Load Forecasting Using Hybrid Models

Iee Proceedings-generation Transmission and Distribution, 1996

A hybrid neural network fuzzy expert system is developed to forecast short-term electric load accurately. The fuzzy membership values of the load and other weather variables are the inputs to the neural network, and the output comprises the membership values of the predicted load. An adaptive fuzzy correction scheme is used to forecast the final load by using a fuzzy rule base and fuzzy inference mechanism. Extensive studies have been performed for all seasons, and a few examples are presented in the paper, including average, peak and hourly load forecasts.

Due to centralized power system and continuous varying nature of load it is very difficult to balance demand and generation at all time. It is due to the fact that generation can't be controlled with the same pace as load due to restriction in instantaneous change in input to power plant. In this paper a work is presented to forecast week ahead loadusing ANN model with wavelet transform signal processing which has outperform all previous method due to the exceptional nature of leaning in neural network. The used datasets in this work are based on the past weather records. To demonstrate the effectiveness of the proposed approach data of load from 132 KV substation of Rajgarh (Dhar), M.P. has been taken to forecast the daily peak load for the Indore City of Madhya Pradesh.The data used in this work is from 1 st Jan 2017 to 31 st Dec 2017. All input variables have per day peak readings, so a total of 365 samples of each parameter are used in study.

2004 International Conference on Power System Technology, 2004. PowerCon 2004., 2004

Absmd-Long term load forecasting presents the first step in planning and developing hture generation, transmission and distribution facilities. Artificial Intelligent Applications have been introduced for load forecasting. Forecast@ procedure considered the correlation variables that have an influence over the demand for electricity, for example: Gross State Product (GSP), Consumers Price Index (CPU and Electricity Tariff (ET). These variables are chosen to enter the model as the inputs of network. T h e output is consumed energy. This paper proposed to apply the nniversd approximation properties of neural and wavelet networks to determine the function that denote relationship between input variables and output energy. The basic back-prapagation algorithm is used as a supervisory learning. Three network models are proposed in this paper: Functional Iink net, Multi-layer perceptron neural network and Wavelet network.

International Journal of Communications, Network and System Sciences, 2010

Electric load forecasting is essential for developing a power supply strategy to improve the reliability of the ac power line data network and provide optimal load scheduling for developing countries where the demand is increased with high growth rate. In this paper, a short-term load forecasting realized by a generalized neuron-wavelet method is proposed. The proposed method consists of wavelet transform and soft computing technique. The wavelet transform splits up load time series into coarse and detail components to be the features for soft computing techniques using Generalized Neurons Network (GNN). The soft computing techniques forecast each component separately. The modified GNN performs better than the traditional GNN. At the end all forecasted components is summed up to produce final forecasting load.

This paper focuses on forecasting electric load consumption using optimized Adaptive Neuro-Fuzzy inference System (ANFIS). It employs the use of Particle Swarm Optimization (PSO) to optimize ANFIS, with aim of improving its speed and accuracy. It determines the minimum error from the ANFIS error function and thus propagates it to the premise part. Wavelet transform was used to decompose the input variables using Daubechies 2 (db2). The purpose is to reduce outliers as small as possible in the forecasting data. The data was decomposed in to one approximation coefficients and three details coefficients. The combined Wavelet-PSO-ANFIS model was tested using weather and load data of Nova Scotia province. It was found that the model can perform more than Genetic Algorithm (GA) optimized ANFIS and traditional ANFIS, which is been optimized by Gradient Decent (GD). Mean Absolute Percentage Error (MAPE) was used to measure the accuracy of the model. The model gives lower MAPE than the other two models, and is faster in terms of speed of convergence.

Short term load forecasting is critically important in modern electricity networks since it helps provide supportive information for reliable power system operation in competitive electricity market environment. In this paper, the wavelet analysis based neural network model is employed to forecast the electricity demand in short-term period. The wavelet analysis helps to decompose the electricity demand data into different frequency bands. The Fourier transform is then employed to reveal the significant lags of these decomposed components. These lags are then used as inputs of neural network model to forecast the future values of each decomposed component. Finally, the forecasted components are combined together to form the electricity demand forecast. A case study has been reported in the paper by acquiring the data for the state of New South Wales, Australia. MAPE is used to validate the proposed model and the results show that the proposed method is promising for short term load forecasting.

IEEE Transactions on Power Systems, 1998

IEEE Transactions on Power Systems, 1995

This paper presents the development of a fuzzy system for short term load forecasting. The fuzzy system has the network structure and the training procedure of a neural network and is called Fuzzy Neural Network (FNN). A FNN initially creates a rule base from existing historical load data. The parameters of the rule base are then tuned through a training process, so that the output of the FNN adequately matches the available historical load data. Once trained, the FNN can be used to forecast future loads. Test results show that the FNN can forecast future loads with an accuracy comparable to that of neural networks, while its training is much faster than that of neural networks.

TJPRC, 2013

Short term load forecasting is a key issue in operating of electricity system. Many operating decisions are based on load forecast such as: dispatch scheduling of generating production, reliability and security analysis and maintenance plan for generators. Artificial neural networks (ANNs) have frequently been proposed for short-term load forecasting (STLF), because of their capabilities for nonlinear modelling of large multivariate datasets. This paper present a short-term load forecasting strategies based on ANN with multi-layer perceptron (MLP) structure and Levenberg–Marquardt (LM) learning algorithm. Also, impact of wavelet transform (WT) in prediction accuracy is studied. Proposed method is implemented in real case study of Zanjan power system.

1995

This paper presents the development of a hybrid neural network to model a fuzzy expert system for time series forecasting of electricc load. The hybrid neural network is trained to develop fuzzy logic rules andjind optimal inputloutput membership values of load and weather parameters. A hybrid learning algorithm consisting of unsupervised and supervised learning phases is used for training the fuzzifred neural network. In the supervised learning phase, both back-propagation and linear Kalman jilter algorithms are used for the adjustment of weights and membership functions. Extensive tests have been performed on a 2-year utility data for the generation of peak and average loadprojiles in 24 h. 48 h. and 168 h ahead time frame during summer and winter seasons. From the simulation results, it is observed that the fuzzy expert system using the Kalman $lter-based algorithm gives faster convergence and more accurate prediction of a load time series.