WO2018069865A2 - 유도전동기의 자속관측기 및 자속 추정 방법 - Google Patents

유도전동기의 자속관측기 및 자속 추정 방법 Download PDF

Info

Publication number
WO2018069865A2
WO2018069865A2 PCT/IB2017/056323 IB2017056323W WO2018069865A2 WO 2018069865 A2 WO2018069865 A2 WO 2018069865A2 IB 2017056323 W IB2017056323 W IB 2017056323W WO 2018069865 A2 WO2018069865 A2 WO 2018069865A2
Authority
WO
WIPO (PCT)
Prior art keywords
flux
model
magnetic flux
angular velocity
calculating
Prior art date
Application number
PCT/IB2017/056323
Other languages
English (en)
French (fr)
Other versions
WO2018069865A3 (ko
Inventor
최종우
조권재
Original Assignee
경북대학교 산학협력단
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 경북대학교 산학협력단 filed Critical 경북대학교 산학협력단
Publication of WO2018069865A2 publication Critical patent/WO2018069865A2/ko
Publication of WO2018069865A3 publication Critical patent/WO2018069865A3/ko

Links

Classifications

    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02P—CONTROL OR REGULATION OF ELECTRIC MOTORS, ELECTRIC GENERATORS OR DYNAMO-ELECTRIC CONVERTERS; CONTROLLING TRANSFORMERS, REACTORS OR CHOKE COILS
    • H02P21/00—Arrangements or methods for the control of electric machines by vector control, e.g. by control of field orientation
    • H02P21/14—Estimation or adaptation of machine parameters, e.g. flux, current or voltage

