• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.66 no.12
• /
• pp.1772-1781
• /
• 2017

In this study, the design methodology for alleviating the overfitting problem of Polynomial Neural Networks(PNN) is realized with the aid of two kinds techniques such as L2 regularization and Sum of Squared Coefficients (SSC). The PNN is widely used as a kind of mathematical modeling methods such as the identification of linear system by input/output data and the regression analysis modeling method for prediction problem. PNN is an algorithm that obtains preferred network structure by generating consecutive layers as well as nodes by using a multivariate polynomial subexpression. It has much fewer nodes and more flexible adaptability than existing neural network algorithms. However, such algorithms lead to overfitting problems due to noise sensitivity as well as excessive trainning while generation of successive network layers. To alleviate such overfitting problem and also effectively design its ensuing deep network structure, two techniques are introduced. That is we use the two techniques of both SSC(Sum of Squared Coefficients) and $L_2$ regularization for consecutive generation of each layer's nodes as well as each layer in order to construct the deep PNN structure. The technique of $L_2$ regularization is used for the minimum coefficient estimation by adding penalty term to cost function. $L_2$ regularization is a kind of representative methods of reducing the influence of noise by flattening the solution space and also lessening coefficient size. The technique for the SSC is implemented for the minimization of Sum of Squared Coefficients of polynomial instead of using the square of errors. In the sequel, the overfitting problem of the deep PNN structure is stabilized by the proposed method. This study leads to the possibility of deep network structure design as well as big data processing and also the superiority of the network performance through experiments is shown.

• Fire Science and Engineering
• /
• v.25 no.2
• /
• pp.33-38
• /
• 2011

This study is a research for analysis the states of flame resistant treatment and its system which important in fire prevention by ignition delay of early stage of fire. Some problems arose in the flame resistant treatment site were analysed through the meeting with fire officer and businessman related with flame resistant treatment, and visiting the flame resistant treatment places. From the meeting, visiting and reviewing the code system we find the following problems. First one is a problem of excessive competition with many company in small market of flame resistant treatment and to reduce the cost, the businessman who have no professional technic do not work along the specification. Second one is the inspection system which do not inspect on site but inspect by sampling the flame resistant treated materials submitted by businessman. And there are some problems arose about the reliability and consistency of flame resistant performance of the sample and the treated site. To solve the above problems, we suggest some countermeasure.

• Computers and Concrete
• /
• v.33 no.2
• /
• pp.175-193
• /
• 2024

Researchers have embarked on an active investigation into the feasibility of adopting alternative materials as a solution to the mounting environmental and economic challenges associated with traditional concrete-based construction materials, such as reinforced concrete. The examination of concrete's mechanical properties using laboratory methods is a complex, time-consuming, and costly endeavor. Consequently, the need for models that can overcome these drawbacks is urgent. Fortunately, the ever-increasing availability of data has paved the way for the utilization of machine learning methods, which can provide powerful, efficient, and cost-effective models. This study aims to explore the potential of twelve machine learning algorithms in predicting the tensile strength of geopolymer concrete (GPC) under various curing conditions. To fulfill this objective, 221 datasets, comprising tensile strength test results of GPC with diverse mix ratios and curing conditions, were employed. Additionally, a number of unseen datasets were used to assess the overall performance of the machine learning models. Through a comprehensive analysis of statistical indices and a comparison of the models' behavior with laboratory tests, it was determined that nearly all the models exhibited satisfactory potential in estimating the tensile strength of GPC. Nevertheless, the artificial neural networks and support vector regression models demonstrated the highest robustness. Both the laboratory tests and machine learning outcomes revealed that GPC composed of 30% fly ash and 70% ground granulated blast slag, mixed with 14 mol of NaOH, and cured in an oven at 300°F for 28 days exhibited superior tensile strength.

• Korean Journal of Construction Engineering and Management
• /
• v.5 no.6 s.22
• /
• pp.101-109
• /
• 2004

The construction engineering field is less than $3\~5\%$ of the gross construction cost but engineering's role is very important because the performance of final facility is decided in the engineering stage. Nevertheless, before now government and private compares are more concentrated on construction works than making strategy to strengthen the competitiveness and technical development of investment in construction engineering sectors. Therefore this research have compared and analyzed the competitiveness of domestic construction engineering companies throughout analytic method 'SWOT' to those developing countries, which are potential big market objects, such as Russia Libya and Indonesia Such research was obtained through business trip & consultation with related specialists for the difficulties of domestic companies in advancing into overseas market. Moreover this research have analyzed the technical improvement of construction engineering companies and reason for competitiveness weaken by various aspects, it provides governmental and private plans to advance into overseas market Especially, to strengthen domestic companies into advancing overseas market it has suggested that tax & subsidy privileges, supporting plans for obtaining information of overseas and strategies cooperation with other organizations.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.57 no.10
• /
• pp.1782-1789
• /
• 2008

