Kusumal Chalermyanont

A Computing Model of Artificial Intelligent Approaches to Mid-term Load Forecasting: a state-of-the-art- survey for the researcher

 Abstract—This article presents the review of the computing models applied for solving problems of midterm load forecasting. The load forecasting results can be used in electricity generation such as energy reservation and maintenance scheduling. Principle, strategy and results of short term, midterm, and long term load forecasting using statistic methods and artificial intelligence technology (AI) are summaried, Which, comparison between each method and the articles have difference feature input and strategy. The last, will get the idea or literature review conclusion to solve the problem of mid term load forecasting (MTLF).

High Frequency Transformer Designs for Improving Cross Regulation in Multiple-Output Flyback Converters

The cross regulation in multiple-output flyback converters is affected by leakage inductances of the transformers. The suitable winding arrangements of the transformers will lead to the improvement of cross regulation. This paper presents the comparative study of various winding arrangements in the three winding flyback transformers. The cross regulation model is considered in form of the resistance matrix related to the leakage inductance parameters. The design guidelines for the transformer for improving cross regulation are given based on the experimental results.

Mid Term Load Forecasting of the Country Using Statistical Methodology: Case Study in Thailand

Abstract-The paper describes the statistical methodology of multiple linear regression (MLR) and autoregressive integrated moving average (ARIMA) methods for mid term load forecasting of the country. The mid term load forecast has many applications such as maintenance scheduling, fuel reserve planning and unit commitment. However, the monthly peak load is a nonlinear, and non-stationary signal. Therefore, this paper proposed a statistical methodology to solve this problem which using multiple linear regression, and autoregressive integrated moving average, based on historical series of electric peak load, weather, and new economic variables such as consumer price index, and industrial index. This paper focuses on the forecasting of monthly peak load for 12 months ahead. This study focused on the mid term load forecasting of peak load demand for Thailand. Finally, we compared between MLR and ARIMA method that the results obtained the autoregressive integrated moving average method proves to be the best accuracy more than the multiple linear regression method.

Solar battery charger using a multi-stage converter

This paper demonstrates a multi-stage DC-DC converter to charge a battery with a three stage charging steps and can simultaneously perform the maximum power point tracking (MPPT). The system uses an auxiliary battery together with an bidirectional DC-DC converter to balance the power flow to complement the input solar power and the required power by the primary battery. The hardware prototypes shows the system functionality under all three charging stages.

Correlation Feature Selection Analysis for Fault Diagnosis of Induction Motors

This paper presents a feature selection method for stator winding fault analysis of induction motors by using a Correlation-based Feature Selection (CFS) method. The 14 original motor parameters are selected from the feature selection method with various searching approaches. The classification efficiency of optimal features obtained from the feature selection method is compared with results from the feature extraction method and the original features. In our experiment, we employ a 2.2 kW delta-connected motor which drives a dc generator as a load. The experimental results demonstrate that 4 common selected features for stator winding fault analysis of induction motors are a percent of load (%Load), a power factor (pf), a negative sequence voltage (V n ), and a negative sequence impedance (Z n ). The accuracy of the classification using this feature subset is higher than using all original features for three classification methods.

Peak load demand forecasting using two-level discrete wavelet decomposition and neural network algorithm

This paper proposed the discrete transform and neural network algorithms to obtain the monthly peak load demand in mid term load forecasting. The mother wavelet daubechies2 (db2) is employed to decomposed, high pass filter and low pass filter signals from the original signal before using feed forward back propagation neural network to determine the forecasting results. The historical data records in 1997-2007 of Electricity Generating Authority of Thailand (EGAT) is used as reference. In this study, historical information of peak load demand(MW), mean temperature(Tmean), consumer price index (CPI), and industrial index (economic:IDI) are used as feature inputs of the network. The experimental results show that the Mean Absolute Percentage Error (MAPE) is approximately 4.32%. This forecasting results can be used for fuel planning and unit commitment of the power system in the future.