Definitions

  • the present invention relates to the magnetic flux estimation method and the magnetic flux observer of the induction motor, and more specifically, the magnetic flux estimation method of the induction motor robust to the parameter error and
  • Electric motors are the most important power source in modern society and are used in many fields such as various industrial fields and household appliances.
  • induction motors with high durability and high efficiency are capable of instantaneous torque control due to the emergence of vector control theory. It is widely used for precision driving.
  • Direct vector control based on rotor flux is mainly used.
  • a typical method is to use a voltage model, a current model, and a state observer.
  • the voltage model has the advantage of being robust to the error of the rotor parameters, but it is implemented in this area because of the low back EMF and the influence of noise in the low speed region.
  • the voltage model is not purely integral because of the town set voltage component, which makes it difficult to implement.
  • the current model is sensitive to rotor parameter errors, so the magnetic flux error is increased.
  • the closed-loop Gopinas model which combines the advantages of the model, has emerged.
  • the Gopinas model has a disadvantage in that the magnetic flux error is larger than that of the current model due to the nature of the characteristic function.
  • the Gopinas model uses a current model in the full speed range, which results in a larger error.
  • the purpose of the present invention is to provide a magnetic flux observer and a magnetic flux estimation method of an induction motor which can efficiently control the induction motor by accurately identifying the position of the rotor magnetic flux by reducing the error of the magnetic flux.
  • a magnetic flux observer and a method of estimating the magnetic flux are proposed. Basically, the magnetic flux observer of the Gopinas model is used and realized through the interpretation of the characteristic function.
  • the present invention is to achieve the above object, the current model flux observer receives the stator current and calculates the current model estimated rotor flux ⁇ ) and the current model r em
  • Gopinas magnetic flux observer including magnetic flux observer stator voltage and the stator
  • Calculating a modified second magnetic flux by multiplying the estimated current magnetic flux calculated by the second model by a second gain value d _) and a first magnetic flux modified by the first adder and the modified second magnetic flux; And calculating the final estimated rotor flux ⁇ ).
  • the final estimated rotor flux ( ⁇ ) may be the estimated rotor flux (s) of the voltage model flux observer in steady state.
  • the first gain value can be calculated as the product of the inverse of the steady state characteristic function of the Gopinas model, the square of the ratio of the synchronous angular velocity to the preset boundary angular velocity (03 ⁇ 4). have.
  • the calculating of the first magnetic flux may include receiving the estimated synchronous angular velocity, calculating the first gain value based on the estimated synchronous angular velocity, and calculating the second magnetic flux, receiving the estimated synchronous angular velocity.
  • the second gain value 1_ ⁇ may be calculated based on the estimated synchronous angular velocity.
  • the present invention receives stator current and inputs current model estimated rotor flux ( ⁇ -). Comprising a current model magnetic flux observer and the current model magnetic flux observer, and receives a stator voltage and the stator current and calculates a pinas model estimated rotor magnetic flux (3 ⁇ 4).
  • the first operator calculates the first magnetic flux multiplied by the Gopinas model estimated rotor flux () by the i gain value, and the second gain value (1) multiplied by the calculated current model estimated rotor flux (-).
  • the final estimated rotor flux ( ⁇ s ) is obtained by adding the second operator that calculates the second flux and the corrected first flux and the modified second flux.
  • a flux observer for an induction motor including a first adder to calculate is provided.
  • the present invention can implement a voltage model that is difficult to implement due to a pure integration problem in a steady state, and a current and voltage hybrid model that outperforms the existing Gopinas model in a low speed region in which the voltage model has poor performance. It is possible to estimate the magnetic flux vector robust to the parameter error of the induction motor.
  • FIG. 1 and 2 are block diagrams showing a flux estimation algorithm of a flux observer according to an embodiment of the present invention.
  • FIG. 3 is a flowchart illustrating a magnetic flux estimation method of the magnetic flux observer according to the exemplary embodiment of the present invention.
  • FIG. 4 is a flowchart showing a method of estimating the synchronous angular velocity.
  • 5 is a block diagram showing an estimator of synchronous angular velocity.
  • FIG. 6 illustrates a characteristic function according to the change in the synchronous angular velocity based on the boundary angular velocity.
  • FIG. 7 is a table showing the specifications and parameters of the induction motor for simulating and experimenting the magnetic flux estimation method of the induction motor according to an embodiment of the present invention.
