• 대한전자공학회논문지TE
• /
• 제39권4호
• /
• pp.363-369
• /
• 2002

본 논문은 적응성 뉴로-퍼지 인터페이스 시스템(Adaptive Neuro-Fuzzy Inference System : ANFIS)과 웨이브렛 변환 다중해상도 분해(multi-resolution Analysis : MRA)을 기반으로 한 웨이브렛 신경망을 가지고 임의의 비선형 함수 학습 근사화를 개선하는 것이다. ANFIS 구조는 벨형 퍼지 소속 함수로 구성이 되었으며, 웨이브렛 신경망은 전파 알고리즘과 역전파 신경망 알고리즘으로 구성되었다. 이 웨이브렛 구성은 단일 크기이고, ANFIS 기반 웨이브렛 신경망의 학습을 위해 역전파 알고리즘을 사용하였다. 1차원과 2차원 함수에서 웨이브렛 전달 파라미터 학습과 ANFIS의 벨형 소속 함수를 이용한 ANFIS 모델 기반 웨이브렛 신경망의 웨이브렛 기저 수 감소와 수렴 속도 성능이 기존의 알고리즘 보다 개선되었음을 확인하였다.

• Environmental Engineering Research
• /
• 제25권4호
• /
• pp.545-553
• /
• 2020

Rainfall-Runoff modeling plays a crucial role in various aspects of water resource management. It helps significantly in resolving the issues related to flood control, protection of agricultural lands, etc. Various Machine learning and statistical-based algorithms have been used for this purpose. These techniques resulted in outcomes with an acceptable rate of success. One of the pertinent machine learning algorithms namely Adaptive Neuro Fuzzy Inference System (ANFIS) has been reported to be a very effective tool for the purpose. However, the computational complexity of ANFIS is a major hindrance in its application. In this paper, we resolved this problem of ANFIS by incorporating one of the evolutionary algorithms known as Particle Swarm Optimization (PSO) which was used in estimating the parameters pertaining to ANFIS. The results of the modified ANFIS were found to be satisfactory. The performance of this modified ANFIS is then compared with conventional ANFIS and another popular statistical modeling technique namely ARIMA model with respect to the forecasting of runoff. In the present investigation, it was found that proposed PSO-ANFIS performed better than ARIMA and conventional ANFIS with respect to the prediction accuracy of runoff.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2004년도 학술대회 논문집 정보 및 제어부문
• /
• pp.523-525
• /
• 2004

Over the last few years, the control of bipedal robot has been considered a promising research field in the community of robotics. But the problems we encounter make the control of a bipedal robot a hard task. The complicated link connection of the bipedal robot makes it impossible to achieve its exact model. In addition, the joint velocity is needed to accomplish good control performance. In this paper a control method using ANFIS as an system approximator is purposed. First a model biped robot of a biped robot with switching leg influence is presented. Unlike classical method, ANFIS approximation error estimator is inserted in the system for tuning the ANFIS. In the entire system, only ANFIS is used to approximate the uncertain system. ANFIS tuning rule is given combining the observation error, control error and ANFIS approximation error. But this needs velocity information which is not available. So a practical method is newly presented. Finally, computer simulation results is presented to show this control method has good position tracking performance and robustness without need for leg switching acknowledgement.

• 대한토목학회논문집
• /
• 제31권5B호
• /
• pp.405-414
• /
• 2011

본 연구에서는 유역에서 관측되는 강우량과 유출량의 시계열 자료를 바탕으로 최근 시계열 예측 및 시스템 제어 분야에서 성공적으로 적용되고 있는 적응형 네트워크 기반 퍼지추론 시스템(ANFIS)을 갑천 유역에 적용하여 시유출량을 모델링하였다. 입력구조, 소속함수 종류와 개수 등을 다양하게 변화시켜 ANFIS 모형을 학습하고, 평균제곱근오차(RMSE), 평균첨두유량오차(PE) 및 평균첨두시간오차(TE)를 이용하여 ANFIS의 유출해석에 대한 적용성을 평가하였다. 현재시간의 시유출량 Q(t)에 대한 ANFIS의 적용성은 우수한 것으로 평가되었으며, ANFIS 모형은 관측유출량을 적절히 모의하였다. 입력구조가 다른 입력모형을 구성하여 최대 8시간까지 ANFIS의 유출예측 적용성을 평가하였다. 예측시간 증가에 따라서 ANFIS의 유출예측 정확도는 감소하여 예측시간 4시간 이상의 시유출량에 대한 ANFIS의 유출예측 적용성은 제한적이었다. ANFIS는 입력과 출력 자료들만 이용하므로 물리기반 모형에 비교하여 모형구축이 비교적 손쉽기 때문에 홍수 유출모델링에 ANFIS을 유용하게 적용할 수 있을 것으로 판단된다.

• Computers and Concrete
• /
• 제32권1호
• /
• pp.27-44
• /
• 2023

The current study targets to apply the adaptive neuro-fuzzy inference system (ANFIS) for the estimation of the shear resistance offered by the externally bonded fiber-reinforced polymer (EB-FRP) U-jackets. A total of 202 groups of data cumulated from previous investigations, were employed for the development and evaluation of the ANFIS model. A relative appraisal between the ANFIS predictions and the results of experiments has shown that the assessments by current ANFIS model are in good concurrence with the latter. In addition, assessment of the accuracy of the ANFIS model was done by relating the ANFIS predictions with the forecasts of eight extensively used design guidelines. Based on the examination of various performance measures, it has been derived that the adequacy of the ANFIS model is better than the available guidelines. A parametric investigation has additionally been done to reconnoiter the influence of individual parameters as well as their combined effects on the shear contribution of EB-FRP. Based on the observations made from the parametric study, it has been witnessed that the ANFIS model has incorporated the effect of different parameters more competently than the considered design guidelines.

