Dissertation > Excellent graduate degree dissertation topics show

Study on Short-term Load Forecasting Based on Weighted LS-SVM

Author: BaiBo
Tutor: WangLinChuan
School: Tohoku Electric Power University
Course: Proceedings of the
Keywords: Short-term load LS-SVM Robustness Bayesian evidence theory Prediction accuracy
CLC: TM715
Type: Master's thesis
Year: 2011
Downloads: 110
Quote: 0
Read: Download Dissertation

Abstract


The short-term load forecasting is a complex task , because it has the apparent randomness and unpredictability of the resulting prediction accuracy is difficult to have increased significantly . Traditional forecasting methods is often considered too simple to make a major impact on short-term load factors did not play its due role . In recent years, with the development of artificial intelligence techniques , machine learning languages ??has become the focus of the study load forecasting staff . Scholars try a variety of intelligent language used for short-term load forecasting , least squares support vector machine (LS-SVM) is one of the highlights . With the depth of its research , LS-SVM is increasingly favored by the experts . This article is based on a comprehensive understanding of the short - term load forecasting , choose the method of least squares support vector machine its prediction research . Taking into account the modern short-term load forecasting will involve a variety of external factors, such as weather , In this paper, the model takes into account the specific circumstances of weather types , maximum and minimum temperature factors . Due to a variety of factors to consider , so the increase in the complexity of the model , and the model established by the LS-SVM defects - robustness loss , so the estimated coefficient value plus a right to consider the robustness of the model error construct a weighted least squares support vector machine (WLS-SVM) model . In addition , model optimization and radial basis function (RBF) as the kernel function of the model parameters using Bayesian evidence theory . Measured load data acquisition to analysis, found a special phenomenon in the load - rest day delay , its adaptive prediction model for the phenomenon of text . Taking into account the load forecasting \Last WLS-SVM method studied in this paper under the Bayesian evidence of actual load forecast. As can be seen by the Matlab programming simulation method used has good predictive model created to meet the short - term load forecasting accuracy requirements .

Related Dissertations

  1. Identification and Robustness Analysis of Nonlinear Hybrid Dynamical System Concerning Glycerol Transport Mechanism,TQ223.162
  2. The Research of Structural System and Arrangement with Good Behavior of Anti-collapse under Large Earthquake,TU352.11
  3. Research on Daily Load-forecasting for Power Grid Based on ANN,TM715
  4. An Improved Method for Short-term Load Forecasting Based on FFT and Chaos,TM715
  5. Large-capacity digital fingerprint system embedding algorithm,TP309.7
  6. ORACLE database system reliability study,TP311.138
  7. Research on High-precision Measurement and Modeling of High Voltage Electric Transformers,TM452
  8. Robust Fault Estimation and Active Fault Tolerant Control for Uncertain System,TP13
  9. Consider short-term load forecasting PV,TM715
  10. Based on environmental parameters of the transition environment predict human thermal sensation,TU831
  11. Based on the H.264 standard video watermarking technology in the network learning platform in the Application and Implementation,TP309.7
  12. Digital watermarking based on neural network algorithm and implementation,TP309.7
  13. Based on sparse representation Face Image Recognition Method,TP391.41
  14. Based on wavelet and neural network theory of power system load forecasting,TM715
  15. A Study of Robust Control Method for Pneumatic Brake ABS on Commercial Vehicles,U463.526
  16. Methodology Study on Prediction and Determinants of Total Health Expenditures in China,R197.1
  17. Robust Controller Design for Gene Regulatory Networks under Intrinsic Parameters Fluctuation and Extrinsic Noises,Q75
  18. Norm Continuity of Solution Semigroup and Robust Controllability for Delay Systems with Unbounded Delay Operators,O231
  19. Application Study on Short-term Load Forecasting Based on Support Vector Regression,TM715
  20. Short - term load forecasting , energy management system based on support vector machines,TM715
  21. Research on Combination Model for Short-term Load Forecasting Based on EMD and GP Regression Theory,TM715

CLC: > Industrial Technology > Electrotechnical > Transmission and distribution engineering, power network and power system > Theory and Analysis > Power system planning
© 2012 www.DissertationTopic.Net  Mobile