• 설비공학논문집
• /
• 제17권8호
• /
• pp.780-788
• /
• 2005

Two different types of algorithms were developed and applied to detect the partial faults of a multi-type air conditioning system. Partial faults include the compressor valve leakage, the refrigerant pipe partial blockage, the condenser fouling, and the evaporator fouling. The first algorithm was developed by using mathematical models and parity relations, and the second algorithm was developed by using mathematical models and a RBF neural network. Test results showed that the second algorithm was better than the first algorithm in detecting various partial faults of the system. Therefore, the algorithm developed by using mathematical models and a RBF neural network may be used for the detection of partial faults of an air-conditioning system.

• 한국통신학회논문지
• /
• 제25권6A호
• /
• pp.885-892
• /
• 2000

MC-COMA 시스템에서 RBF 네트워크를 이용한 다중사용자 수신기를 제안하고자 한다. 제안된 수신기에서는 주파수 다양성을 효과적으로 이용할 수 있도록 RBF 네트워크가 주파수축에 연결된 구조를 갖는다. 본 논문에서는 주파수 선택적 다중 경로 페이딩 채널에 대한 모의 실험을 수행하였다. 이러한 모의 실험을 통해, 제안된 수신기가 near-far 효과에 대한 대응성과 사용자 용량증가를 통한 성능향상을 얻고자 하는데 사용될 수 있음을 확인되었다. 그리고 제안된 수신기가 도플러 주파수는 10Hz이고 경로수가 5일 경우 $10^{-3}$의 BER에서 병렬 간섭제거 기법을 이용한 다중사용자 수신기보다 약 1.8배의 사용자를 수용할 수 있음을 확인하였다.

• Asian Journal of Atmospheric Environment
• /
• 제9권1호
• /
• pp.12-21
• /
• 2015

Air pollution is a challenging issue in some of the large cities in developing countries. In this regard, data interpretation is one of the most important parts of air quality management. Several methods exist to analyze air quality; among these, we applied the Multilayer Perceptron (MLP) and Radial Basis Function (RBF) methods to predict the hourly air concentration of benzene in 14 districts in the municipality of Tehran. Input data were hourly temperature, wind speed and relative humidity. Both methods determined reliable results. However, the RBF neural network performance was much closer to observed benzene data than the MLP neural network. The correlation determination resulted in 0.868 for MLP and 0.907 for RBF, while the Index of Agreement (IA) was 0.889 for MLP and 0.937 for RBF. The sensitivity analysis related to the MLP neural network indicated that the temperature had the greatest effect on prediction of benzene in comparison with the wind speed and humidity in the study area. The temperature was the most significant factor in benzene production because benzene is a volatile liquid.

• 한국정밀공학회지
• /
• 제34권2호
• /
• pp.125-131
• /
• 2017

2-Axis Pan and Tilt Motion Platform, a complex multivariate non-linear system, may incur any disturbance, thus requiring system controller with robustness against various disturbances. In this study, we designed an adaptive backstepping compensated controller by estimating the disturbance and error using the Radial Basis Function Neural Network (RBF NN). In this process, Uniformly Ultimately Bounded (UUB) was demonstrated via Lyapunov and stability was confirmed. By generating progressive disturbance to the irregular frequency and amplitude changes, it was verified for various environmental disturbances. In addition, by setting the RBF NN input vector to the minimum, the estimated disturbance compensation process was analyzed. Only two input vectors facilitated compensatory function of RBF NN via estimating the modeling and control error values as well as irregular disturbance; the application of the process resulted in improved backstepping controller performance that was confirmed through simulation.

• 전자공학회논문지S
• /
• 제36S권8호
• /
• pp.39-47
• /
• 1999

본 논문에서는 포항제철의 열연 권취 온도 공정을 제어하기 위한 방법으로 온라인 재학습 가능한 RBF 네트워크(Radial Basis Function network)를 제안한다. 새로 제안된 신경회로망 모델을 이용해 권취 온도 제어 시스템의 열전달 계수를 생성해주는 기조의 규칙 기반 계수표를 대체할 수 있다. 열연 공정 작업의 시간에 따른 변화를 고려하도록 지존의 RBF 네트워크에 부가적인 온라인 재학습용 시냅스 가중치를 도입한다. 부가적인 가중치들로 인해 이미 학습된 전체 데이터들의 특성 정보를 유지하면서 새로 들어오는 데이터들에 대해 온라인 재학습이 이루어진다. 따라서, 제안된 RBF 네트워크는 파국적 간섭(catastrophic interference)효과를 상당히 감소시킬 수 있다. 그리고 거부 네트워크을 도입하여 제어기의 신뢰도을 높일 수 있었고, 실제 현장에 적용한 실험 결과는 기존의 방법보다 평균 2.2퍼센트이상 향상된 성능을 보였다.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2005년도 ICCAS
• /
• pp.925-930
• /
• 2005

