Shang Gao

Mr. Shang Gao

orcid

Research Focus & Metrics

Academic Career Length

1Year (2026-Present)

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

Representative Publications

View All Publications
24 June 2026 in Applied Sciences
Verified
Journal Article
Open access
PDF file
YLYiteng LIN
Yiteng LIN
QSQingtian Su
Qingtian Su
Fawas Omogbolahan MatanmiFawas Omogbolahan Matanmi
Fawas Omogbolahan Matanmi
XYXingfei Yan
Xingfei Yan
SGShang Gao
Shang Gao
This study focuses on a novel three-span hybrid continuous beam bridge, analyzing the force performance and key design parameters of the non-cellular post-support plate joint. A finite element model and parametric analysis were used to reveal the stress distribution patterns, the load-bearing characteristics of the connectors, and the load transfer path under negative bending moments. The study shows that the axial force within the joint is equitably shared among three load paths: the top slab concrete (20.7%), the bearing plate (40.1%), and the shear connectors (39.2%). Although interfacial friction contributes approximately 27.1% to the total shear resistance, it is conservatively recommended to neglect this effect in design due to inherent uncertainties. Parametric analysis reveals distinct marginal effects and efficiency thresholds: increasing the bearing plate thickness from 20 mm to 100 mm results in a mere 1.0 MPa reduction in the peak concrete stress, while extending the joint length beyond 1.0 times the beam height renders the central connectors ineffective. Furthermore, reducing the connector stiffness effectively lowers the non-uniformity coefficient from 2.3 to below 2.0. Notably, the first row of web PBLs carries 34.8% to 47.2% of the total shear force, with a stable non-uniformity coefficient of 1.05–1.06, establishing it as the critical control section for simplified design. These findings provide a theoretical basis and practical guidance for the design of similar joints in hybrid girder bridges.

22 June 2026
Verified
Conference Paper
Open access
JAJatin Aggarwal
Jatin Aggarwal
SGShang Gao
Shang Gao
ABAmelie Bonde
Amelie Bonde
ZHZhili He
Zhili He
JGJulia Gersey
Julia Gersey
HNHae Young Noh
Hae Young Noh
Homelessness in U.S. cities has reached unprecedented levels, and effectively allocating limited social service resources requires timely, fine-grained information about where unhoused populations are located and the conditions they face, especially when people are living in vehicles. Identifying these lived-in vehicles, however, is not straightforward. As a first step, it is necessary to reliably detect vehicles in complex urban environments. Existing approaches rely on periodic field surveys or fixed surveillance systems, which fail to capture the highly dynamic and dispersed nature of this population. Prior drive-by sensing efforts have also been limited in scope, focusing mainly on air pollution rather than broader environmental conditions. We introduce NeighborDrive, a multi-modal mobile sensing system that operates on utility and non-profit vehicles already embedded in the communities they serve. The system combines RGB, thermal, gas, air quality, temperature, and GPS sensors to build spatiotemporal maps of neighborhood-level environmental stressors. Because deployments like this must preserve trust and protect privacy, we design a pipeline that anonymizes faces, license plates, and private property before using the data to train a YOLO-based vehicle detection model. In a deployment in Palo Alto, CA, the first stage of our system achieves an F1 score of 0.73 for vehicle detection, more than 10 times higher than an off-the-shelf YOLO baseline. While performance is lower than that of standard RGB-based detection, this gap reflects a trade-off we describe as the cost of privacy.