• 응용통계연구
• /
• 제27권5호
• /
• pp.773-786
• /
• 2014

본 연구는 ROC 곡선에서 형성되는 면적 형태로 나타나는 분류정확도기준인 오분류율곱(multiplication of false rates; MFR)를 제안한다. MFR 기준과 다른 기준로부터 구한 최적분류점의 분류성과에 대하여 비교 분석한다. 다양한 분포함수에 대하여 최적분류점을 구하고 이에 대응하는 FNR과 FPR을 비교하면서 MFR의 특징과 장점을 유도한다. 일반적인 비용함수를 바탕으로 분류점에 대한 비용비율을 다양한 분류기준을 이용하여 구한다. 비용곡선에 대한 비용비율의 관계를 정리하여 MFR 기준의 장점을 탐색한다. MFR 기준의 정의를 다차원 ROC 분석으로 확장하고 다차원의 다른 분류기준과의 관계를 설명하면서 토론한다.

• 전자공학회논문지SC
• /
• 제46권3호
• /
• pp.15-21
• /
• 2009

본 논문에서는 변수 불확실성을 가지는 이산시간 특이시스템의 강인 안정화 기법과 강인 보장비용 제어기법을 다룬다. 제안하는 제어기법은 제어기 존재조건에서 준정부호조건(semi-definite condition)이나 시스템 행렬의 분해 없이 볼록최적화(convex optimization)가 가능한 선형행렬부등식 접근방법을 이용하여 제안한다. 먼저, 강인 안정화 상태궤환 제어기는 폐루프 시스템의 정규성, 코잘 및 안정화를 만족하는 제어기의 존재조건과 설계방법을 선형행렬부등식으로 제시한다. 그리고 보장비용 함수의 상한치의 최소화를 보장하는 강인 보장비용 제어기 설계방법은 강인 안정화 제어기 설계를 기반으로 제안한다. 예제를 통하여 제안한 제어기 설계기법의 타당성을 확인한다

• International Journal of Naval Architecture and Ocean Engineering
• /
• 제11권1호
• /
• pp.606-623
• /
• 2019

The path following problem of a ship sailing in restricted waters under wind effect is investigated based on Robust $H_{\infty}$ Guaranteed Cost Control (RHGCC). To design the controller, the ship maneuvering motion is modeled as a linear uncertain system with norm-bounded time-varying parametric uncertainty. To counteract the bank and wind effects, the integral of path error is augmented to the original system. Based on the extended linear uncertain system, sufficient conditions for existence of the RHGCC are given. To obtain an optimal robust $H_{\infty}$ guaranteed cost control law, a convex optimization problem with Linear Matrix Inequality (LMI) constraints is formulated, which minimizes the guaranteed cost of the close-loop system and mitigates the effect of external disturbance on the performance output. Numerical simulations have confirmed the effectiveness and robustness of the proposed control strategy for the path following goal of a ship sailing in restricted waters under wind effect.

• 제어로봇시스템학회논문지
• /
• 제20권9호
• /
• pp.900-907
• /
• 2014

This paper proposes a MPC (Model Predictive Control) method for the torque and flux controls of induction motor. The proposed MPC method selects the optimized voltage vector for the matrix converter control using the predictive modeling equation of the induction motor and cost function. Hence, the reference voltage vector that minimizes the cost function of the torque and flux error within the control period is selected and applied to the actual system. As a result, it is possible to perform the torque and flux control of induction motor using only the MPC controller without a PI (Proportional-Integral) or hysteresis controller. Even though the proposed control algorithm is more complicated and has lots of computations compared with the conventional MPC, it can perform torque ripple reduction by synthesizing voltage vectors of various magnitude. This feature provides the reduction of amount of calculations and the improvement of the control performance through the adjustment of the number of the unit vectors n. The proposed control method is validated through the PSIM simulation.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권4호
• /
• pp.1504-1526
• /
• 2018

For uplink multi-user massive MIMO systems, conventional minimum mean square error (MMSE) linear detection method achieves near-optimal performance when the number of antennas at base station is much larger than that of the single-antenna users. However, MMSE detection involves complicated matrix inversion, thus making it cumbersome to be implemented cost-effectively and rapidly. In this paper, we first summarize in detail the state-of-the-art simplified MMSE detection algorithms that circumvent the complicated matrix inversion and hence reduce the computation complexity from ${\mathcal{O}}(K^3)$ to ${\mathcal{O}}(K^2)$ or ${\mathcal{O}}(NK)$ with some certain performance sacrifice. Meanwhile, we divide the simplified algorithms into two categories, namely the matrix inversion approximation and the classical iterative linear equation solving methods, and make comparisons between them in terms of detection performance and computation complexity. In order to further optimize the detection performance of the existing detection algorithms, we propose more proper solutions to set the initial values and relaxation parameters, and present a new way of reconstructing the exact effective noise variance to accelerate the convergence speed. Analysis and simulation results verify that with the help of proper initial values and parameters, the simplified matrix inversion based detection algorithms can achieve detection performance quite close to that of the ideal matrix inversion based MMSE algorithm with only a small number of series expansions or iterations.

