SHORT-TERM LOAD FORECASTING USING HYBRID FUZZY-NEURAL NETWORK IN "PLN PJB REGION IV" SYSTEM
Total View This Week0
Total View This Week0
Institusion
Institut Teknologi Nasional Malang
Author
Saputra, Andy
Subject
Electrical Engineering
Datestamp
2022-05-25 02:08:26
Abstract :
One of the function of planing and operation of an Electric Power System is short-term load forecasting (57LF) that is load forecast from several hour until several day Forecast accuracy has an economic impact on Electric Power Companies. Therefore there is need for an accurate so that there is congruent betwen generation and power demand In this paper we analyse Short-term loud forecasting using Hyhrid Fuzzy Neural Network (FNN which takes into account the influence of Temperature, Humidity and Wind Speed to improve forecast accuracy. Fuzzy Inference System (FIS) structure whanism is based on 30 Lapers in the architecture and the implemented timbership function as the tip Linear Representation. That Artificial Neural Network ANN raining uses the perceptron network architecture with mans ligger cand the Backpropagation metode for the training algorithm where we apply the Feedforward and Backward stages to change the trummed weight to obtain allero out Input to the fiary inference system mechanism and the training process of neural network stilizes the trend of the historical data Hybrid F Neural Network (FVN) can forcast with the per hours and 0.75% error per days. The effectiveness of short-term forecasting is demonstrated by its application PLN PUB REGION I rem
One of the function of planing and operation of an Electric Power System is short-term load forecasting (57LF) that is load forecast from several hour until several day Forecast accuracy has an economic impact on Electric Power Companies. Therefore there is need for an accurate so that there is congruent betwen generation and power demand In this paper we analyse Short-term loud forecasting using Hyhrid Fuzzy Neural Network (FNN which takes into account the influence of Temperature, Humidity and Wind Speed to improve forecast accuracy. Fuzzy Inference System (FIS) structure whanism is based on 30 Lapers in the architecture and the implemented timbership function as the tip Linear Representation. That Artificial Neural Network ANN raining uses the perceptron network architecture with mans ligger cand the Backpropagation metode for the training algorithm where we apply the Feedforward and Backward stages to change the trummed weight to obtain allero out Input to the fiary inference system mechanism and the training process of neural network stilizes the trend of the historical data Hybrid F Neural Network (FVN) can forcast with the per hours and 0.75% error per days. The effectiveness of short-term forecasting is demonstrated by its application PLN PUB REGION I rem
Download
book
BibTex
Latex, Jabref
cloud_download
Original Resource
url resource Institution
