Electric load forecasting using an artificial neural network (original) (raw)
Related papers
European Transactions on Electrical Power, 2007
ABSTRACT The paper illustrates two different Artificial Neural Networks (ANN) architectures for electric Short-Term Load Forecasting (STLF). Two multi-layer perceptron ANN using the back-propagation learning algorithm have been implemented which provide different, although complementary, forecasting approaches (static and dynamic). In order to test the potentialities of the architectures implemented, the ANN have been applied to the Short-Term Forecasting of Italian hourly electric load. The importance of this load (peak demands up to about 38 000 MW) requires tools for STLF which must be as more accurate and precise as possible. This fact has imposed the adoption of some algorithmic enhancements to the basic back-propagation algorithm formulation. Since an adequate formulation of the influence exerted on hourly electric load by the main meteorological and climatic factors is not known at present, the data set used for ANN training phase has concerned only historical series of electric hourly demand. The paper illustrates the two ANN architectures as well as the computational platforms used for implementation. Finally, some results obtained from the application of the two ANN to the short-term forecasting of Italian electric load relevant to three different weeks of the year 1993 are comparatively reported.
Accurate Electricity Load Forecasting with Artificial Neural Networks
International Conference on Computational Intelligence for Modelling, Control and Automation and International Conference on Intelligent Agents, Web Technologies and Internet Commerce (CIMCA-IAWTIC'06), 2005
In this paper we present a simple yet accurate model to forecast electricity load with Artificial Neural Networks (ANNs). We analyze the problem domain and choose the most adequate set of attributes in our model. To obtain the best performance in prediction, we follow an experimental approach analyzing the entire ANN design space and applying different training strategies. We found that when little data is available, applying this approach is critical to obtain the best results. Our experiments also show that a simple ANN-based prediction model appropriately tuned can outperform other more complex models. Our feed-forward ANN-based model obtained 29% improvement in prediction accuracy when compared to the best results presented in the 2001 EUNITE competition.
An efficient approach for short term load forecasting using artificial neural networks
International Journal of …, 2006
In previous work, we applied artificial neural networks (ANN) for short term load forecasting using real load and weather data from the Hydro-Quebec databases where three types of variables were used as inputs to the neural network: (a) hour and day indicators, (b) weather related inputs and (c) historical loads. In general, for forecasting with a lead time of up to a few days ahead, load history (for the last few days) is not available, and therefore, estimated values of this load are used instead. However, a small error in these estimated values may grow up dramatically and lead to a serious problem in load forecasting since this error is fed back as an input to the forecasting procedure. In this paper, we demonstrate ANN capabilities in load forecasting without the use of load history as an input. In addition, only temperature (from weather variables) is used, in this application, where results show that other variables like sky condition (cloud cover) and wind velocity have no serious effect and may not be considered in the load forecasting procedure.
ARTIFICIAL NEURAL NETWORKS APPROACH BASED SHORT TERM ELECTRIC LOAD FORECASTING
IJCIRAS, 2020
The term load prediction refers to the projected load requirement using systematic process of defining load in sufficient quantitative detail so that vital power system expansion decisions can be made. In recent years, there is an emphasis on Short-Term Load Forecasting (STLF), the essential part of power system planning and operation. Rudimentary operating functions such as unit commitment, economic transmit, and unit preservation can be performed efficiently with a precise forecast. Short-term forecasting can assist in predicting the flow and making decisions that prevent overloading. This paper implements the STLF as a 24-hour forecast whose result is an hourly electric forecast. This paper uses the method of Artificial Neural Network (ANN) to create a STLF process. The inputs to the ANN are load profiles of one month previous days and the weather variables of that days. Correlation analysis between load and weather variables will be used for all predictor input data to the ANN to optimize in size and accuracy. MATLAB programming language is used to implement this system.
Short term electric load forecast using artificial neural networks
Proceedings of MELECON '94. Mediterranean Electrotechnical Conference, 1994
Artificial Neural Network (ANN) Method is applied to forecast the short-term load for a large power system. The load has two distinct patterns: weekday and weekend-day patterns. The weekend-day pattern include Saturday, Sunday, and Monday loads. A nonlinear load model is proposed and several structures of ANN for short-term load forecasting are tested. Inputs to the ANN are past loads and the output of the ANN is the load forecast for a given day. The network with one or two hidden layers are tested with various combination of neurons, and results are compared in terms of forecasting error. The neural network, when grouped into different load patterns, gives good load forecast.
