I. introductionLoad forecast is very important to the operation of electrical system for economic and security reasons, 131 if an accurate evaluation of the electrical system with a lead of at least 24 hours, then it is necessary to coordinate the economic power by selecting create a sufficient supply of Se loaded khopri much: curity check box.Analyzing power system using prdcted to load, to assess the status of the system in advance, so that future commitments may not be able to additionally load forecasting system is important as an analysis tool that is used to restore the power system. Author of several people and then are sent using artificial neural networks. (ANNs) load forecasting problem [1, 2, 4] However, the majority of jobs are in this area, considering the short term IEEE 0-7803-5272-4/96 $5.00 @ 1996 load forecasting (STLJ) with load prediction whlch is borrowed time to 24-hour weather variables (temperature meter unly) has been used in a large part of the program, click.These at the ICA (Research Center for use of C0mputatic), PUC-Rio Brazil intelligence nal), We've been using Backpropagation neural networks [5] in the process in several steps in order to predict the load time up to 744 hours forecast are made by feeding neural networks with load, and has hours of the day. We have introduced four different neural mesh loading prediction of last week's group, no date dfferent used temperature data, due to the characteristics of the consumers, that is supplied by the electric company in the study. Our objective is to show that the correct prebctions using tlvs, and to determine such limits, we add time to the leadership of the following section predctions analysis previ.ous series loading made to identify the neural network structure is the most appropriate. In section 3, we present a neural network structure cho'sen and section 4 presents the case studes, and results. A summary of this work are presented in section 5.
การแปล กรุณารอสักครู่..
