• 한국통신학회논문지
• /
• 제19권12호
• /
• pp.2418-2425
• /
• 1994

기울기를 따라가는 방식(gradient descent method)에 바탕을 둔 오류 역전파(EBP : Error Back Propagation) 방법이 가장 널리 사용되는 신경회로망의 학습 방법에서 문제가 되는 지역 최소값(local minima), 느린 학습 시간, 신경망 구조(structure), 그리고 초기의 연결 강도(interconnection weight) 등을 기존의 다층 신경 회로망에 지역적인 학습 능력을 가진 가우시안 포텔샵 네트워크(GPFN : Gaussian Potential Function Networks)를 병렬적으로 부가하여 해결함으로써 지역화된 오류 학습 패턴들이 나타내는 문제에 대하여 학습 성능을 향상시킬 수 잇는 새로운 학습 방법을 제시한다. 함수 근사화 문제에서 기존의 EBP 학습 방법과의 비교 실험으로 제안된 학습 방법이 보다 개선된 일반화 능력과 빠른 학습 속도를 가짐을 보여 그 효율성을 입증한다.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 1998년도 추계종합학술대회 논문집
• /
• pp.1281-1284
• /
• 1998

This paper proposes an efficient time series prediction of the nonlinear dynamical discrete-time systems using multilayer neural networks of a hybrid learning algorithm. The proposed learning algorithm is a hybrid backpropagation algorithm based on the steepest descent for high-speed optimization and the dynamic tunneling for global optimization. The proposed algorithm has been applied to the y00 samples of 700 sequences to predict the next 100 samples. The simulation results shows that the proposed algorithm has better performances of the convergence and the prediction, in comparision with that using backpropagation algorithm based on the gradient descent for multilayer neural network.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2018년도 학술발표회
• /
• pp.448-448
• /
• 2018

In this study, the regression models (Load ESTimator and eight-parameter model) were evaluated to estimate instantaneous pollutant loads under various criteria and optimization methods. As shown in the results, LOADEST commonly used in interpolating pollutant loads could not necessarily provide the best results with the automatic selected regression model. It is inferred that the various regression models in LOADEST need to be considered to find the best solution based on the characteristics of watersheds applied. The recently developed eight-parameter model integrated with Genetic Algorithm (GA) and Gradient Descent Method (GDM) were also compared with LOADEST indicating that the eight-parameter model performed better than LOADEST, but it showed different behaviors in calibration and validation. The eight-parameter model with GDM could reproduce the nitrogen loads properly outside of calibration period (validation). Furthermore, the accuracy and precision of model estimations were evaluated using various criteria (e.g., $R^2$ and gradient and constant of linear regression line). The results showed higher precisions with the $R^2$ values closed to 1.0 in LOADEST and better accuracy with the constants (in linear regression line) closed to 0.0 in the eight-parameter model with GDM. In hence, based on these finding we recommend that users need to evaluate the regression models under various criteria and calibration methods to provide the more accurate and precise results for pollutant load estimations.

• 한국지능시스템학회논문지
• /
• 제12권4호
• /
• pp.334-340
• /
• 2002

본 논문에서는 상대적으로 큰 퍼지 엔트로피를 갖는 입력-출력 데이터 집단에 다중 회귀 분석을 적용하여 다차원 평면 클러스터를 생성하고, 이 클러스터를 새로운 퍼지 모델의 규칙으로 추가한 후 모델 파라미터의 개략 동조와 정밀 동조를 반복 수행하는 자기구성 퍼지 모델링을 제안한다 Weighted recursive least squared 알고리즘과 fuzzy C-regression model 클러스터링에 의해 퍼지 모델의 파라미터를 개략적으로 동조한 후 gradient descent 알고리즘에 의해 파라미터를 정밀 동조하면서 감수분열 유전 알고리즘을 이용하여 최적의 학습률을 탐색한다. 그리고, 자기구성 퍼지 모델링 기법을 이용하여 Box-Jenkins의 가스로 데이터, 비선형 다변수 정적 함수의 데이터, 하수처리 활성오니 공정과 Mackey-Glass 시계열 데이터의 모델링을 수행하고, 기존의 방법에 의한 모델링 결과와 비교하여 그 성능을 입증한다.

• Structural Engineering and Mechanics
• /
• 제45권1호
• /
• pp.53-67
• /
• 2013

This study aimed to develop a model to accurately predict the acceleration of structural systems during an earthquake. The acceleration and applied force of a structure were measured at current time step and the velocity and displacement were estimated through linear integration. These data were used as input to predict the structural acceleration at next time step. The computation tool used was the Volterra/Wiener neural network (VWNN) which contained the mathematical model to predict the acceleration. For alleviating problems of relatively large-dimensional and nonlinear systems, the VWNN model was utilized as the signal processing tool, including the Taylor series components in the input nodes of the neural network. The number of the intermediate layer nodes in the neural network model, containing the training and simulation stage, was evaluated and optimized. Discussions on the influences of the gradient descent with adaptive learning rate algorithm and the Levenberg-Marquardt algorithm, both for determining the network weights, on prediction errors were provided. During the simulation stage, different earthquake excitations were tested with the optimized settings acquired from the training stage to find out which of the algorithms would result in the smallest error, to determine a proper simulation model.

