Dissertation > Excellent graduate degree dissertation topics show

The Studies on Identification of Electric Load Model

Author: ZhangHongBin
Tutor: HeRenMu
School: North China Electric Power University (Beijing)
Course: Proceedings of the
Keywords: Load Model Overall measurement resolution method Load characteristics clustering Load characteristics Consolidated Kohonen neural network Induction motor Analytical sensitivity analysis T - S fuzzy model Adaptive neuro-fuzzy inference system , global load model
CLC: TM743
Type: PhD thesis
Year: 2003
Downloads: 1296
Quote: 60
Read: Download Dissertation

Abstract


Clustering about the dynamic characteristics of the power load in this thesis overall resolution measurement method of load modeling and synthesis load model parameters dispersible power system nonlinear variable structure load modeling and other key issues. Difficulties caused by the study of the time-varying power load, variable structural load modeling work to the overall resolution measurement method has important significance. The main content of the paper is as follows: the dispersion model parameters for load modeling of the overall measurement resolution method, we first conducted a more in-depth and meticulous research, the results showed that: whether the static load model, the differential equation load model, The model structure selection is incorrect, untrue, and load the presence of noise is the root cause of lead to the dispersion of the load model parameters. This paper presents an effective solution to the dispersion of load model parameters - load characteristics integrated. Dynamic Simulation Experiments and field data modeling examples demonstrate the effectiveness and feasibility of the method. Clustering and integrated implementation of the dynamic characteristics of power load towards practical one important indicator of load modeling, which is determined by the power load characteristics and the power load modeling principles. This paper proposes the use Kohonen neural network to solve the clustering problem of the dynamic characteristics of the power load, load dynamic characteristics clustering algorithm using Kohonen neural network. By cluster analysis of field data with integrated processing, can be found, despite the presence of load composition variability and randomness, but the load characteristics still show a certain regularity. The same time, the instance also verified the validity of a Kohonen neural network to solve the load dynamic characteristics clustering. For easy identification, and some parameters relative illegible induction motor load model parameter identification in practice there are some parameters, the first induction motor load model parameters analytical sensitivity analysis and identification of research results show that: Sensitivity easy identification of the parameters, and stability of the identification results and sensitivity parameters is relatively difficult to identify, and each parameter has enhanced gradually easily identifiable trend with excitation. According to these characteristics, this paper first proposed a comprehensive strategy for a new induction motor load model, and by measured data modeling to verify the effectiveness and feasibility of the policy. When the power load face a larger span of voltage and frequency disturbances, nonlinear variable structure characterized by fairly obvious, especially in the dynamic process of the power system and long-term. Structure for existing single load model can not accurately describe the non-linearity of the power load, variable structure characteristics, this paper first proposed the establishment of a global load model (TS fuzzy model) to describe the non-linearity of the power load, variable structure characteristics, and gives specific algorithm for global load model. Simulation data modeling examples demonstrate the effectiveness and feasibility of the method. The use and promotion of global load model can improve the accuracy of medium and long-term dynamic simulation of power systems. Proposed for the first time the nonlinear adaptive neuro-fuzzy inference system (ANFIS) to establish global load model to describe the power load, variable structure characteristics. It uses the back-propagation algorithm and the method of least squares hybrid algorithm lt; WP = 6 gt; adjust the antecedent parameters and consequent parameters, and automatically generate If-then rules, greatly improving the speed of convergence of the algorithm. Simulation data modeling instance to verify the correctness of the method.

Related Dissertations

  1. Study on DSP-Based Four-Quadrant Control System of Induction Motor,TM346
  2. Research on Vector Control System of Induction Motor Based on DSP,TM346
  3. Fault Diagnosis Research on Three-Tank System,TP277
  4. The Design of Experimental Device for Mine Motor Soft Starter,TM343.2
  5. The Research and Application of Modified Algorithms About Fuzzy Predictive Functional Control,TP273
  6. Study of Low Speed Control Strategy on Induction Motor Speed Sensorless,TM343
  7. Networked Control System Fault Diagnosis and Fault Tolerant Control,TP273
  8. Dual power switch switching Key Issues,TM762.1
  9. Application of Communication Strategies to Consecutive Interpreting of Sino-US Joint Press Conference,H315.9
  10. Study on Optimal Control for Chaotic System Based on T-S Fuzzy Model,O232
  11. Rare earth separation process of Life Cycle Assessment,TF845
  12. Speed Controller and Observer for Induction Motor Servo System,TM346
  13. Transient Stabilityanalysis and Calculation of Offshore Oil Platform,TM712
  14. The Study on Optimization Sootblowing System Based on Fuuzy Neural Net,TM621.2
  15. The Research on Induction Motor VVVF Speed Regulation in Electric Traction,TM343
  16. Control Strategy Optimization of Induction Motor for Fuel Cell City Bus,TM343
  17. The Optimization of Stator’s Structure of Large-Scale Induction Motor,TM343
  18. A Research on Long-span Bridge Vehicle Design Load,U441.2
  19. Study on the Fault Diagnosis of Asynchronous Induction Motor Base on the Wavelet Analysis,TH165.3
  20. Research on Characters and Parameters Optimization Method of LLC Resonant Converter,TM46
  21. Research on AC Variable Frequency Speed Regulation Experimental Platform Based on Direct Torque Control,TM921.51

CLC: > Industrial Technology > Electrotechnical > Transmission and distribution engineering, power network and power system > Power system simulation and calculation > Modeling and Simulation
© 2012 www.DissertationTopic.Net  Mobile