In this paper, evolutionary multi-objective selection method of RBF networks structure is considered. The candidates of RBF network structure are encoded into the chromosomes in GAs. Then, they evolve toward Pareto-optimal front defined by several objective functions concerning with model accuracy and model complexity. An ensemble network constructed by such Pareto-optimal models is also considered in this paper. Some numerical simulation results indicate that the ensemble network is much robust for the case of existence of outliers or lack of data, than one selected in the sense of information criteria.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제2권3호
• /
• pp.174-178
• /
• 2002

This paper discusses a blind equalization technique for FIR channel system, that might be minimum phase or not, in digital communication. The proposed techniques consist of two parts. One is to estimate the original channel coefficients based on fourth-order cumulants of the channel output, the other is to employ RBF neural network to model an inverse system fur the original channel. Here, the estimated channel is used as a reference system to train the RBF. The proposed RBF equalizer provides fast and easy teaming, due to the structural efficiency and excellent recognition-capability of R3F neural network. Throughout the simulation studies, it was found that the proposed blind RBF equalizer performed favorably better than the blind MLP equalizer, while requiring the relatively smaller computation steps in tranining.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1998년도 하계학술대회 논문집 G
• /
• pp.2299-2301
• /
• 1998

In this paper, a modified RBF (Radial Basis Function) neural network structure is suggested for the prediction of time series with non-linear, non-stationary characteristics. Conventional RBF neural network predicting time series by using past outputs is for sensing the trajectory of the time series and for reacting when there exists strong relation between input and hidden neuron's RBF center. But this response is highly sensitive to level and trend of time serieses. In order to overcome such dependencies, hidden neurons are modified to react to the increments of input variable and multiplied by increments(or decrements) of out puts for prediction. When the suggested structure is applied to prediction of Lorenz equation, and Rossler equation, improved performances are obtainable.

• Advances in environmental research
• /
• 제5권3호
• /
• pp.153-167
• /
• 2016

We applied multilayer perceptron (MLP) and radial basis function (RBF) neural network in upstream and downstream water quality stations of the Karaj Reservoir in Iran. For both neural networks, inputs were pH, turbidity, temperature, chlorophyll-a, biochemical oxygen demand (BOD) and nitrate, and the output was dissolved oxygen (DO). We used an MLP neural network with two hidden layers, for upstream station 15 and 33 neurons in the first and second layers respectively, and for the downstream station, 16 and 21 neurons in the first and second hidden layer were used which had minimum amount of errors. For learning process 6-fold cross validation were applied to avoid over fitting. The best results acquired from RBF model, in which the mean bias error (MBE) and root mean squared error (RMSE) were 0.063 and 0.10 for the upstream station. The MBE and RSME were 0.0126 and 0.099 for the downstream station. The coefficient of determination ($R^2$) between the observed data and the predicted data for upstream and downstream stations in the MLP was 0.801 and 0.904, respectively, and in the RBF network were 0.962 and 0.97, respectively. The MLP neural network had acceptable results; however, the results of RBF network were more accurate. A sensitivity analysis for the MLP neural network indicated that temperature was the first parameter, pH the second and nitrate was the last factor affecting the prediction of DO concentrations. The results proved the workability and accuracy of the RBF model in the prediction of the DO.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제13권1호
• /
• pp.39-49
• /
• 2013

This paper presents a method of improving the performance of a day-ahead 24-h load curve and peak load forecasting. The next-day load curve is forecasted using radial basis function (RBF) neural network models built using the best design parameters. To improve the forecasting accuracy, the load curve forecasted using the RBF network models is corrected by the weighted sum of both the error of the current prediction and the change in the errors between the current and the previous prediction. The optimal weights (called "gains" in the error correction) are identified by differential evolution. The peak load forecasted by the RBF network models is also corrected by combining the load curve outputs of the RBF models by linear addition with 24 coefficients. The optimal coefficients for reducing both the forecasting mean absolute percent error (MAPE) and the sum of errors are also identified using differential evolution. The proposed models are trained and tested using four years of hourly load data obtained from the Korea Power Exchange. Simulation results reveal satisfactory forecasts: 1.230% MAPE for daily peak load and 1.128% MAPE for daily load curve.