Search

This paper presents a regression based daily peak load forecasting method with a transformation technique. In order to forecast the load precisely through a year, we should consider seasonal load change, annual load growth and the latest daily load change. To deal with these characteristics in the load forecasting, a transformation technique is presented. This technique consists of a transformation function with translation and reflection methods. The transformation function is estimated with the previous year's data points, in order that the function converts the data points into a set of new data points with preserving the shape of temperature-load relationships in the previous year. Then, tire function is slightly translated so that the transformed data points will fit the shape of temperature-load relationships in the year. Finally, multivariate regression analysis with the latest, daily loads and weather observations estimates the forecasting model. Large forecasting errors caused by the weather-load nonlinear characteristic in the transitional seasons such as spring and fall are reduced. Performance of the technique which is verified with simulations on actual load data of Tokyo Electric Power Company is also described. Keywords - Load Forecasting, Multivariate Regression

Describes the development of partial discharges generated in a gas insulated substation (GIS) using a neural network. Collection of neural network (NN) input data; Evaluation of the similarity of injected pulses to actual abnormalities; Practical application of neural network; Checking of NN on an actual GIS; Conclusions.