Short-Term Electric Load Forecasting Using Neural Networks
Computer as a Tool, …
Short term electric load forecasting with a neural network based on fuzzy rules is presented. In this network, fuzzy membership functions are represented using combinations of two sigmoid functions. A new scheme for augmenting the rule base is proposed. The network employs outdoor temperature forecast as one of the input quantities. The influence of imprecision in this quantity is investigated. The model is shown to be capable of also making reasonable forecasts in exceptional weekdays. Forecasting simulations were made with three different time series of electric load. In addition, the neurofuzzy method was tested at two electricity works, where it was used to produce forecasts with 1-24 hour lead times. The results of these one month real world tests are represented. Comparative forecasts were also made with the conventional Holt-Winters exponential smoothing method. The main result of the study is that the neuro-fuzzy method requires stationarity fi~om the time series with respect to training data in order to give clearly better forecasts than the Holt-Winters method.
Neural network based approach for short-term load forecasting
2009 IEEE/PES Power Systems Conference and Exposition, 2009
Short-term load forecast is an essential part of electric power system planning and operation. Forecasted values of system load affect the decisions made for unit commitment and security assessment, which have a direct impact on operational costs and system security. Conventional regression methods are used by most power companies for load forecasting. However, due to the nonlinear relationship between load and factors affecting it, conventional methods are not sufficient enough to provide accurate load forecast or to consider the seasonal variations of load. Conventional ANN-based load forecasting methods deal with 24-hour-ahead load forecasting by using forecasted temperature, which can lead to high forecasting errors in case of rapid temperature changes. This paper presents a new neural network based approach for short-term load forecasting that uses the most correlated weather data for training, validating and testing the neural network. Correlation analysis of weather data determines the input parameters of the neural networks. The suitability of the proposed approach is illustrated through an application to the actual load data of the Egyptian Unified System.
Energy and Buildings, 2013
This paper presents the upgrading of a method for predicting short-term power building consumption that was previously developed by the authors (EUs method). The upgrade uses a time temperature curve (TTC) forecast model. The EUs method involves the use of artificial neural networks (ANN) for predicting each independent process-end-uses (EUs). End-uses consume energy with a specific behaviour in function of certain external variables. The EUs method obtains the total consumption by the addition of the forecasted end-uses. The inputs required for this method are the parameters that may affect consumption, such as temperature, type of day, etc. Historical data of the total consumption and the consumption of each end-use are also required. A model for prediction of the time temperature curve has been developed for the new forecast method (TEUs method). The temperature at each moment of the day is obtained using the prediction of the maximum and minimum daytime temperature. This provides various benefits when selecting the training days and in the training and forecasting phases, thus improving the relationship between expected consumption and temperatures. The method has been tested and validated with the consumption forecast of the Universitat Politècnica de València for an entire year.
Short Term Load Forecasting Using Artificial Neural Networks
Artificial neural network (ANN) has been used for many years in sectors and disciplines like medical science, defence industry, robotics, electronics, economy, forecasts, etc. The learning property of ANN in solving nonlinear and complex problems called for its application to forecasting problems. This report present the development of an ANN based short-term load forecasting model for the 132/33KV sub- Station, Kano, Nigeria. The recorded daily load profile with a lead time of 1-24 hours for the year 2005 was obtained from the utility company. The Levenberg-Marquardt optimization technique which has one of the best learning rates was used as a back propagation algorithm for the Multilayer Feed Forward ANN model using MATLAB® R2008b ANN Toolbox. Experiences gained during the development of the model regarding the selection of the input variables, the ANN structure, and the training parameters are described. The forecasted next day 24 hourly peak loads are obtained based on the stationary output of the ANN with a performance Mean Squared Error (MSE) of . and compares favorably with the actual Power utility data. The results have shown that the proposed technique is robust in forecasting future load demands for the daily operational planning of power system distribution sub-stations in Nigeria.
Short Term Load Forecasting in Electric Power Systems with Artificial Neural Networks
The demand in electric power should be predicted with the highest possible accuracy as it affects decisively many of power system's operations. Conventional methods for load forecasting were built on several assumptions, while they had to cope with relations between the data used that could not be described analytically. Artificial Neural Networks (ANNs) gave answers to many of the above problems and they became the predominant load forecasting technique. In this chapter the reader is first introduced to Artificial Neural Networks and their usage in forecasting the load demand of electric power systems. Several of the major training techniques are described with their pros and cons being discussed. Finally, feed-forward ANNs are used for the short-term forecasting of the Greek Power System load demand. Various ANNs with different inputs, outputs, numbers of hidden neurons etc. are examined, techniques for their optimization are proposed and the obtained results are discussed.