PURPOSES :This study aims to improve complex modeling of multivariable, nonlinear, and overdispersion data with an artificial neural network that has been a problem in the civil and transport sectors.METHODS :Deep learning, which is a technique employing artificial neural networks, was applied for developing a large bus fuel consumption model as a case study. Estimation characteristics and accuracy were compared with the results of conventional multiple regression modeling.RESULTS :The deep learning model remarkably improved estimation accuracy of regression modeling, from R-sq. 18.76% to 72.22%. In addition, it was very flexible in reflecting large variance and complex relationships between dependent and independent variables.CONCLUSIONS :Deep learning could be a new alternative that solves general problems inherent in conventional statistical methods and it is highly promising in planning and optimizing issues in the civil and transport sectors. Extended applications to other fields, such as pavement management, structure safety, operation of intelligent transport systems, and traffic noise estimation are highly recommended.

선박의 항행에 의해 발생되는 항주파의 특성은 선박의 속도와 수심 조건에 따라 크게 달라진다. 연안 항로에서 발생된 항주파는 주변 해안으로 전파됨에 따라 항만 내의 정온 수역을 교란하여 정박 중인 소형선박, 수영객 등에 돌발적이고도 심각한 위험을 가져다 줄뿐만 아니가 해안의 침식, 호안의 결괴 등의 피해를 주기도 한다. 지금까지 항주파에 관한 연구의 관심사는 일정 수심 조건에 대해 조파저항이나 조선에 미치는 영향을 분석하는 등 주로 조선공학도의 관점에서 검토가 대상이 되어 왔으며, 가변 수심을 가진 실제 해역에 있어서의 항주파 발생과 전파에 기인한 주변 해역의 영향은 그다지 검토되지 못하였다. 최근 고속선 등의 발달로 인해 천해역에서의 항주파로 인한 인근 해역의 피해가 더욱 우려되고 있는 추세이다. 따라서 실제 수역에서의 항주파의 발달과 그 전파과정은 조사할 필요가 있는 것이다. 본 연구에서는 연안해역의 얕고 복잡한 수로와 다양한 선속 조건에 대한 항주파의 발생 및 전파를 예측하기 위하여 고정 좌표계에서 Boussinesq 방정식을 토대로 항주파 수치예측 모형을 구축하였다. 제안된 모형은 수리모형실험 결과와의 비교를 통하여 검증하였으며, 또한 실제 수로를 토대로 한 가변 수심역에 개발된 모형을 적용하여 수신 변화 고려의 중요성을 확인하였다.

To design a coastal structure in the nearshore region, engineers must have means to estimate wave climate. Waves, approaching the surf zone from offshore, experience changes caused by combined effects of bathymetric variations, interference of man-made structure, and nonlinear interactions among wave trains. This paper has attempted to find out the effects of two of the more subtle phenomena involving nonlinear shallow water waves, amplitude dispersion and secondary wave generation. Boussinesq-type equations can be used to model the nonlinear transformation of surface waves in shallow water due to effects of shoaling, refraction, diffraction, and reflection. In this paper, generalized Boussinesq equations under the complex bottom condition is derived using the depth averaged velocity with the series expansion of the velocity potential as a product of powers of the depth of flow. A time stepping finite difference method is used to solve the derived equation. Numerical results are compared to hydraulic model results. The result with the non-linear dispersive wave equation can describe an interesting transformation a sinusoidal wave to one with a cnoidal aspect of a rapid degradation into modulated high frequency waves and transient secondary waves in an intermediate region. The amplitude dispersion of the primary wave crest results in a convex wave front after passing through the shoal and the secondary waves generated by the shoal diffracted in a radial manner into surrounding waters.