Research And Design Of Improved Dynamic Voltage Restorer

In order to solve the power quality problem existing in the power grid,the power electronic equipment of the type of power compensation has been more and more widely researched and applied.In order to solve the voltage sag problem in power quality problems,Dynamic Voltage Restorer(DVR)has been rapidly developed as the most ideal compensation device.But,the randomness of the voltage sag depth and the transient nature of the occurrence time are very high on the design of the DVR detection and control method.In the DVR system structure,the unit of detection detects the signal of the grid voltage and the load voltage in real time;gets the command signal in the shortest time by comparing the characteristic quantity of the voltage and then generates the compensation voltage to the inverter through the output control signal of the control unit to ensure that the voltage at the load terminal is stable.Therefore,a more suitable detection algorithms and control strategies can achieve the best compensation effect of the DVR.In order to develop the compensation effect of DVR better,the thesis takes detection algorithm and control strategy as the main research obj ect,and proposes improved strategies.The main research contents are as follows:Firstly,the thesis studies the basic structure of DVR;analyzes the functions of each component.Simultaneously,it establishes the equivalent circuit model through the analysis of working principle and derives its system equation.The thesis briefly analyzes three compensation strategies which are most commonly used,and choose the minimum energy compensation method that requires less energy storage capacity.Secondly,the thesis studies the detection algorithm of DVR.Since the traditional detection algorithm does not fully consider the fluctuation of voltage dip and harmonic distortion,the detection time gets longer.Thus,the thesis proposes a detection algorithm,which fully considers the fluctuation of voltage sag,harmonic distortion and other factors,based on improving transform.It can ensure the detection of the characteristic value of voltage sag in the shortest time,so that the detection time of the DVR system gets shorter.The simulation results show that the proposed algorithm can quickly and accurately detect the characteristic value of voltage dip in the absence of harmonic components and harmonic components.Then,in order to optimize the response time of traditional fuzzy PID control,this thesis proposes a PID control strategy which is supposed to improve the neural network based on fuzzy PID controller and fuzzy neural network of Mamdani fuzzy system.This strategy is that making the parameters of the neural network get trained by the improved hybrid optimization algorithm,so they can be optimized and adjusted.By simulation experiments,the simulation results show that this strategy can bring better control effects and meet the control needs of DVR system.Finally,in order to verify the effectiveness and superiority of the detection algorithm and control strategy,the thesis designs the topology structure of the improved DV R system;builds a simulation model in the Simulink environment and analyzes the two components of the non-harmonic component and the 10th,5th and 7th harmonic components.The simulation result shows the designed DVR can meet the voltage sag compensation and harmonic suppression effect.