• 한국건축시공학회지
• /
• 제23권1호
• /
• pp.103-111
• /
• 2023

건설 프로젝트의 초기단계에서 공사비를 정확하게 예측하는 것은 프로젝트를 성공적으로 수행하기 위해 매우 중요하다. 본 연구에서는 ANFIS 모델을 활용하여 건설프로젝트의 초기단계에 건축공사비를 예측할 수 있는 모델을 제시하였다. 모델의 활용도를 높이기 위해 공개된 공사비 데이터를 활용하였으며 프로젝트 초기단계의 제한된 정보를 바탕으로 예측할 수 있는 모델을 제시하고자 하였다. ANFIS와 관련된 기존 연구를 분석하여 최근의 동향을 파악하였으며 ANFIS의 기본 구조를 고찰한 후 건축공사비 예측을 위한 ANFIS 모델을 제시하였다. ANFIS의 모델의 소속함수의 종류와 개수에 따라 달라지는 예측 성능을 분석하여 가장 성능이 우수한 모델을 제시하였으며, 대표적인 기계학습 모델의 예측 정확도와 비교분석하였다. 적용결과 ANFIS 모델을 다른 기계학습 모델과 비교한 결과 동등 이상으로 성능을 나타내 프로젝트 초기단계 공사비 예측에 적용 가능할 것으로 판단된다.

• Computers and Concrete
• /
• 제26권5호
• /
• pp.397-409
• /
• 2020

The current study targets to estimate the contribution of the end-anchored FRP composites in resisting shear force using a soft computing tool i.e., adaptive neuro-fuzzy inference system (ANFIS). A total of 107 sets of data accumulated from literature was utilized for the development and evaluation of the current ANFIS model. A comparative analysis between the ANFIS predictions and the acquired experimental results has shown that the ANFIS predictions are in very good agreement with that of experimental ones. Additionally, the accuracy of the current ANFIS model has been weighed up against the estimates of nine widely adopted design guidelines. Based on various statistical parameters, it has been deduced that the effectiveness of the current ANFIS model is better than the considered design guidelines. Besides this, a parametric study was carried out to explore the combined effect of different parameters as well as the impact of individual parameters.

• Structural Engineering and Mechanics
• /
• 제45권6호
• /
• pp.779-802
• /
• 2013

In this paper, adaptive neuro-fuzzy inference system (ANFIS) and artificial neural networks (ANNs) techniques are developed and applied to identify damage in a model steel girder bridge using dynamic parameters. The required data in the form of natural frequencies are obtained from experimental modal analysis. A comparative study is made using the ANNs and ANFIS techniques and results showed that both ANFIS and ANN present good predictions. However the proposed ANFIS architecture using hybrid learning algorithm was found to perform better than the multilayer feedforward ANN which learns using the backpropagation algorithm. This paper also highlights the concept of ANNs and ANFIS followed by the detail presentation of the experimental modal analysis for natural frequencies extraction.

• Structural Engineering and Mechanics
• /
• 제47권3호
• /
• pp.401-419
• /
• 2013

In this study, the efficiency of adaptive neuro-fuzzy inference system (ANFIS) and genetic expression programming (GEP) in predicting the effects of infill walls on base reactions and roof drift of reinforced concrete frames were investigated. Current standards generally consider weight and fundamental period of structures in predicting base reactions and roof drift of structures by neglecting numbers of floors, bays, shear walls and infilled bays. Number of stories, number of bays in x and y directions, ratio of shear wall areas to the floor area, ratio of bays with infilled walls to total number bays and existence of open story were selected as parameters in GEP and ANFIS modeling. GEP and ANFIS have been widely used as alternative approaches to model complex systems. The effects of these parameters on base reactions and roof drift of RC frames were studied using 3D finite element method on 216 building models. Results obtained from 3D FEM models were used to in training and testing ANFIS and GEP models. In ANFIS and GEP models, number of floors, number of bays, ratio of shear walls and ratio of infilled bays were selected as input parameters, and base reactions and roof drifts were selected as output parameters. Results showed that the ANFIS and GEP models are capable of accurately predicting the base reactions and roof drifts of RC frames used in the training and testing phase of the study. The GEP model results better prediction compared to ANFIS model.

• 지능정보연구
• /
• 제15권3호
• /
• pp.151-165
• /
• 2009

퍼지신경망 모형은 인공신경망의 네트워크 구조 표현방법 및 학습알고리듬과 퍼지시스템의 추론방법을 통합한 모형으로 제어 및 예측분야에 성공적으로 적용되고 있다. 본 연구에서는 퍼지신경망 모형 중 우수한 예측정확도로 인해 최근 각광받고 있는ANFIS (Adaptive Network-based Fuzzy Inference System)모형을 기반으로 하는 분류모형을 설계하고 기존의 분류기법(C5.0 의사결정나무)과 비교하여 분류 정확성 관점에서 평가한다. ANFIS 추론의 경우, 최종 결과값이 계급값이 아닌 연속형 변수값을 취하게 되므로 산출된 결과값을 이용하여 적절한 계급값을 할당하는 과정이 필요하다. 본 연구에서는 의사결정나무기법을 이용하여 계급값을 할당하는 방식과 군집분석을 이용하여 계급값을 할당하는 두 가지 방식을 제안하고 두 가지 데이터 세트에 적용하여 ANFIS를 기반으로 한 분류모형의 정확도를 평가하였다.