  • FIGS. 8 and 9 are graphs showing simulation results of a flux estimation method of an induction motor according to an exemplary embodiment of the present invention.
  • 10 and 11 are graphs showing the results of experiments using the magnetic flux estimation method of the induction motor according to an embodiment of the present invention.
  • first and second refer to various components, They can be formulated regardless of their order and / or importance, and are used to distinguish one component from another, but do not limit those components.
  • a first user device and a second user device Regardless, it is possible to represent different user devices.
  • a first component can be named as a second component without departing from the scope of rights described in this document. It can be renamed as an element.
  • 1 and 2 are magnetic flux estimation of the magnetic flux observer 100 according to an embodiment of the present invention.
  • the induction motor magnetic flux observer 100 is a current model.
  • the magnetic flux observer 10, the Gopinas model magnetic flux observer 30, the first operator 110, the second operator 120 and the first adder 130 may be included.
  • the current model magnetic flux observer 10 can calculate the current model estimated rotor flux by receiving the stator current.
  • the voltage model magnetic flux observer 20 receives the stator current and the stator voltage to calculate the voltage model estimated rotor flux. Can be calculated.
  • the gopinas model magnetic flux observer 30 may include the current model magnetic flux observer 10 and the voltage model magnetic flux observer 20, and receives the stator voltage and the stator current. Magnetic flux can be calculated.
  • the gopinas model estimated rotor flux is represented by a high pass filter and a low pass filter, respectively, for the voltage model and the current model estimated rotor flux.
  • the DC component of the gopinas model estimated rotor flux is based on the DC component of the voltage model.
  • the cutoff frequency can also be interpreted as the transition frequency that is the boundary between the voltage model and the current model. It is difficult to estimate the flux from the model, and the Gopinas model does not perform well at medium speeds.
  • the magnetic flux observer 100 according to an embodiment of the present invention according to FIG. 1 is as follows.
  • the characteristic function of the magnetic flux observer 100 is as follows, and becomes 1 in the normal state.
  • FIG. 3 illustrates a method of estimating magnetic flux by the magnetic flux observer 100 of FIGS. 1 and 2.
  • the current model flux observer 10 may receive a stator current to calculate a current model estimated rotor flux (S310).
  • a Gopinas model including the current model flux observer 10 The magnetic flux observer 30 receives the stator voltage, the stator current and the current model estimated rotor flux, and calculates the pinas model estimated rotor flux.
  • the corrected first flux may be calculated by multiplying the first gain value by the Gopinas model estimated rotor flux from which the first operator 110 is calculated.
  • the second operator 120 may calculate the modified second magnetic flux by multiplying the calculated current model estimated rotor flux by a second gain value.
  • the first adder 130 may calculate the final estimated rotor flux by adding the modified first flux and the modified second flux.
  • FIG. 4 is a flowchart showing a method of estimating the synchronous angular velocity.
  • the synchronous angular velocity e is required for the calculation of the first gain ( ⁇ ) for implementing the magnetic flux observer 100 according to the embodiment of the present invention.
  • the synchronous angular velocity is the synchronous reference frame of the rotor. Voltage equation
  • the subtractor 210 receives the first magnetic flux angle of the final estimated rotor flux and the second magnetic flux angle fed back from the integrator 240 to calculate an error of the magnetic flux angle (S410).
  • the proportional integral controller 220 may calculate the compensation angular velocity ⁇ ⁇ ⁇ from the error of the calculated magnetic flux angle (S420).
  • the second adder 230 calculates the calculated angular velocity and the parameters of the induction motor.
  • the synchronous angular velocity may be calculated by adding the forward compensating angular velocity (S430).
  • the integrator 240 may receive the calculated synchronous angular velocity to calculate the second magnetic flux angle (S440).
  • the estimated synchronous angular velocity is a flux estimation method of the magnetic flux observer 100, and the step of calculating the first magnetic flux receives a first gain value ( ⁇ ) based on the estimated synchronous angular velocity based on the estimated synchronous angular velocity.
  • the calculating and calculating the second magnetic flux may be performed by receiving the estimated synchronous angular velocity and calculating a second gain value (1 _) based on the estimated synchronous angular velocity.
  • FIG. 5 is a block diagram illustrating a synchronous angular velocity estimator 200.
  • the synchronous angular velocity estimator 200 includes a subtractor 210, a proportional integral controller 220, and a second
  • It may include an adder 230 and an integrator 240.
  • the subtractor 210 receives feedback from the first magnetic flux angle of the final estimated rotor flux and the integrated second magnetic flux angle of the estimated synchronous angular velocity from the integrator 240 to obtain an error of the magnetic flux angle.
