Font Size: a A A

Research On Speed-sensorless Vector Control Of Induction Motor Based On Full-order Flux Observer

Posted on:2020-11-23Degree:MasterType:Thesis
Country:ChinaCandidate:M L SongFull Text:PDF
GTID:2392330596977263Subject:Electrical engineering
Abstract/Summary:PDF Full Text Request
With the continuous development of AC speed regulation technology,the dynamic performance and steady-state accuracy of the AC speed regulation system have been greatly improved.On the basis,people put forward higher requirements for system reliability.As a relatively precise mechanical component in the system,the speed sensor becomes the weakest link.So the speed-sensorless vector control technology has received extensive research and attention.In this paper,the state-space models of induction motor and full-order flux observer are established firstly.The vector control technology and the model reference adaptive speed identification based on the full-order flux observer are introduced.The traditional speed adaptive law is derived using the Lyapunov's second method and Popov's hyperstability criterion separately.The assumptions for ensuring the stability of the speed identification system are also given.Based on the traditional speed adaptive law,two kinds of feedback matrix design method are introduced in detail.And evaluating the performance of three feedback matrixes from four aspects.They include the smoothness of switching between the current-model-based observer and the voltage-model-based observer,the stability of the identification speed,the convergence speed of the full-order flux observer and the parameter sensitivity of flux observation.Aiming at the instability of the speed identification system based on the traditional speed adaptive law in low speed regenerating mode,the causes for the instability are obtained through theoretical analysis and unstable range is deduced.According to the causes of the instability,a new speed adaptive law suitable for regenerating mode is proposed.Besides,the parameter constraints that guarantee the global stability of the speed identification system are given.The parameter sensitivity analysis is carried out on the speed identification system based on the improved speed adaptation law.At the same time,the theory that the rotor speed is unobservable when the stator current frequency is reduced to zero is confirmed.In the digital realization of the control system,the discretization of full-order flux observer is indispensable.This paper analyzes the advantages and disadvantages of the Euler method and bilinear method from three aspects which are calculated amount,stability range and discretization error.According to the characteristics of every method,a new discretization strategy is proposed,which is applying differentmethod according to the synchronization frequency.In order to obtain a high-performance vector control system,this paper proposes a PI parameters tuning method for direct rotor flux orientation vector control system regulators based on motor parameters,sampling frequency,desired bandwidth and stability margin.This method has wide versatility.Compared with the traditional SVPWM digital implementation method,this paper introduces an equivalent implementation method of injecting zero-sequence component into sinusoidal modulated waves.The effectiveness of the method is verified by the simulation.Finally,based on the designed feedback matrix,speed adaptive law and PI parameters,a simulation model of speed-sensorless vector control system based on full-order observer is built on Matlab/Simulink platform,which verifies the correctness of previous theoretical analysis and the feasibility of the control method.At same time,lots of experiments are carried out on the hardware platform with TMS320F2812 as the main controller and EP1K30QC208 as the co-controller.Experimental results verify that the speed-sensorless vector control system has good static and dynamic performance.
Keywords/Search Tags:Induction motor, Vector Control, Full-order flux observer, Feedback matrix, Speed adaptive law, Parameter sensitivity
PDF Full Text Request
Related items