Daily Peak Load Research Articles

A New Training Method for Analyzable Structured Neural Network and Application of Daily Peak Load Forecasting

This paper presents a daily peak load forecasting method using an analyzable structured neural network (ASNN) in order to explain forecasting reasons. In this paper, we propose a new training method for ASNN in order to explain forecasting reason more properly than the conventional training method. ASNN consists of two types of hidden units. One type of hidden units has connecting weights between the hidden units and only one group of related input units. Another one has connecting weights between the hidden units and all input units. The former type of hidden units allows to explain forecasting reasons. The latter type of hidden units ensures the forecasting performance. The proposed training method make the former type of hidden units train only independent relations between the input factors and output, and make the latter type of hidden units train only complicated interactions between input factors. The effectiveness of the proposed neural network is shown using actual daily peak load. ASNN trained by the proposed method can explain forecasting reasons more properly than ASNN trained by the conventional method. Moreover, the proposed neural network can forecast daily peak load more accurately than conventional neural network trained by the back propagation algorithm.

Daily Peak Load Forecasting by Structured Representation on Genetic Algorithms for Non-linear Function Fitting

Daily Peak Load Forecasting by Structured Representation on Genetic Algorithms for Non-linear Function Fitting

Bayesian Modeling and Forecasting of Intraday Electricity Load

The advent of wholesale electricity markets has brought renewed focus on intraday electricity load forecasting. This article proposes a multi-equation regression model with a diagonal first-order stationary vector autoregresson (VAR) for modeling and forecasting intraday electricity load. The correlation structure of the disturbances to the VAR and the appropriate subset of regressors are explored using Bayesian model selection methodology. The full spectrum of finite-sample inference is obtained using a Bayesian Markov chain Monte Carlo sampling scheme. This includes the predictive distribution of load and the distribution of the time and level of daily peak load, something that is difficult to obtain with other methods of inference. The method is applied to several multiequation models of half-hourly total system load in New South Wales, Australia. A detailed model based on 3 years of data reveals trend, seasonal, bivariate temperature/humidity, and serial correlation components that all vary intraday, justifying the assumption of a multiequation approach. Short-term forecasts from simple models highlight the gains that can be made if accurate temperature predictions are exploited. Bayesian predictive means for half-hourly load compare favorably with point forecasts obtained using iterated generalized least squares estimation of the same models.

Electric load forecasting for northern Vietnam, using an artificial neural network

AbstractThis paper employs a feed‐forward neural network with a back‐propagation algorithm for the short‐term electric load forecasting of daily peak (valley) loads and hourly loads in the northern areas of Vietnam. A large set of data on peak loads, valley loads, hourly loads and temperatures was used to train and calibrate the artificial neural network (ANN). The calibrated network was used for load forecasting. The mean percentage errors for the peak load, valley load, one‐hour‐ahead hourly load and 24‐hour‐ahead hourly load were −1.47%, −3.29%, −2.64% and −4.39%, respectively. These results compare well with similar studies.

KW‐value of a photovoltaic power generation system evaluated from actual data on electricity demand and insolation

AbstractThis paper proposes a new method of evaluating the kW‐value of photovoltaic power generation systems (PV systems). This evaluation is based on actual insolation data, collected by the Japan Meteorological Agency at four locations in the Chubu area in Japan, and the actual electrical load data of the Chubu area in the 1990s. We selected specific days on which the daily peak load was relatively high. For the selected days, we estimated the average output of PV systems using the IEA's method for considering the temperature dependency of cell efficiency. Then, we calculated the reduction of the peak load ΔLp(d). The resultant ΔLp(d) was distributed in the range from 0.4 to 0.5 kW/kWp.We evaluated the average value of the smallest ΔLp(d) as the kW‐value of PV systems. When the kW‐value was evaluated for the two or three days with the smallest ΔLp(d) among the selected 10 days for 1 year, the kW‐value of the PV systems was calculated to be about 30%. This was close to the value obtained by the conventional loss of load probability (LOLP) method. Our study evaluates the relationship between the kW‐value of PV systems and the number of days on which the insolation was low but the daily peak demand was relatively high. © 2003 Wiley Periodicals, Inc. Electr Eng Jpn, 144(1): 42–49, 2003; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/eej.10089

Artificial neural network-based peak load forecasting using conjugate gradient methods

The daily electrical peak load forecasting (PLF) has been done using the feed forward neural network (FFNN)-based upon the conjugate gradient (CG) back-propagation methods, by incorporating the effect of 11 weather parameters, the previous day peak load information, and the type of day. To avoid the trapping of the network into a state of local minima, the optimization of user-defined parameters, namely, learning rate and error goal, has been performed. The training dataset has been selected using a growing window concept and is reduced as per the nature of the day and the season for which the forecast is made. For redundancy removal in the input variables, reduction of the number of input variables has been done by the principal component analysis (PCA) method of factor extraction. The resultant dataset is used for the training of a 3-layered NN. To increase the learning speed, the weights and biases are initialized according to the Nguyen and Widrow method. To avoid over fitting, an early stopping of training is done at the minimum validation error.