  • the proportional integral controller 220 calculates the compensation angular velocity ⁇ ⁇ from the error of the magnetic flux angle.
  • the second adder 230 obtains the synchronous angular velocity obtained by using the parameters of the induction motor.
  • the transient state is compensated for while the second flux angle converges to the first flux angle (). After this process, it is output at the synchronous angular velocity ('' ⁇ ;).
  • FIG. 6 is a graph showing a characteristic function according to the change in the synchronous angular velocity based on the boundary angular velocity.
  • the synchronous angular velocity e is less than the preset boundary angular velocity, it can be implemented as a hybrid model of the voltage model and the current model.
  • the final estimated rotor flux is the voltage model estimated rotor flux, which consists of the inverse of the characteristic function of the logo pinas model.
  • the characteristic function of the mixed model in the steady state obtained by using the first gain is as follows.
  • the estimated flux vector is located in the voltage model above the boundary frequency, and below the boundary frequency, it is a mixed model located at the intermediate point between the current model and the voltage model.
  • ⁇ ⁇ 2 As the characteristic function is designed to be a real number, the magnetic flux observer model according to the embodiment of the present invention has a small magnetic flux error even at a medium speed unlike the existing Gopinas model, so that a more accurate magnetic flux can be estimated.
  • FIG. 7 is a table showing the specifications and parameters of the induction motor for simulating the magnetic flux estimation method of the induction motor according to an embodiment of the present invention.
  • the comparison targets of the magnetic flux observer are the existing voltage model, the current model, the Gopinas model, and the model of the magnetic flux observer 100 according to the embodiment of the present invention.
  • the magnetic flux observer (100) model is a normal voltage model.
  • the cut-off frequency of the Gopinas model and the flux observer 100 is designed to be 10 Hz so that the gain of the proportional controller is the same so that the conditions are the same.
  • the boundary frequency of the mixed model and the voltage model of the magnetic flux observer 100 model according to an embodiment of the present invention is 10 Hz.
  • the flux error that is evaluated for performance is defined as the distance between the actual rotor flux vector and the estimated rotor flux vector of the magnetic flux observer.A half load is applied during operation at the first no load.
  • the parameter error is the rotor parameter cycle. An error was caused in the electronic resistance over mutual inductance.
  • the induction motor is operated at 300 rpm.
  • the flux observer 100 is implemented as a voltage model.
  • the Gopinas model partially uses the voltage model, but the magnetic flux error is larger than the current model due to the phase of the characteristic function.
  • the induction motor is driven at 1000 rpm.
  • the Gopinas model has a much smaller magnetic flux error than the current model because it is significantly higher than the voltage model at high speeds above the transition frequency.
  • the flux observer 100 model according to one embodiment of the present invention operates almost identically to the voltage model.
  • 10 to 11 are graphs showing the results of experiments using the magnetic flux estimation method of the induction motor according to an embodiment of the present invention.
  • the conditions of the experiment are the same as the conditions of the simulation. Similar to the simulation, the comparison targets are the current model, the Gopinas model and the flux observer (100) model according to one embodiment of the present invention.
  • FIG. 10 shows a case where there is an error in the rotor resistance. There is no difference at no load, but the difference appears after the load is applied.
  • the magnetic flux observer 100 is implemented as a voltage model in a steady state, and a magnetic flux error is 0 Wb.
  • the magnetic flux observer 100 model according to an embodiment of the present invention Has a smaller magnetic flux error than the Gopinas model current model at medium and high speeds.
  • the magnetic flux error is lower than the conventional model because it is designed to be out of phase with the characteristic function unlike the existing Gopinas model.
  • the flux observer 100 is implemented as a voltage model, so both no-load and half loads are close to 0 Wb of flux error.
  • the flux error of the flux observer 100 model according to an embodiment is lower than that of the conventional model.
  • the error of the estimated flux in the medium velocity is large depending on the property of the characteristic function.
  • the magnetic flux observer 100 according to the embodiment of the present invention has the estimated flux. Since errors are used at medium speeds and high speeds, errors rarely occur in steady state, and mixed models are used at low speeds, resulting in fewer errors than Gopinas models. Therefore, when the magnetic flux observer 100 according to the embodiment of the present invention is used, it can be seen that the error of the estimated flux rarely occurs in the full speed range. Therefore, the rotor parameter error of various induction motors is robust.
  • a magnetic flux observer 100 is provided.

