Dissertation > Excellent graduate degree dissertation topics show

Power Load Forecast under Intelligent Grid Environment

Author: ZhangZuoZuo
Tutor: ZhangYouGang
School: Southwest Jiaotong University
Course: Electrical Engineering
Keywords: Stationary wavelet transform Bad data identification Wavelet clustering Elman Neural network WNN
CLC: TM715
Type: Master's thesis
Year: 2011
Downloads: 417
Quote: 1
Read: Download Dissertation

Abstract


Smart grid is the development direction of future grid to self-healing, security, power generation resources compatible, power user interaction, electric power market coordination and resource optimize efficiency and power quality quality, information system integration as the main characteristics of the realization of intelligent power grid, without accurate load forecasting technology support. The development of the electric power industry on the one hand it directly restricts the national economic and social development, and the correct power load forecasting that the development of the national economy can provide enough power for power system, also can help the development of itself, especially for the power system planning is concerned, precise load forecast is the whole planning the basis and premise for work. This paper in smart grid environment to load forecast of load signals at the same time, increase the reliability of the forecast results.Introduced the wavelet transform and the neural network, the basic principle of clustering theory and the basic concept of intelligent power grid. The discrete wavelet transform smooth wavelet transform, stationary wavelet transform the redundancy and panning invariability of the time-frequency transform, in the process, to avoid the sampling processing signal distortion. Using wavelet transformation noise reduction processing of load signals, is able to on signal processing while, not classify the trend. Change signal On load signal wavelet decomposition, used statistical methods after using probability theory, the method to weed out poor data. In the load forecast this step, use wavelet clustering of data load classification, then use Elman neural network algorithm forecast.Due to the complexity of the communication network, may introduce the interference of bad data, this paper deal with the problem of load forecast, the influence of bad data adopts the automatic identification method, i.e. used statistical methods to solve the mathematics paper, this is the first innovations.For intelligent power grid, AMI’s main function is to provide a intelligence platform, through this platform accurately grasp the characteristics of load node, the main method is to use wavelet clustering algorithms for load classification. Because the purpose of doing that for industrial enterprises and civil point speaking, its power load forecasting is different, the result of power load node to classify and separately predict, will greatly enhance the load forecasting results accuracy and dependability. This is the second innovation points. This method is far superior to expedite the load forecast intelligent master load forecast, and the whole of actual cannot consider different demands of electricity load.After a great deal of simulation results prove the validity and reliability of this algorithm, and the results show that, using WNN to load signals of multilayer signal decomposition, with periodic components to carry on the forecast, the result is accurate and reliable, and the load forecasting process based on intelligent power grid, the senior measure system provides massive data support, the use of wavelet clustering method to load, classification precision load forecasting technology to provide support. This method is AMI load forecasting methods and development direction of one of the feasibility.

Related Dissertations

  1. Study on the Buildings Deformation Around Expansive Soil Foundation Pit,TU443
  2. Data Mining for Forest Space Information Features from Remote Sensing Images,TP751
  3. Monitoring Leaf Nitrogen Status and Grain Protein Content in Wheat Using Near Infrared Spectroscopy,S512.1
  4. Intelligent Power Transformer Fault Diagnosis,TM41
  5. The Research of Color Recognition Based on Elman Neural Network and Clustering Algorithm,TP391.41
  6. High Fill Embankment Settlement Analysis and Forecasting and Control Standards,U416.1
  7. Genetic optimization wavelet neural network in the application of integrated navigation,V249.3
  8. Technology Research on AGV Traffic Sign Recognition Based on Image,TP391.41
  9. The Development on Intelligent Recognition Software of the Icing Thickness of Transmissionline,TM75
  10. Research on Methods of Fault Diagnosis Based on Wavelet Neural Network and Its Application in Variable Frequency Speed Regulation System,TP183
  11. A Study of Option Pricing Based on Neural Networks Method,F830.9
  12. The Research on Image Compression Based on BP Neural Network,TP391.41
  13. Parameter Optimization for Wavelet Neural Network and Its Application,TP183
  14. WNN neural network based short-term load forecasting,TM715
  15. Wavelet theory in Dam Deformation Monitoring Data Analysis Research,TV698.11
  16. Impulse signal detection method based on wavelet network,TP274.4
  17. Wavelet neural network stock prediction,TP183
  18. Research on Iris Recognition Methods Based on Wavelet Packet Neural Network,TP391.41
  19. Hospital -oriented design of mobile robot navigation system,TP242
  20. Forecast Research of Real Estate Price Index Based on WNN,F293.3
  21. Based on wavelet neural network modeling method of microwave nonlinear scattering function,TN62

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