Font Size: a A A

Research On Electromagnetic Characteristics Analysis And Active Disturbance Rejection Decoupling Control Of Six-Pole Radial-Axial Active Magnetic Bearing

Posted on:2021-02-02Degree:MasterType:Thesis
Country:ChinaCandidate:S S WangFull Text:PDF
GTID:2392330629987205Subject:Electrical engineering
Abstract/Summary:PDF Full Text Request
Magnetic bearing is a non-contact bearing technology,which aims to eliminate the disadvantages of rolling bearings and ordinary bearings.Magnetic bearing is also a bearing technology that supports the rotor with magnetic force without any physical contact between the rotor and the static supporting structure.Compared with traditional bearings,magnetic bearings have many advantages,such as no contact,no lubrication,no wear,high speed and reduced vibration.In military industry,chemical industry,medical tools,compressors and petroleum industries have a very wide application prospects.Due to the low price and mature technology of the inverter,the three-pole magnetic bearings driven by the inverter have become the focus of research.In order to further reduce the cost and improve the bearing capacity,this dissertation proposes a six-pole radial-axial active magnetic bearing(AMB)driven by the inverter,and research the technical problems around its mathematical modeling,parameter design,suspension force characteristic analysis,decoupling control and experiment.The main :Firstly,the research status of magnetic bearing at home and abroad is reviewed,then the working principle of magnetic bearings,the influence of different magnetic poles on magnetic bearings,the characteristics and classification of magnetic bearings,the application of magnetic bearings are introduced,and the current mainstream magnetic bearing control technology is analyzed.Secondly,working principle of the six-pole radial-axial AMB are analyzed,and its mathematical model and suspension force expression are derived.According to the mathematical model and design objective of the six-pole radial-axial AMB,the basic parameters are designed.The six-pole radial-axial AMB is a symmetrical structure,which increases the bearing capacity,reduces the cost and improves the suspension force current characteristics.To verify the performance of the six-pole radial-axial AMB and the correctness of the designed parameters,the six-pole radial-axial AMB and the three-pole radial-axial AMB are compared and analyzed by finite element simulation.The analysis is mainly carried out from three aspects: maximum bearing capacity,relationship between radial suspension force and current,and electromagnetic coupling between radial two degrees of freedom.Thirdly,although the coupling between the six-pole radial-axial AMB radial two degrees of freedom is weaker than that between the three-pole radial-axial AMB,there is still a strong coupling between the radial two degrees of freedom when the rotor offset balance position is far away.Furthermore,the disturbances sustained by the magnetic bearings are often not fixed in practical applications.In this dissertation,an active disturbance rejection controller(ADRC)based on BP neural network is designed.Based on the characteristics of the BP neural network wireless approximation nonlinear function,the parameters of the extended state observer in the ADRC are adjusted according to the magnitude of the disturbance to which the magnetic axis is subjected,thus improving the tracking accuracy of the extended state observer for the disturbance.Compared with conventional ADRC,the simulation results show the superiority of ADRC based on BP neural network.Finally,through the above analysis,the six-pole radial-axial AMB has superior performance.In this dissertation,the digital control system of six-pole radial-axial AMB is introduced from hardware and software,and the experimental platform is built.Nonlinear coupling,lifting and disturbance experiments are carried out on the experimental platform to verify the superiority of six-pole radial-axial AMB.
Keywords/Search Tags:Six-pole active magnetic bearing, analysis of electromagnetic characteristics, decoupling control, active disturbance rejection control, BP neural networks
PDF Full Text Request
Related items