced data sources. The research team will first characterize analytical space-time distributions of different traffic states at both macroscopic and microscopic scales, and further develop time-geography-oriented optimization for quantifying spatial observability for dynamic networks. A new class of ubiquitous sensor network design problems is studied for the traffic state estimation stage, and the integration of the well-fused crowdsourced data with optimized fixed and active mobile sensor data is investigated under different levels of activity/penetration rates. Utilizing the structure of underlying dynamic transportation networks, this research aims to develop computationally efficient optimization algorithms to create a distributed and scalable computing framework, which can solve joint scheduling and routing problems of active mobile sensors to increase coverage and accuracy. The research team will develop generic measures of spatial network observability that can provide additional theoretical findings for general civil engineering systems such as earthquake impact detection, ground water pollution source identification, and critical infrastructure monitoring.

Kuilin Zhang
Kuilin Zhang
Colin N. Brooks
Colin N. Brooks

Exploratory Investigation of Thermally-Induced Water Flow in Soils

SPONSOR:  NATIONAL SCIENCE FOUNDATION (NSF)

PI:  Zhen Liu

This project aims to answer a very fundamental yet very old scientific question: “Why and how does water move due to temperature gradients in porous materials?” This thermally induced water flux ubiquitously exists in porous materials, whenever both heat transfer and water movement are present. A scientific understanding of this phenomenon is an essential base for many important scientific and social challenges: climate effects on geomaterials, geothermal energy applications, behavior of porous materials under extreme conditions, and recovery of non-conventional fossil fuels such as gas hydrates and shale gas. However, despite the significance, this phenomenon has been an historically unsolved and perplexing issue affecting many science and engineering areas involving porous materials from traditional applications in civil engineering, soil science and petroleum engineering to emerging needs in microfluidics, material processing and biomechanics.

This award supports the exploration of a new research concept/methodology and its application to reveal the physical mechanisms underlying thermally induced water flux for a complete scientific description and analysis framework for this phenomenon. As an exploratory study, which pioneers a very high-risk but possibly high-return concept, the success of the study may provide the geotechnical community a new understanding to tackle many issues which are hard to solve in the existing frameworks, and also provide a way to integrate porous material research which is currently distributed in various disciplines. In addition to supporting a doctoral student, the project will support outreach activities for rural, low-socioeconomic students and native tribal communities in the Upper Peninsula of Michigan. An annual summer program will be established to engage K-12 students in hands-on-learning for understanding of porous materials.

Zhen Liu
Zhen Liu