This paper presents the design of acoustic resonance(AR) free and dimmable electronic ballast for 1kW Metal-Halide Lamp(MHL). The proposed Ballast consists of a Full-Bridge(FB) rectifier, a passive power factor correction(PFC) circuit, a full-bridge inverter, an ignitor using LC resonance and a control circuit for frequency modulation and dimming control. Whereas a passive PFC provides advantages in terms of high reliability and low cost for constructing the circuit, it is difficult to supply a stable voltage because of the output voltage ripple that occurs with a period of 120Hz. Although the ballast can be designed with a small size and a light weight if it is driven at a switching frequency between 1 and 100 kHz, AR will occur if the eigenvalue frequency of the lamp coincides with the inverter's operation frequency. The operation frequency was modulated in real time according to the output voltage ripple to compensate for the variation in power supplied to the lamp and eliminate AR. For dimming, the method, which modulated drive frequency of FB inverter using the control of DC level by microprocessor, was used. The Dimming ranged at least from 600W to 1kw as rated power of the lamp with 4 stages. Performance of the proposed technique was validated through numerical analysis, computer simulation using Pspice and by applying it to an electronic ballast for a prototype 1kW MHL.

• The Journal of the Korea Contents Association
• /
• v.17 no.2
• /
• pp.269-277
• /
• 2017

The theses is an analytical comparison research between 'Skills Link' and 'Do Dream' : the former is an effective program to establish a variety of intervention strategies and the institutional foundation for employment support of disadvantaged youth, performed by ESDC in the level of the nation, and the latter is the program which is carried out by K-Dream Center of MGEF in South Korea. As for the research method, comparison and analysis have been fulfilled, focusing on the employment support program. The research includes program background, goal, participation subject, selection standard, program contents, support period, support cost, and program effect. It suggests that we should design the systematic program for the youth in the transition to independency and perform constant support. It is necessary to select beneficiary on the consideration of the characteristics and situation of the disadvantaged youth. In addition, it is required to secure consistency of the program and enlarge the support period. For the sake of successful performance of the independent living program, it is a necessity to motivate the youth and introduce the ways to encourage the business sector to take part in.

• Journal of Navigation and Port Research
• /
• v.43 no.6
• /
• pp.395-402
• /
• 2019

The development and operation of the MASS (Maritime Autonomous Surface Ship) is being actively discussed for more efficient and safer maritime transportation solutions. The autonomous navigation technology has positive aspects such as the prevention of marine accidents, improvement of fuel efficiency of ships and cost reduction, and negative aspects such as job loss, task change, and security problems. It is expected that there will be new human element issues such as the situation awareness of remote operators, because the shore-based control will be conducted when fully autonomous ships are in operation. In this paper, we consider major human element issues that should be factored in the development and operation of MASS, and suggest a method of HRA (Human Reliability Analysis) for P IFs (Performance Influencing Factors) of the remote operators that are expected according to the shore-based control.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.17 no.11
• /
• pp.2736-2744
• /
• 2013

In this paper, we present a Smart Water Treatment Plant using Zigbee USN devices and a real-time monitoring system in K-water Flow Meter Calibration Center Building. For verification, the data of vertical type WTP such as flow rate, pressure, water level and water temperature are obtained by the Zigbee USN devices, operating in 2.45 GHz band, and be wirelessly surveilled by the real-time monitoring system. The received data from the sensor is transmitted to the data processing device, and then the processed data can be monitored on a smart phone. Consequently, the pilot plant based on the low-cost and high-efficiency USN has been developed with the performance analysis for the communication network and remote monitoring system on mobile devices.

• Journal of Internet Computing and Services
• /
• v.21 no.1
• /
• pp.159-167
• /
• 2020

The Korean government has pursued various policies to revitalize the cloud industry since 2009. In particular, the government enacted the 'Cloud Computing and Privacy Security Act' for the first time in the world in 2015, creating institutions and support programs to boost the cloud industry. As a result, the volume of the cloud industry has increased significantly, and the number of companies and professionals have increased. In this paper, the researches analyzed the situation of Korean cloud market, usage of cloud computing in enterprises, effects and problems of cloud-enabled companies based on the industrial survey conducted from 2016 to 2019, In this study, the results showed concerns of security, uncertainty about cost savings, and IT capabilities of enterprises as problems in revitalizing the cloud industry. The researches also propose strategies, perception changes, and development processes for companies to resolve the problems.

• Korean Journal of Construction Engineering and Management
• /
• v.12 no.3
• /
• pp.31-41
• /
• 2011

The $LCCO_2$ evaluation programs previously developed in Korea involve limitations in establishing strategies of reducing environmental loads to optimal level in a way to put in materials directly after designing. Therefore, this study has the purpose to extract and propose elements required for the establishment of highly accurate system by counteracting swiftly in a method with high energy efficiency over cost at planning stage. To that end, existing $LCCO_2$ evaluation programs in Korea and abroad were compared and analyzed, and in the planning stage, GEM-21P and Carbon-navigator intended for the establishment of building energy performance improving strategy were selected as the evaluation program for survey. On such basis, after comparison and analysis between $LCCO_2$ calculating methods and system structures of the two programs, elements required for system establishment that can evaluate life-cycle environmental loads of building in planning stage were proposed.