• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.16 no.3
• /
• pp.163-172
• /
• 2016

Electric load forecasting is essential for effective power system planning and operation. Complex and nonlinear relationships exist between the electric loads and their exogenous factors. In addition, time-series load data has non-stationary characteristics, such as trend, seasonality and anomalous day effects, making it difficult to predict the future loads. This paper proposes a locally-weighted polynomial neural network (LWPNN), which is a combination of a polynomial neural network (PNN) and locally-weighted regression (LWR) for daily shortterm peak load forecasting. Model over-fitting problems can be prevented effectively because PNN has an automatic structure identification mechanism for nonlinear system modeling. LWR applied to optimize the regression coefficients of LWPNN only uses the locally-weighted learning data points located in the neighborhood of the current query point instead of using all data points. LWPNN is very effective and suitable for predicting an electric load series with nonlinear and non-stationary characteristics. To confirm the effectiveness, the proposed LWPNN, standard PNN, support vector regression and artificial neural network are applied to a real world daily peak load dataset in Korea. The proposed LWPNN shows significantly good prediction accuracy compared to the other methods.

• The Journal of the Acoustical Society of Korea
• /
• v.39 no.4
• /
• pp.255-262
• /
• 2020

In modern passive sonar systems, a towed array sensor is used to minimize the effects of own ship noise and to get a higher SNR. The thin and long towed array sensor can be guided in a non-linear form according to the maneuvering of tow-ship. If this change of the array shape is not considered, the performance of beamformer may deteriorate. In order to properly beamform the elements in the array, an accurate estimate of the array shape is required. Various techniques exist for estimating the shape of the linear array. In the case of a method using a heading sensor, the estimation performance may be degraded due to the effect of heading sensor noise. As means of removing this potential error, weighted polynomial fitting technique for estimating array shape is developed here. In order to evaluate the performance of proposed method, we conducted computer simulation. From the experiments, it was confirmed that the proposed method is more robust to noise than the conventional method.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.22 no.1
• /
• pp.118-123
• /
• 2008

This paper proposes a frequency transfer function synthesis for approximating a high-order model with resonance to a low-order model in the frequency domain. The presented model approximation method is based on minimizing the error function weighted by the numerator polynomial of approximated models, which is used of the RLS(Recursive Least Square) technique to estimate the coefficient vector of approximated models. The proposed method provides better fitting in a low frequency and peak resonance. And an example is given to illustrate feasibilities of the suggested schemes.