Research On Active Island Detection Method For Photovoltaic Grid-Connected System

With the increasingly tense energy in the world and the increasing global environmental problems,the sustainable development and recycling of clean energy have become the inevitable trend of the current development.Solar energy is recognized as the most ideal new energy source,and photovoltaic power generation is one of the main ways to use solar energy.The grid connected photovoltaic power generation has brought many problems to the control and protection of the power grid and the photovoltaic system.Island detection as a necessary safety measure for photovoltaic grid-connected systems is one of the focuses of photovoltaic research.The active islanding detection algorithm for photovoltaic research direction,mainly from the active frequency drift detection method design,signal feature analysis,isolated grid connected photovoltaic power generation system structure and simulation model building and other aspects of the exploration and research.Firstly,this dissertation introduces the development and current situation of domestic and international photovoltaic power grid detection and island,expounds the purpose and significance of islanding detection,summarizes the classification and characteristics of islanding detection method.The main object of this dissertation is to establish the active frequency drift method.Secondly,this dissertation briefly introduces the principle and islanding detection.Secondly,this dissertation briefly introduces the principle and islanding detection,to evaluate the detection performance of the standard are analyzed and explained,according to the characteristics and control principle of photovoltaic grid system,using Simulink design to build a grid-connected model of single-phase photovoltaic power generation with a capacity of 5k W as an experimental platform for the study of islanding detection algorithm.Furthermore,by analyzing the principle of Active Frequency Drift with Positive Feedback(AFDPF)algorithm and detecting the influence of non-detection zones and parameters on the detection performance of AFDPF algorithm,this dissertation proposes a parameter-free adaptive AFDPF algorithm.Through the contrast simulation experiments under a variety of load conditions,the effectiveness of AFDPF algorithm is proved,and the characteristics of adaptive AFDPF algorithm without blind spots and low-power pollution are verified.Finally,this dissertation uses wavelet packet transform analysis characteristics of grid voltage signal in the island,the design of wavelet packet feature extraction as the introduction of parameter adaptive AFDPF detection algorithm to predict the condition detection algorithm is proposed based on wavelet packet and frequency drift combination,through different load conditions and non-island disturbances simulation experiment,verified the anti-interference performance and the superiority of the detection algorithm.