In this paper, we presented the experimental basis for the theoretical background and robustness of the Convolutional Neural Network for object recognition based on artificial intelligence. An experimental result was performed to visualize the weighting filters and feature maps for each layer to determine what characteristics CNN is automatically generating. experimental results were presented on the trend of learning error and identification error rate by checking the relevance of the weight filter and feature map for learning error and identification error. The weighting filter and characteristic map are presented as experimental results. The automatically generated characteristic quantities presented the results of error rates for moving and rotating robustness to geometric changes.

This aim of this study examined the relationship among the interest in anti-aging, health promotion behaviors and the need for anti-aging services by their 20s. Survey was conducted among adults from their 20s in the Changwon City. 228 responses were used for analysis. The overall average score of the interest and effort of anti-aging were 2.97 point and 2.62 point (out of 5), respectively. And those were both higher in female than men (p<0.01~p<0.001). The overall average score of need for anti-aging service was 3.50 point(total score is 5). In The demand for each area of anti-aging service were ≥3.5 point in all 5 areas, and stress management (4.00 point) was the highest, while the beauty management (3.60 point) was the lowest. There were significant differences in all five areas by sex (p<0.01~p<0.001). The overall score of the performance of health promotion behaviors was 2.44 point(total score is 4), and the interpersonal relationship score (2.85 point) was the highest, while the health responsibility score (2.08 point) was the lowest. The interest in anti-aging and performance of health promotion behaviors showed positive relationship to anti-aging services, and their explanation powers were 34.6% (p<0.001). The results of this study suggest be used as data to establish strategies revitalizing various anti-aging service in the twenties.

In the Port Logistics system, Container Stowage planning is an important issue for cost-effective efficiency improvements. At present, Planners are mainly carrying out Stowage planning by manual or semi-automatically. However, as the trend of super-large container ships continues, it is difficult to calculate an efficient Stowage plan with manpower. With the recent rapid development of artificial intelligence-related technologies, many studies have been conducted to apply enhanced learning to optimization problems. Accordingly, in this paper, we intend to develop and present a Deep Q-Learning Network model for the Master Stowage planning of Container ships.

In this study, approximate design optimization using various meta-models was performed for the structural design of active type deck support frame. The active type deck support frame was newly developed to facilitate both transportation and installation of 20,000 ton class offshore plant topside. Structural analysis was carried out using the finite element method to evaluate the strength performance of the active type deck support frame in its initial design stage. In the structural analysis, the strength performances were evaluated for various design load conditions that were regulated in ship classification organization. The approximate optimum design problem based on meta-model was formulated such that thickness sizing variables of main structure members were determined by achieving the minimum weight of the active type deck support frame subject to the strength performance constraints. The meta-models used in the approximate design optimization were response surface method, Kriging model, and Chebyshev orthogonal polynomials. The results from approximate design optimization were compared to actual non-approximate design optimization. The Chebyshev orthogonal polynomials among the meta-models used in the approximate design optimization represented the most pertinent optimum design results for the structure design of the active type deck support frame.

The purpose of this study is to suggest the optimal shape for a local air ventilation system for fume removal, which is operated in a zinc galvanizing factory, and to propose the improvement plan for a ventilation system used in a zinc galvanizing factory through flow analysis. A part of the air sprayed by an air curtain goes out. It will be necessary to research the position of an air curtain, its spray angles, and its nozzle shape. In addition, additional research needs to be conducted on the shape of the fan installed before a hood in order to make it easy to induce fume. In a local air ventilation system, air is inhaled from the outside. The higher an inlet negative pressure is, the easier fume is removed. It was found that it was necessary to install an appropriate hole in the wall on the back of a push nozzle in order to reduce an inlet negative pressure.

The purpose of this study is to develop a damper that protects against the dangers of tidal waves since there's no function to block the inflow of large amounts of water into the inside When natural disasters such as tidal waves occur. Therefore, it intended to derive the design data by simulating through flow analysis in order to predict the pressure that a damper configured to open and close manually or automatically receives. It examined the preceding researches first and conducted the flow analysis, to predict the force of the damper installed on the bottom of the building's outside to prevent the inflow of seawater into the inside when natural disaster occurring. As a result, it showed that, in the event of a tsunami, it moved about 170m and the time impacting the damper occurred within about eight seconds, and, at the moment, the damper door was pressured about 17bar. Also, it could identify that the load was approximately 900kN and the force by the fluid was applied to the damper door.

With the recent development of bathemetry technology, the hydrographic surveying method has been changed from single beam depth device use to multi beam acoustic sounding technology. Also, various studies have been reported to obtain high accuracy and precision in the process of river bed topographic data. Especially south korea is geographically on three sides of the sea and the river topography is very developed. To build information about the underwater, and riverbed status, the public investigations has been continuously progressed. In this study, We investigasted the riverbed topography near Baeckma river leisure park. for this purpose, In this study, as the first preliminary survey, location of navigational dangerous objects and reefs and the dangerous areas are identified. Also, ground control points is selected for the optimal GPS surveying. Secondary, through test surveying the Gain, TVG, and pulse length are determined. In addition, the investigation of dangerous objects for navigation is also conducted. As the last step, the error analysis are conducted for the acquired data, and this process involves the removal and adjustments of errors. This section includes the analysis of tide level and navigational contributions, and finally generates a submarine topographic map.

The widely used low-cost design methodology for IoT devices is very popular. In such a networked device, memory is composed of flash memory, SRAM, DRAM, etc., and because it processes a large amount of data, memory design is an important factor for system performance. Therefore, each device selects optimized design factors such as function, performance and cost according to market demand. The design of a memory architecture available for low-cost IoT devices is very limited with the configuration of SRAM, flash memory, and DRAM. In order to process as much data as possible in the same space, an architecture that supports parallel processing units is usually provided. Such parallel architecture is a design method that provides high performance at low cost. However, it needs precise software techniques for instruction and data mapping on the parallel architecture. This paper proposes an instruction/data mapping method to support optimized parallel processing performance. The proposed method optimizes system performance by actively using hardware and software parallelism.