Power Research

Open Access Open Access  Restricted Access Subscription Access
Open Access Open Access Open Access  Restricted Access Restricted Access Subscription Access

Short Term Load Forecasting using Soft Computing Techniques


Affiliations
1 Professor, Dept. of Electrical EngineeringFaculty of Engineering, D.E.I. Dayalbagh, Agra, U.P.-282 005, India
2 Faculty, Dept. of Electrical Engineering, D.E.I. Dayalbagh, Agra, U.P.-282 005, India
3 Energy Efficiency and Renewable Energy Division, Central Power Research Institute, Bangalore-560080, India
     

   Subscribe/Renew Journal


Soft computing techniques are extensively used for electrical load forecasting in the past such as ANN, Fuzzy Systems, GA etc.. ANN has some limitations, such unknown structure of ANN, Decision of neuron type, problem of training data and time, stuck in local minima etc. To overcome the drawbacks of ANN, a Generalized Neural Network (GNN) has been proposed. In this paper, different variants of GNN have been proposed to improve its performance such as GNN integrated with wavelet transform and trained with adaptive genetic algorithm and fuzzy system to forecast the short term week day electrical load. Performance of the proposed algorithm is compared with other GNN and its other variants on the basis of prediction error.

Keywords

Load Forecasting, ANN, Generalized neural network, Wavelet, Adaptive Genetic algorithms, Fuzzy systems.
User
Subscription Login to verify subscription
Notifications
Font Size

Abstract Views: 240

PDF Views: 0




  • Short Term Load Forecasting using Soft Computing Techniques

Abstract Views: 240  |  PDF Views: 0

Authors

D. K. Chaturvedi
Professor, Dept. of Electrical EngineeringFaculty of Engineering, D.E.I. Dayalbagh, Agra, U.P.-282 005, India
A. P. Sinha
Faculty, Dept. of Electrical Engineering, D.E.I. Dayalbagh, Agra, U.P.-282 005, India
Vikas Pratap Singh
Energy Efficiency and Renewable Energy Division, Central Power Research Institute, Bangalore-560080, India

Abstract


Soft computing techniques are extensively used for electrical load forecasting in the past such as ANN, Fuzzy Systems, GA etc.. ANN has some limitations, such unknown structure of ANN, Decision of neuron type, problem of training data and time, stuck in local minima etc. To overcome the drawbacks of ANN, a Generalized Neural Network (GNN) has been proposed. In this paper, different variants of GNN have been proposed to improve its performance such as GNN integrated with wavelet transform and trained with adaptive genetic algorithm and fuzzy system to forecast the short term week day electrical load. Performance of the proposed algorithm is compared with other GNN and its other variants on the basis of prediction error.

Keywords


Load Forecasting, ANN, Generalized neural network, Wavelet, Adaptive Genetic algorithms, Fuzzy systems.



DOI: https://doi.org/10.33686/prj.v9i4.189537