Font Size: a A A

Research On Fault Tolerant Control Strategy Of Three-Level Inverter

Posted on:2023-03-14Degree:MasterType:Thesis
Country:ChinaCandidate:Y J QinFull Text:PDF
GTID:2532307145965179Subject:Electrical engineering
Abstract/Summary:PDF Full Text Request
Three level inverter is an important equipment in the field of new energy power generation and other industrial production fields.The three-level inverter used in actual industrial production is in a bad geographical environment,and has to bear high electrical stress and high thermal stress for a long time,which is very prone to failure and affect industrial production.Therefore,the research on fault diagnosis technology and fault-tolerant control strategy for inverter is of great significance.This paper focuses on neutral point clamped(NPC)three-level inverter.The research contents are as follows:Firstly,the topology and working principle of NPC three-level inverter are analyzed,as well as its space vector pulse width modulation(SVPWM)strategy.The open circuit fault mechanism of power switch is studied and classified,and the simulation model is built by Matlab/Simulink.The three-phase output current signal is transformed into two-dimensional signal I_αand I_βby 3/2 transformation,then I_αand I_βthe signal is decomposed by wavelet packet decomposition algorithm,the energy values of each frequency band of the third layer are obtained and normalized as the fault eigenvector.Secondly,a BP neural network model suitable for open circuit fault diagnosis of NPC three-level inverter is designed,and the neural network is trained by fault eigenvector.Because BP neural network has some disadvantages,such as slow convergence speed and easy to fall into local minimum,this paper proposes to use the improved grey wolf optimizer(IMGWO)to determine its initial weight and threshold,because the global and local search capabilities of the traditional grey wolf optimizer(GWO)are inconsistent,it is optimized by cosine of the convergence factor and introducing variable proportional weight.Then use the adaptive moment estimation algorithm(ADAM)to optimize the learning rate of BP neural network,it can avoid BP neural network falling into local minimum.Finally,the fault-tolerant control strategy of inverter after open circuit fault is studied,and a three-phase four leg fault-tolerant control strategy is proposed.By paralleling a two-level bridge arm in the inverter,when the device has open circuit fault,the bridge arm of fault phase is separated by bidirectional thyristor,and the load of fault phase is connected to the midpoint of two-level bridge arm,the NPC three-level inverter can still operate in full in case of open circuit fault.In order to effectively suppress the fluctuation of capacitor neutral point voltage,a"vector replacement SVPWM algorithm"is proposed to readjust the switching order and action time of SVPWM algorithm.The simulation results show that the proposed fault-tolerant control strategy can still maintain the neutral point voltage balance after the inverter fault,and the output side voltage amplitude will not be affected,and the reliability is high.
Keywords/Search Tags:Three Level Inverter, Fault Diagnosis, Improved Gray Wolf Algorithm, BP Neural Network, Adaptive Moment Estimation Algorithm, Fault Tolerant Control
PDF Full Text Request
Related items