Devin K. Harris

Bio: Devin K. Harris is an academic researcher from University of Virginia. The author has contributed to research in topics: Structural load & Girder. The author has an hindex of 19, co-authored 92 publications receiving 1039 citations. Previous affiliations of Devin K. Harris include Michigan Technological University & Amirkabir University of Technology.

Bond Performance between Ultrahigh-Performance Concrete and Normal-Strength Concrete

Abstract: Ultrahigh-performance concrete (UHPC) exhibits several properties that make it appropriate for the rehabilitation of concrete structures. In this investigation, the application is focused on bridge deck overlays, but the study is equally applicable to other rehabilitation applications. Its negligible permeability makes this material suitable as a protective barrier that prevents any water or chemical penetration into the substrate. In addition, its ultra-high compressive strength and post-cracking tensile capacity could provide an improvement to the bearing capacity. However, for extensive acceptance, it has to be demonstrated that the bond between UHPC and normal strength concrete (NSC) offers a good long-term performance under a variety of operating conditions. The UHPC-NSC interface can experience high tensile, shear, and compressive stresses at both early and later life stages and the environmental conditions inherent to the operating environment. The success of the rehabilitation will depend ...

Evaluation of Commercially Available Remote Sensors for Highway Bridge Condition Assessment

Abstract: Improving transportation infrastructure inspection methods and the ability to assess conditions of bridges has become a priority in recent years as the transportation infrastructure continues to age. Current bridge inspection techniques consist largely of labor-intensive subjective measures for quantifying deterioration of various bridge elements. Some advanced nondestructive testing techniques, such as ground- penetrating radar, are being implemented; however, little attention has been given to remote sensing technologies. Remote sensing technologies can be used to assess and monitor the condition of bridge infrastructure and improve the efficiency of inspection, repair, and rehabilitation efforts. Most important, monitoring the condition of a bridge using remote sensors can eliminate the need for traffic disruption or total lane closure because remote sensors do not come in direct contact with the structure. The purpose of this paper is to evaluate 12 potential remote sensing technologies for assessing the bridge deck and superstructure condition. Each technology was rated for accuracy, commercial availability, cost of measurement, precollection preparation, complexity of analysis and interpretation, ease of data collection, stand-off distance, and traffic disruption. Results from this study demonstrate the capabilities of each technology and their ability to address bridge challenges.

Robust Pixel-Level Crack Detection Using Deep Fully Convolutional Neural Networks

Abstract: This paper introduces the idea of using deep fully convolutional neural networks for pixel-level defect detection in concrete infrastructure systems. Although coarse patch-level deep learni...

Learning without Forgetting

Abstract: When building a unified vision system or gradually adding new capabilities to a system, the usual assumption is that training data for all tasks is always available. However, as the number of tasks grows, storing and retraining on such data becomes infeasible. A new problem arises where we add new capabilities to a Convolutional Neural Network (CNN), but the training data for its existing capabilities are unavailable. We propose our Learning without Forgetting method, which uses only new task data to train the network while preserving the original capabilities. Our method performs favorably compared to commonly used feature extraction and fine-tuning adaption techniques and performs similarly to multitask learning that uses original task data we assume unavailable. A more surprising observation is that Learning without Forgetting may be able to replace fine-tuning with similar old and new task datasets for improved new task performance.

Adhesion and Adhesives

Abstract: Adhesion and Adhesives Edited by Dr. R. Houwink and Dr. G. Salomon. Vol. 1: Adhesives. Second, completely revised edition. Pp. xvi + 548. (Amsterdam, London and New York: Elsevier Publishing Company, 1965.) 135s.