Research On Finite Set Model Predictive Control Of Drive Motor For High-Altitude Ventilator

Stratospheric Airships have the advantages of low manufacturing costs,low maintenance costs,long standing time,good stealth and wide coverage.They can be widely used in reconnaissance and surveillance,early warning of war zones,communication relay and other fields,and have become a hot research topic in many countries.Keeping shape and decline is the key technology of stratospheric airship recovery,and it is also one of the difficulties of stratospheric airship technology.The high-altitude ventilator electric drive system for stratospheric airship is the key equipment and core technology to realize the decline.Aiming at the special application background of stratospheric airship,the main problems of induction motor drive system for high-altitude ventilator include: research on mathematical modeling and load characteristics of high altitude ventilator with variable altitude and pressure difference,research on low switching frequency modulation technology of induction motor for high-altitude ventilator,research on flux weakening control technology of induction motor considering load characteristics of high altitude ventilator,research on robust control technology of induction motor based on realtime variation of atmospheric parameters,research on high performance speed sensorless control technology of induction motor.In recent years,finite set model predictive control has been widely studied in the field of power electronics and electric drive because of its intuitive concept,fast dynamic response and flexible handling of nonlinear constraints.In this paper,finite set model predictive control is applied to the induction motor drive system for high altitude ventilator,and the research difficulties in the induction motor drive system are deeply studied.The main contents of this paper are as follows:(1)Different from the ground fan,the inlet and outlet of high-altitude fan are connected with the atmosphere and the auxiliary airbag of the airship,respectively.The operating height of the high-altitude fan changes from the ground to the stratosphere.Therefore,the operating states of the high-altitude fan include four state variables,namely,altitude,pressure difference,speed and flow.Firstly,a four-dimensional mathematical model of high-altitude fan is established,which describes the coupling relationship between the four state variables.Secondly,based on the principle of fan similarity,the flow-pressure difference-speed characteristics and flowpressure difference-power characteristics of the high-altitude ventilator at different heights are studied.And the speed-torque characteristic curves of the high-altitude fan are derived.The research results show that the ventilator is in a light load at high speed due to the thin air at high altitude.(2)Induction motor for high-altitude ventilator are driven by voltage source inverter.Because of the thin air at high altitude and poor heat dissipation conditions,the voltage source inverter must be equipped with complex cooling system.In order to solve this problem,this paper proposes two-vector predictive current control and two-vector predictive torque control of induction motor based on optimized duty cycle,which method reduces the switching frequency of voltage source inverter,the switching loss,and the harmonic content of stator current of induction motor.This method applies two voltage vectors in each control cycle.Firstly,three sets of double vectors are selected according to deadbeat voltage vector,then duty cycle of each set of double vectors is calculated based on optimization method,and finally,the optimal double vectors are selected based on the cost function.In the experimental part,deadbeat predictive current control,single-vector predictive current control,double-vector predictive current control,deadbeat predictive torque control,single-vector predictive torque control and double-vector predictive current control are compared in detail from the aspects of dynamic response,steadystate performance,inverter switching frequency and stator current harmonics.Experimental results show that the proposed method can simultaneously reduce the switching frequency and stator current harmonics.(3)Limited by the airship’s size and carrying capacity,the airship carries limited energy.In order to solve this problem,this paper proposes a flux weakening predictive control of induction motor based on maximum torque per ampere(MTPA),which method reduces the amplitude of stator current and the copper consumption of induction motor,and improves the efficiency of electric drive system of high altitude ventilator.Based on the characteristic of the high-altitude ventilator in the state of light load at high speed,a nonlinear constrained optimization equation of induction motor based on MTPA is established,which takes the voltage limit circle and current limit circle as constraints,and the minimum stator current as the optimization objective.Since the global analytical expressions of nonlinear constrained optimization equations can not be obtained at present,this paper proposes the zoning principle to solve the optimal stator current reference values in the constant torque region,the constant power region and the constant voltage region,respectively.Because the expression of the optimal stator current reference values depend on the parameters of induction motor,this paper studies the robustness of the optimal stator current reference value on the parameters of induction motor.In order to quickly track the optimal stator current reference value,a predictive current control strategy is proposed to achieve the fast dynamic response of the stator current.The experimental results show that compared with the field weakening control based on maximum torque output control,this method reduces the stator current amplitude of induction motor,reduces copper consumption and improves the efficiency of induction motor.(4)Under the influence of stratospheric atmosphere parameters,the realtime changes of induction motor parameters and load torque will lead to the reduction of control performance of induction motor.To solve this problem,a robust predictive torque control based on disturbance feedforward compensation is proposed,which reduces the influence of internal and external disturbances and improves the control performance of induction motor drive system.Firstly,the influence mechanism of mismatched parameters on electromagnetic torque prediction error and stator flux magnitude prediction error is discussed.The disturbance rejection performance of predictive torque control based on proportional integral regulator is studied.The results show that the proportional integral regulator can not achieve optimal control of dynamic performance and robust performance.Based on this,a disturbance observer is proposed to estimate the total disturbance,including load disturbance,parameter perturbation disturbance,predictive error disturbance and so on.Then the optimal control of dynamic performance and robust performance is realized based on the total disturbance feedforward compensation strategy.The experimental results show that compared with the traditional proportional-integral regulator,the method improves the dynamic performance,steady state performance and robust performance of the induction motor.(5)In the 0-20 km operating environment of the high-altitude fan electric drive system,the installation and use of the speed sensor of the induction motor reduces the reliability of the electric drive system of the high-altitude fan.To solve this problem,a speed sensorless predictive control of induction motor based on dual reference frame is proposed in this paper.In the traditional adaptive observer,the speed estimation of induction motor is coupled with the stator flux observation.When the parameters of induction motor change,they affect each other,which reduces the estimation accuracy of both.The observer based on dual reference frame in this paper abandons the traditional adaptive observer which takes the rotor speed as the adaptive variable.The proposed observer represents the stator flux equation and the rotor flux equation in the stator reference frame and the rotor flux reference frame,respectively,which decouples the flux observation and speed estimation.Secondly,a predictive torque control based on dual reference frame is proposed to eliminate the influence of speed estimation error on electromagnetic torque prediction.Aiming at the influence of mismatched parameters of induction motors,a correction term is inserted into the predictive torque equations based on dual reference frame,which is derived from the correction term of the proposed observer.The experimental results show that compared with the predictive torque control based on the Luenberger observer,this method improves the estimation accuracy of the induction motor speed in the high speed,low speed and reverse states.