• 한국소음진동공학회논문집
• /
• 제14권4호
• /
• pp.267-275
• /
• 2004

This research is concerned with the development of a real-time adaptive PPF controller for the active vibration suppression of smart structure. In general, the tuning of the PPF controller is carried out off-line. In this research, the real-time learning algorithm is developed to find the optimal filter frequency of the PPF controller in real time and the efficacy of the algorithm is proved by implementing it in real time. To this end, the adaptive algorithm is developed by applying the gradient descent method to the predefined performance index, which is similar to the method used popularly in the optimization and neural network controller design. The experiment was carried out to verify the validity of the adaptive PPF controller developed in this research. The experimental results showed that adaptive PPF controller is effective for active vibration control of the structure which is excited by either impact or harmonic disturbance. The filter frequency of the PPF controller is tuned in a very short period of time thus proving the efficiency of the adaptive PPF controller.

• 한국컴퓨터정보학회논문지
• /
• 제22권7호
• /
• pp.63-74
• /
• 2017

Recently in the area of ICT, the M2M and IoT are in the spotlight as a cutting edge technology with the help of advancement of internet. Among those fields, the smart home is the closest area to our daily lives. Smart home has the purpose to lead a user more convenient living in the house with WLAN (Wireless Local Area Network) or other short-range communication environments using automated appliances. With an arrival of the age of IoT, this can be described as one axis of a variety of applications as for the M2H (Machine to Home) field in M2M. In this paper, we propose a novel technique for estimating the location of a terminal that freely move within a specified area using the RSSI (Received Signal Strength Indication) in the WLAN environment. In order to perform the location estimation, the Fingerprint and KNN methods are utilized and the LMS with the gradient descent method and the proposed algorithm are also used through the error correction functions for locating the real-time position of a moving user who is keeping a smart terminal. From the estimated location, the nearest fixed devices which are general electric appliances were supposed to work appropriately for self-operating of virtual smart home. Through the experiments, connection and operation success rate, and the performance results are analyzed, presenting the verification results.

• 자동차안전학회지
• /
• 제13권4호
• /
• pp.129-143
• /
• 2021

This paper proposes actuator fault detection and adaptive fault-tolerant control algorithms using performance index and human-like learning for longitudinal autonomous vehicles. Conventional longitudinal controller for autonomous driving consists of supervisory, upper level and lower level controllers. In this paper, feedback control law and PID control algorithm have been used for upper level and lower level controllers, respectively. For actuator fault-tolerant control, adaptive rule has been designed using the gradient descent method with estimated coefficients. In order to adjust the control parameter used for determination of adaptation gain, human-like learning algorithm has been designed based on perceptron learning method using control errors and control parameter. It is designed that the learning algorithm determines current control parameter by saving it in memory and updating based on the cost function-based gradient descent method. Based on the updated control parameter, the longitudinal acceleration has been computed adaptively using feedback law for actuator fault-tolerant control. The finite window-based performance index has been designed for detection and evaluation of actuator performance degradation using control error.

• 전력전자학회:학술대회논문집
• /
• 전력전자학회 1998년도 Proceedings ICPE 98 1998 International Conference on Power Electronics
• /
• pp.293-297
• /
• 1998

This paper presents a newly developed speed sensorless drive using RLS based on Neural Network Training Algorithm. The proposed algorithm has just the time-varying learning rate, while the wellknown back-propagation algorithm based on gradient descent has a constant learning rate. The number of iterations required by the new algorithm to converge is less than that of the back-propagation algorithm. The theoretical analysis and experimental results to verify the effectiveness of the proposed control strategy are described.

• 한국정보과학회:학술대회논문집
• /
• 한국정보과학회 1998년도 가을 학술발표논문집 Vol.25 No.2 (2)
• /
• pp.360-362
• /
• 1998

In the original mixtures of experts framework, the parameters of the network are determined by gradient descent, which is naturally slow. In [2], the Expectation-Maximization(EM) algorithm is used instead, to obtain the network parameters, resulting in substantially reduced training times. This paper presents the new EM algorithm for prediction. We show that an Efficient training algorithm may be derived for the HME network. To verify the utility of the algorithm we look at specific examples in time series prediction. The application of the new EM algorithm to time series prediction has been quiet successful.