Research On Photovoltaic Grid-connected Inverter Detection Technology

With the increasing demand for energy,problems of energy utilization are becoming more and more prominent.Solar energy technology is the most widely used energy support in the utilization of renewable energy.Nowadays,the power market is changing with each passing day.The country has proposed that green development,energy conservation and environmental protection should be the main direction of energy and power development in the future.Solar power generation has gradually become a popular power production technology.Among them,the development of photovoltaic power generation technology has been growing,and it can supply continuous power to the load and the power grid simultaneously when it is paralleled with the grid system.At the same time,photovoltaic grid-connected networks also face many technical challenges,such as the stability of power quality and the control methods of grid-connected systems.The island effect affects the operation and safety of the power grid system.As an important technical difficulty,the corresponding island detection method is a great attention to vast scholars.In order to reduce the harm of the island effect,the research of islanding detection technology has far-reaching significance.Firstly,this paper expounds the background and research status quo of the generation of the islanding effect,and the blind zone judgment of islanding detection,and the description of he control of grid-connected inverter.Secondly,this paper studies separately active island detection technology and passive island detection technology.Active island detection technology is based on traditional reactive current disturbance islanding detection technology.This paper proposes an improved islanding detection method for frequency positive feedback reactive current disturbance,by periodically applying positive feedback function of reactive disturbance current and disturbance functions.to achieve the anti-islanding effect protection function and increas amounts of reactive perturbation momenturm,and making the frequency out of limit,islanding detection technology.For the research of passive island detection technology,this paper adopts the optimization algorithms for island detection without adding any disturbances,and the extraction of feature vectors and the selection of classification algorithms.The feature extraction method based on wavelet packet transform and the feature extraction method based on variable mode decomposition(VMD)are adopted,combining respectively with extreme learning machine(ELM)and genetic membrane.Which constitutes an algorithm-based passive islanding detection technology,The islanding detection technology is simulated and compared by MATLAB simulation platform,and optimized for genetic membrane based on wavelet packet transform.The BP neural network islanding detection algorithm is verified and analyzed.By extracting the common point voltage signals under different working conditions,the identification and judgment of the islands are carried out.The simulation shows that the intelligent passive island detection method has the characteristics of no blind zone,no disturbance,and accurate judgment.