Artificial Neural Network-Based Peak Load Forecasting Using Conjugate Gradient Methods

Daily electrical peak load forecasting has been done using the feed forward neural network based upon the conjugate gradient back propagation methods by incorporating the effect of eleven weather parameters, the previous day's peak load information, and the type of day. To avoid the trapping of the network into a state of local minima, the optimization of user-defined parameters viz., leaming rate and error goal has been performed. The training data-set has been selected using a growing window concept and is reduced per the nature of the day and the season for which the forecast is made. For redundancy removal in the input variables, reduction of the number of input variables has been done using the principal component analysis method of factor extraction. The resultant data set is used for the training of a three-layered neural network To increase the leaming speed, the weights and biases are initialized according to the Nguyen and Widrow method. To avoid overfitting, an early training is stopped early at the minimum validation error.

Short-term hourly load forecasting using time-series modeling with peak load estimation capability

This paper presents a new time series modeling for short term load forecasting, which can model the valuable experiences of the expert operators. This approach can accurately forecast the hourly loads of weekdays, as well as, of weekends and public holidays. It is shown that the proposed method can provide more accurate results than the conventional techniques, such as artificial neural networks or Box-Jenkins models. In addition to hourly loads, daily peak load is an important problem for dispatching centers of a power network. Most of the common load forecasting approaches do not consider this problem. It is shown that the proposed method can exactly forecast the daily peak load of a power system. Obtained results from extensive testing on the Iran's power system network confirm the validity of the developed approach.

Enhanced removal of ammonium by combined nitrification/adsorption in expanded clay aggregate filters

Long-term experiments were carried out on a nitrifying biofilter with a zeolite-containing expanded clay aggregate filter media (Filtralite ZL) receiving daily peak loads of ammonia. The filter removed ammonium from domestic wastewater by a combination of nitrification and sorption of ammonium. The filter was biologically regenerated continuously and there was no deterioration in performance after 10 months of continuous experiments. Isotherm experiments confirmed that the Filtralite ZL granules that had been used for more than 8 months had about the same capacity as the unused material. Desorption experiments that were carried out indicated that desorption of ammonium from the filter material was mainly caused by an exchange of cations from the wastewater even at low cation concentrations.

構造的GAによるGMDHを用いた翌日最大電力需要予測

Recently, the independent power producers (IPPs) and the distributed power generations (DGs) are increase on by the electric power system with the power system deregulation. And the power system becomes more complicated. It is necessary to carry out the electric power demand forecasting in order to the power system is operated for the high economical and the high-efficient. For the improvement of electric power demand forecasting, many methods, such as the methods using fuzzy theory, neural network and SDP data, are proposed. In this paper, we proposed the method using STROGANOFF (STructured Re-presentation on Genetic Algorithms for Non-linear Function Fitting) that approximate the value of predictive to the future data by the past data is obtained. Also, the weather condition was considered for the forecasting that is improvement, and the daily peak load forecasting in next day on Chubu district in Japan was carried out, and the effectiveness of proposed method was examined.

Peak load forecasting using multiple-year data with trend data processing techniques

This paper presents a regression-based daily peak load forecasting method using multiple-year data with trend cancellation and trend estimation techniques. Daily peak load heavily depends on daytime temperature and is influenced by the other weather factors such as humidity. Since the characteristic of the load is varying, peak loads just before a forecasting day are more significant for the forecasting. The regression model can represent relationships between these weather factors and peak loads. However, the forecasting model is sometimes not adequate for precise load forecasting. The regression model is well matched with the late data, but the model causes large forecasting errors in transitional seasons because of seasonal change of load characteristics. In order to forecast precisely through a year, a method of using seasonal or whole year data from past years is proposed. In this paper, two kinds of trend data processing techniques are described. The first is trend cancellation. The second is trend estimation. The trend cancellation technique removes annual load growth by means of division or subtraction processes with morning load on the forecasting day. The trend estimation technique estimates the trend between the forecasting year's load and the past year's load by using the variable transformation techniques. The performance of both techniques, verified with simulations on actual load data, is also described. © 1998 Scripta Technica, Electr Eng Jpn, 124(1): 7–16, 1998

Economic impact analysis of load forecasting

This paper examines the economic impact of inaccurate load forecasts. Monte Carlo simulations are used to study the effect of different load forecasting accuracy. Investigations into the effect of improving the daily peak load forecasts, effect of different seasons of the year and effect of utilization factors are also presented.

