• 품질경영학회지
• /
• 제29권3호
• /
• pp.66-78
• /
• 2001

This paper deals with problem of determining process target value under automated visual inspection(AVI) system. Three independent error sources - digitizing error, illumination error, and positional error - which have a close relationship with the performance of the AVI system, are considered. Assuming that digitizing error is uniformly or normally distributed and illumination and positional errors are normally distributed, respectively, the distribution function for the error of measured lengths is derived when the length of a product is measured by the AVI system. Then, Optimal target values under two error models of AVI system are obtained by minimizing the total expected cost function which consists of give away, rework and penalty cost. To validate two process setting models, AVI system for drinks filling process is made up and test results are discussed.

• 정보처리학회논문지D
• /
• 제15D권2호
• /
• pp.207-212
• /
• 2008

본 논문은 소프트웨어 비용추정 모델의 적합성을 평가하고, 가장 적합한 모델을 제시하였다. 먼저, 해당 모델의 함수를 변수변환시켜 선형식으로 만든다. 다음으로 실제 개발 소프트웨어의 비용 데이터가 모델의 선형식에 얼마나 적합한지로 모델의 성능을 평가한다. 모델 성능평가에는 절대오차 대신 상대오차 개념인 MMRE를 적용하였다. 기존의 소프트웨어 비용추정 모델은 Weibull, Gamma와 Rayleigh 함수를 따르고 있다. 본 논문에서는 성장곡선의 일종인 Gompertz 곡선 모델을 제안하였다. 추가로 다른 성장곡선들도 적합성을 검증하였다. 모델 성능평가 결과 Gompertz 성장곡선이 소프트웨어 비용추정 모델로 가장 적합한 성능을 보였다.

• 한국지능시스템학회:학술대회논문집
• /
• 한국지능시스템학회 2007년도 추계학술대회 학술발표 논문집
• /
• pp.183-186
• /
• 2007

We introduction some properties for fuzzy power function of performance of a test. First we define fuzzy type I error and type II error for the probability of the two types of error. And we show that an fuzzy error probability of one kind can only be reduced at cost of increasing the other fuzzy error probability.

• 제어로봇시스템학회논문지
• /
• 제5권7호
• /
• pp.808-813
• /
• 1999

A fuzzy learning method to control an unstable and multivariable system is presented in this paper, Because the multivariable system has generally a coupling effect between the inputs and outputs, it is difficult to find its modeling equation or parameters. If the system is unstable, initial condition rules are needed to make it stable because learning is nearly impossible. Therefore, this learning method uses the initial rules and introduces a cost function composed of the actual error and error-rate of each output without the modeling equation. To minimize the cost function, we experimentally got the Jacobian matrix in the operating point of the system. From the Jacobian matrix, we can find the direction of the convergence in the learning, and the optimal control rules are finally acquired when the fuzzy rules are updated by changing the portion of the errors and error rates.

• 대한수학회지
• /
• 제33권2호
• /
• pp.235-247
• /
• 1996

Many numerical computations in science and engineering can only be solved approximately since the available infomation is partial. For instance, for problems defined ona space of functions, information about f is typically provided by few function values, $N(f) = [f(x_1), f(x_2), \ldots, f(x_n)]$. Knwing N(f), the solution is approximated by a numerical method. The error between the true and the approximate solutions can be reduced by acquiring more information. However, this increases the cost. Hence there is a trade-off between the error and the cost.

• 대한기계학회논문집A
• /
• 제20권2호
• /
• pp.399-407
• /
• 1996

A fuzzy learning algorithm to get the optimal fuzzy rules is presented in this paper. The algorithm introduces a reference model to generate a desired output and a performance index funtion instead of the performance index table. The performance index funtion is a cost function based on the error and error-rate between the reference and plant output. The cost function is minimized by a gradient method and the control input is also updated. In this case, the control rules which generate the desired response can be obtained by changing the portion of the error-rate in the cost funtion. In SISO(Single-Input Single- Output)plant, only by the learning delay, it is possible to experss the plant model and to get the desired control rules. In the long run, this algorithm gives us the good control rules with a minimal amount of prior informaiton about the environment.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2000년도 추계학술대회 논문집 학회본부 D
• /
• pp.747-749
• /
• 2000

