Influence Assessment and Location Research of Voltage Sags Based Intelligent Algorithm
|School||Zhejiang Normal University|
|Course||Computer Software and Theory|
|Keywords||Voltage dips Identification Locate Impact Assessment Intelligent algorithm|
Energy has become the basic resources of the national economy and social development, with the development of industrial and agricultural production, people put forward higher requirements for the quality of the power supply. Voltage sag is the most serious power quality problems occur frequently, affecting a wide range, seriously affecting the normal operation of the grid sensitive electrical equipment, causing huge economic losses and casualties voltage sag has become the main power quality complaints reason is the research focus of the electrical field at home and abroad. From voltage dips for the sensitive load operation characteristics of sag source identification, sag source positioning three carried out research in terms of voltage sag. The paper first introduces the definition of voltage dips and dips characteristic values ??of the voltage sag hazards, voltage sag and voltage dips mathematical models and simulation models have been studied. Voltage dips can seriously affect the operation of the sensitive equipment, In this paper, there is a wide range of applications in industry and agriculture DC speed control system, for example, analysis and evaluation of dips for sensitive load operation characteristics. Based on the transfer function of the control system, model DC speed control system, the voltage drop value, the duration of the system of DC speed control system operating characteristics (speed characteristics, torque characteristics, armature current). MATLAB Simulink and the use of open-loop, double loop DC motor control system voltage sag impact of simulation analysis. The results show that the voltage sag cause decreased to varying degrees of motor speed and output torque fluctuations affect the stable operation of the DC speed control system. Established by the neural network for dual closed-loop system, according to sag under the influence of system operating characteristics, dips impact assessment model. The test data show that the model can accurately describe the sag under the influence of system operating characteristics. Sag source identification for voltage sag prevention and treatment is important. Cause voltage dips main reasons are: short-circuit fault, induction motor starting, transformer switching. Different causes voltage dips have different characteristics, voltage sag source identification based on these features can be achieved. This paper presents a voltage sag source identification method based on S transform and PSO-SVM. S transform is an excellent time-frequency analysis, the process is simple, intuitive results, you can extract a variety of eigenvalues ??of the time domain and frequency domain. First, the S transform to extract the sag characteristic value, and then use the PSO algorithm for the sag source identification feature selection and SVM parameter optimization and select the voltage sag source identification, and finally by the SVM. We also adopt the GA-SVM for sag source identification, test results show that GA-SVM algorithm and PSO-SVM algorithm can achieve accurate identification of sag source, but relative to the GA-SVM, PSO-SVM in feature temporarily The drop source localization quickly ruled out the circuit failure and power supply dispute is of great significance. The essence of the voltage sag source localization is to determine the sag source relative to the location of the monitoring point, at home and abroad for sag source localization studies mainly focused on the use of dips process sensitive to changes in the electrical quantities positioning. This paper presents an idea of ??intelligent classification to determine the the sag source on the downstream positioning method, the basic idea is to learn from the voltage sag source positioning criterion as positioning characteristics, the use of intelligent classification method to establish sag source upstream and downstream Categories optimal classification surface. First, a brief analysis of the existing several dips source positioning method and its criterion, then extract location characteristic values ??on this basis to construct the optimal classification surface based on machine learning, using the two-class support vector machine to achieve a temporary Data reduction positioning. The test data show that this method can effectively sag source location, location with high accuracy and short computation time required.