Effect of probabilistic inputs on neural network-based electric load forecasting

This paper presents a novel method to include the uncertainties or the weather-related input variables in neural network-based electric load forecasting models. The new method consists of traditionally trained neural networks and a set of equations to calculate the mean value and confidence intervals of the forecasted load. This method was tested for daily peak load forecasts for one year by using modified data from a large power system. The tests indicate that in addition to the confidence interval, the new method provides a more accurate mean forecast than a multilayer perceptron networks alone.

Fuzzy logic for short term load forecasting

The results of an investigation of a fuzzy logic model for short term load forecasting are presented. The proposed methodology uses fuzzy rules to incorporate historical weather and load data. These fuzzy rules are obtained from the historical data using a learning-type algorithm. Test results from daily peak and total load forecasts for one year of data from a large scale power system indicate that the fuzzy rule bases can produce results similar in accuracy to more complicated statistical and back propagation neural network methods.

Forecasting daily peak load by a deterministic prediction method employing Gram‐Schmidt orthonormalization

AbstractA technique for forecasting daily peak load in a utility power system is presented. After embedding time series data of daily peak load into a reconstructed state space, a nonlinear mapping is constructed by a local approximation method based on the orthonormal Gram‐Schmidt bases. This method utilizes only the past load data for short‐term prediction of the daily peak load, while many conventional methods make predictions with various kinds of data such as temperature and weather. The quality of prediction by the proposed method is as good as those with other prediction methods. Moreover, the results of short‐term prediction by this method are satisfactory even with data as small as 250 points.

Development of peak load forecasting system using neural networks and fuzzy theory

AbstractThis paper presents a peak load forecasting system using multilayer neural networks and fuzzy theory. Electric load forecasting in power systems is a very important task from the perspective of reliability and economic operation. Daily peak load forecasting is one of the basic operations of generation scheduling for the following day. Therefore, many statistical methods have been developed and used for such forecasting even though it has been difficult to construct a proper functional model.The developed system is applied by neural network and fuzzy theory to forecast for daily, weekly and monthly peak load. The system consists of an engineering workstation (EWS) and a personal computer (PC). The EWS is for learning and data‐bases, and the PC is for man‐machine interface such as forecasting operation.The system has been used since June 1993. The result evaluated with an absolute mean error is 1.63 percent for 10 months. From the results shown here, the system applied by neural network and fuzzy theory has high validity.

Daily peak electric load forecasting using an artificial neural network and an improvement method for reducing the forecasting errors

AbstractThis paper proposes a forecasting method for shortterm peak electric loads using a 3‐layer neural network of locally active units. Each unit in the hidden layer of the neural network is activated only by input vectors in a bounded domain of vector space. This characteristic enables additional learning. Furthermore, it is supposed to provide the network structure with information that helps to improve forecasting accuracy. The neural network is applied to daily peak load forecasting simulations in summer. The results show that the proposed method is superior to a conventional neural network with the backpropagation algorithm. To make the best use of the neural network, an error‐oriented method of parameter modification is also examined.

Application of radial basis function neural network model for short-term load forecasting

A description and original application of a type of neural network, called the radial basis function network (RBFN), to the short-term system load forecasting (SLF) problem is presented. The predictive capability of the RBFN models and their ability to produce accurate measures that can be used to estimate confidence intervals for the short-term forecasts are illustrated, and an indication of the reliability of the calculations is given. Performance results are given for daily peak and total load forecasts for one year using data from a large-scale power system. A comparison between results from the RBFN model and the back-propagation neural network are also presented.

グラムシュミット直交化法を用いた決定論的予測手法に基づく翌日最大電力予測

グラムシュミット直交化法を用いた決定論的予測手法に基づく翌日最大電力予測

ニューロ・ファジー応用最大電力予測システムの開発

This paper presents a peak load forecasting system using multilayer neural networks and fuzzy theory. Electric load forecasting in power systems is a very important task from the perspective of reliability and economic operation. Daily peak load forecasting is one of the basic operations of generation scheduling for the following day. Therefore, many statistical methods have been developed and used for such forecasting even though it has been difficult to construct a proper functional model. The developed system is applied by neural network and fuzzy theory to forecast for daily, weekly and monthly peak load. The system consists of an engineering workstation (EWS) and a personal computer (PC). The EWS is for learning and data-bases, and the PC is for man-machine interface such as forecasting operation. The system has been used since June 1993. The result evaluated with an absolute mean error is 1.63 percent for 10 months. From the results shown here, the system applied by neural network and fuzzy theory has high validity.