In this paper, we proposed color stereo matching algorithm using dynamic programming. The conventional gray stereo matching algorithms show blur at depth discontinuities and non-existence of matching pixel in occlusion lesions. Also it accompanies matching error by lack of matching information in the untextured region. This paper defines new cost function makes up for the problems happening in conventional gray stereo matching algorithm. New cost function contain the following properties. I) Edge points are corresponded to edge points. ii) Non-edge points are corresponded to non-edge points. iii) In case of exiting the amount of edges, the cost function has some weight in proportion to path distance. Proposed algorithm was applied in various images obtained by parallel camera model. As the result, proposed algorithm showed improved performance in the aspect of matching error and processing in the occlusion region compared to conventional gray stereo matching algorithms.

• 한국수자원학회논문집
• /
• 제35권2호
• /
• pp.187-194
• /
• 2002

하천유역 면적강우량 산정의 정확도를 개선하기 위하여 기존 강우관측자료의 통계적 특성을 이용한 강우관측망의 최적설계방법을 연구하였다. 최적설계를 위한 목적함수는 면적강우량의 추정오차 및 지점강우량 관측비용의 항으로 구성하고, 그 값이 최소인 관측망은 선정하였다. 통계f7파의 추정방법으로는 통계적 분산 산정방법인 크리깅 모형을 채택하였다. 비용은 강우관측소의 설치비와 연간운영 비론 적용하고, 오차항과 비용항의 통합에는 등치매개변수를 이용하였다. 연구된 최적설계방법을 댐 신설로 강우관측소 증설이 필요한 용담댐 유역에 적용하여, 대상유역의 최적 강우관측망을 제안하였다.

• ETRI Journal
• /
• 제17권1호
• /
• pp.11-22
• /
• 1995

This paper proposes a modified error function to improve the error back-propagation (EBP) algorithm for multi-Layer perceptrons (MLPs) which suffers from slow learning speed. It can also suppress over-specialization for training patterns that occurs in an algorithm based on a cross-entropy cost function which markedly reduces learning time. In the similar way as the cross-entropy function, our new function accelerates the learning speed of the EBP algorithm by allowing the output node of the MLP to generate a strong error signal when the output node is far from the desired value. Moreover, it prevents the overspecialization of learning for training patterns by letting the output node, whose value is close to the desired value, generate a weak error signal. In a simulation study to classify handwritten digits in the CEDAR [1] database, the proposed method attained 100% correct classification for the training patterns after only 50 sweeps of learning, while the original EBP attained only 98.8% after 500 sweeps. Also, our method shows mean-squared error of 0.627 for the test patterns, which is superior to the error 0.667 in the cross-entropy method. These results demonstrate that our new method excels others in learning speed as well as in generalization.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제14권1호
• /
• pp.40-52
• /
• 2020

In this paper, we consider a precoder design problem for multiuser interference channel. Most of the conventional schemes for precoder design utilize a signal-to-interference-plus-noise ratio (SINR) as a cost function. However, since the SINR metric of a desired transmitter-receiver pair is a function of precoding vectors of all transmitters in the multiuser interference channel, an analytic closed-form solution is not available for the precoding vector of a desired transmitter that maximizes the SINR metric. To eliminate coupling between the precoding vectors of all transmitters and to find a closed-form solution for the precoding vector of the desired transmitter, we use a signal-to-leakage-plus-noise ratio (SLNR) instead as a cost function because the SLNR at a transmitter is a function of the precoding vector of the desired transmitter only. In addition, channel estimation errors for undesired links are considered when designing the precoding vector because they are inevitable in a multiuser interference channel. In this case, we propose a design scheme for the precoding vector that is robust to the channel estimation error. In the proposed scheme, the precoding vector is designed to maximize the worst-case SLNR. Through computer simulation, we show that the proposed scheme has better performance than the conventional scheme in terms of SLNR, SINR, and sum rate of all users.