• 대한교통학회지
• /
• 제19권5호
• /
• pp.61-70
• /
• 2001

급변하는 교통환경 하에서 신뢰성 있는 기·종점통행량(O/D)를 구축하는 것은 매우 어려운 일이다. 특히, 시간과 비용의 제약하에서 교통계획을 수립해야 하는 경우라면 전통적인 O/D 구축방법인 O/D조사를 통해 O/D를 구축하는 것은 현실적으로 불가능한 일이다. 이로 인해 관측통행량을 이용하여 O/D를 추정하는 연구가 활발하게 이루어져 왔다. 그러나 현재까지 진행되어 온 연구들이 개별차량 O/D를 추정하는 대에만 국한되어 있어, 관측통행량으로 대중교통 O/D를 추정하는 연구는 거의 없는 실정이다. 본 연구는 이미 개별차량 O/D 추정에서 신뢰성이 검증된 Gradient방법을 대중교통 O/D 추정에 적용하여 그 신뢰성을 평가하는데 목적이 있다. 본 연구에서 대중교통 통행배정은 Spiess-Florian의 대중교통 통행배정규칙을 이용하였고, Gradient방법의 신뢰성을 평가하는 기준으로는 기본 O/D와 추정 O/D의 분포형태를 이용하였다.

• International journal of advanced smart convergence
• /
• 제11권1호
• /
• pp.111-116
• /
• 2022

After COVID-19, the number of people with sleep disorders around the world is increasing. In particular, in the flow of the 4th industrial revolution, the differentiation of types and characteristics of the sleep industry is accelerating. Therefore, in this study, the characteristics of each type of sleep-related industry were reclassified from an industrial point of view, and based on this, an attempt was made to review the classification system that can help companies develop sleep products and improve related national systems. Based on the 10th standard industry classification, we compared input cost, value, and usability and analyzed common characteristics, treatments, and preventive effects based on this. A comprehensive taxonomy using matrix analysis was reviewed. As a result, in terms of cost (A), the most common sleeping products are general mattresses and general bedding. It is an IOT device (auxiliary device), and the value aspect (B, B/D) included sleep cafe, bedding rental and management service, and sleep consulting. In terms of utility (A/B), a total of 6 product groups including sleep aids (health functional foods) belong to this category, and in terms of treatment (A/C), a total of 3 product groups including sleep clinics (medical services) belong to this category. As for the product group (A/D) with both properties, it was found that non-insurance sleep treatment medical devices, sleep-related over-the-counter drugs, and some sleep monitoring applications belong to this category. Ultimately, it was found that the sleep industry classification enables the most active product development and composition according to the relative relationship between cost and utility, and treatment and utility. appeared to be necessary.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2004년도 학술대회 논문집 정보 및 제어부문
• /
• pp.708-710
• /
• 2004

This paper studies a multirate sampled-data control for LTI systems by using the digital redesign (DR) method. In this note, to well tackle the problem associated with both the state matching and the stabilization, our nobel strategy is to minimize the linear quadratic cost function. The main features of the proposed method are that i) the delta-operator-based descretization method is applied to improve the state-matching performance in the fast sampling limit and/or the large input multiplicity; ii) the proposed multirate control scheme can improve the state-matching performance in the long sampling limit; iii) some sufficient conditions that guarantee the stability of the closed-loop discrete-time system and provide a guarantee cost for the cost function can be formulated in the LMIs format; and iv) an optimal sampled-data controller in the sense of minimizing the upper bound of the cost function is also given by means of an LMI optimization procedure.

• 한국관광식음료학회지:관광식음료경영연구
• /
• 제16권2호
• /
• pp.17-31
• /
• 2005

Recently, hotel industry has realized the importance of food and beverage sales for the profit maximization, and the focuses on restaurant management has been growing. Accordingly, menu management in the F/B department is one of the most key factors determining the success or failure of business. Therefore, in this study, entree menu items of french restaurant in the deluxe hotel was analysed with presently theorized model of menu analysis, classified into four menu items. Also it was analyzed how those classified menu items influence on sales, number of sold, food cost percentage, contribution margin And, proper ways was presented to make restaurant managers and menu planner in order to increase food operation sales through proper modifications and methods on various menu analysis matrix. In Pavesic's menu analysis method, all of menu items have impact on the sales, number of sold, contribution margin and Primes, Sleepers do so on the food cost. The finding of this study was that Pavesic's menu analysis is superior to menu analysis in terms of the sales, number of sold, food cost percentage, contribution margin. Therefore, Pavesic's menu analysis is useful and efficient method in order to conduct menu engineering.

• 전기학회논문지
• /
• 제60권6호
• /
• pp.1091-1096
• /
• 2011

Currently, the concern about the environment is the issue all over the world, and in particular, carbon emissions of the power plants will not be able to disregard from the respect of generation cost. This paper proposes DR (demand response) as a method of reducing carbon emissions and therefore, carbon emissions cost. There are a number of studies considering DR, and in this paper, the effect of DR is focused on the side of carbon emission reduction effect considering Time-Of-Use (TOU) program, which is one of the most important economic methods in DSM. Demand-price elasticity matrix is used in this paper to model and analyze DR effect. Carbon emissions is calculated by using the carbon emission coefficient provided by IPCC (Intergovernmental Panel on Climate Change), and generator's input-output characteristic coefficients are also used to estimate carbon emission cost as well as the amount of carbon emissions. Case study is conducted on the RBTS IEEE with six buses. For the TOU program, it is assumed that parameters of time period partition consist of three time periods (peak, flat, off-peak time period).