Landscapes

  • Engineering & Computer Science (AREA)
  • Power Engineering (AREA)
  • Control Of Ac Motors In General (AREA)
  • Control Of Electric Motors In General (AREA)

Abstract

유도전동기의 자속관측기 및 자속 추정 방법이 개시된다. 유도전동기의 자속 추정방법은 전류 모델 자속관측기가 고정자 전류를 입력받아 전류 모델 추정 회전자 자속을산출하는 단계와 전류 모델 자속관측기를 포함하는 고피나스 자속관측기가 고정자전압과 고정자 전류를 입력받아 고피나스 모델 추정 회전자 자속을 산출하는 단계와 제1연산기가 산출된 고피나스 모델 추정 회전자 자속에 제1 이득값을 곱한 수정된 제1자속을 산출하는 단계와 제2 연산기가 산출된 전류 모델 추정 회전자 자속에 제2이득값을 곱한 수정된 제2 자속을 산출하는 단계와 제1 가산기가 수정된 제1 자속 및수정된 제2 자속을 더하여 최종 추정 회전자 자속(I)을 산출하는 단계를 포함하는것을 특징으로 한다.

Description

명세서
발명의명칭:유도전동기의자속관측기및자속추정방법 기술분야
[1] 본발명은유도전동기의자속추정방법및자속관측기에관한것으로서 ,더욱 상세하게는파라미터오차에강인한유도전동기의자속추정방법및
자속관측기에관한것이다.
배경기술
[2] 전동기는현대사회에서가장중요한동력원으로다양한산업분야및생활 가전등많은영역에사용되고있다.특히,내구성과효율이높은유도전동기는 백터제어이론의등장으로인해순시토크제어가가능해져전동기의정밀 구동에많이이용되고있다.회전자자속기준직접백터제어가주로
사용되는데,자속정보를알아내기위해자속관측기를이용하여이를추정하는 방식이주로이용된다.
[3] 대표적인방식으로전압모델,전류모델과상태관측기를이용하는방법이 있다.전압모델은회전자파라미터의오차에강인하다는장점이 있으나,저속 영역에서는역기전력이낮고노이즈의영향이크기때문에이영역에서는 구현이어려운단점이 있다.추가적으로,전압모델은읍셋전압성분때문에 순수적분이불가능하여실제구현하는것이어렵다.한편,전류모델은회전자 파라미터오차에민감하기때문에자속의오차가커지는단점이있다.그래서두 모델의장점을흔합한폐루프고피나스모델이등장하였다.그러나,전동기 파라미터의오차가있을경우,고피나스모델은특성함수의성질때문에중속 영역에서오히려전류모델보다자속오차가커지는단점이있다.또한, 고피나스모델은전속도영역에서전류모델을이용하므로오차가더커지는 단점이 있다.
발명의상세한설명
기술적과제
[4] 본발명의목적은자속의오차를줄임으로써회전자자속의위치를정확하게 파악하여유도전동기를효율적으로제어할수있는유도전동기의자속관측기 및자속추정방법을제공함에 있다.
과제해결수단
[5] 상기문제점을해결하기위하여,읍셋전압에영향이없는전압모델
자속관측기및자속추정방법을제안한다.기본적으로고피나스모델의 자속관측기를이용하고,특성함수의해석을통해서실현된다.
[6] 본발명은상기목적을달성하기위해,전류모델자속관측기가고정자전류를 입력받아전류모델추정회전자자속 π )을산출하는단계와상기전류모델 r em
자속관측기를포함하는고피나스자속관측기가고정자전압과상기고정자 전류를입력받아고피나스모델추정회전자자속 ( ^ )을산출하는단계와 제 1연산기가산출된상기고피나스모델추정회전자자속에제 1이득값 을 곱한수정된제 1자속을산출하는단계와제 2연산기가산출된상기전류모델 추정회전자자속에제 2이득값 d _ )을곱한수정된제 2자속을산출하는 단계와제 1가산기가상기수정된제 1자속및상기수정된제 2자속을더하여 최종추정회전자자속 Π )을산출하는단계 ;를포함하는유도전동기의자속 추정방법을제공한다.
[7] 동기각속도 (∞ 가기설정된경계각속도 ( o^)보다큰경우,상기최종추정 회전자자속 ( ^ )은정상상태에서전압모델자속관측기의추정회전자자속 ( s )일수있다.
! u r_vm
[8] 상기제 1이득값으)은, 1 에의해산출되고,상기 F s)는 kf=——— \s =Jm e ^ 고피나스모델의특성함수인 2 이다.
τρ S
[9] 동기각속도 가기설정된경계각속도 보다작은경우,상기최종 추정회전자자속 ( ^ )은전압모델자속관측기와전류모델자속관측기의
Λ r new
혼합된추정회전자자속일수있다.
[10] 동기각속도 가기설정된경계각속도 보다작은경우,상기제 1 이득값 은,기설정된경계각속도 ( 0¾)에대한동기각속도 의비의 제곱에고피나스모델의정상상태특성함수의역수의곱으로산출될수있다.
[11] 감산기가상기최종추정회전자자속 ( Λ λ ^ )의제 1자속각과적분기로부터
Λ r_new
피드백된제 2자속각을입력받아자속각의오차를산출하는단계와비례적분 제어기가산출된상기자속각의오차로부터보상동기각속도 (Δω 를산출하는 단계와제 2가산기가상기산출된보상동기각속도와상기유도전동기의 파라미터를이용하여산출된전향보상각속도를더하여상기동기각속도를 추정하는단계와상기적분기가상기산출된동기각속도를입력받아상기제 2 자속각을산출하는단계 ;를더포함하고,상기제 1자속을산출하는단계는상기 추정된동기각속도를입력받아상기추정된동기각속도를기초로상기제 1 이득값 를산출하고,상기제 2자속을산출하는단계는상기추정된동기 각속도를입력받아상기추정된동기각속도를기초로상기제 2이득값 ( 1 _^)를 산출할수있다.
[12] 본발명은고정자전류를입력받아전류모델추정회전자자속 ( ^ - )을 산출하는전류모델자속관측기와상기전류모델자속관측기를포함하며 , 고정자전압과상기고정자전류를입력받아고피나스모델추정회전자자속 ( ¾ - )을산출하는고피나스모델자속관측기와산출된상기고피나스모델 추정회전자자속 ( )에제 i이득값 )을곱한수정된제 1자속을 산출하는제 1연산기와산출된상기전류모델추정회전자자속 ( - )에제 2 이득값 ( 1 을곱한수정된제 2자속을산출하는제 2연산기와상기수정된제 1 자속및상기수정된제 2자속을더하여최종추정회전자자속 ( ^ s )을
r_new
산출하는제 1가산기；를포함하는유도전동기의자속관측기를제공한다.
발명의효과
[13] 본발명은순수적분문제로인하여구현이어려운전압모델을정상상태에서 구현을할수있으며,전압모델의성능이좋지않은저속영역에서는기존의 고피나스모델보다성능이뛰어난전류및전압흔합모델을제공하여, 유도전동기의파라미터오차에강인한자속백터의추정이가능하다.
도면의간단한설명
[14] 도 1및도 2는본발명의일실시예에따른자속관측기의자속추정알고리즘을 나타내는블록선도 (Block diagram)이다.
[15] 도 3은본발명의일실시예에따른자속관측기의자속추정방법을나타내는 순서도이다.
[16] 도 4는동기각속도를추정하는방법을나타내는순서도이다.
[17] 도 5는동기각속도의추정기를나타내는블록선도이다.
[18] 도 6은경계각속도를기준으로동기각속도의변화에따른특성함수를
나타내는그래프이다.
[19] 도 7은본발명의일실시예에따른유도전동기의자속추정방법을시뮬레이션 및실험하기위한유도전동기의사양및파라미터를나타낸표이다.
[20] 도 8및도 9는본발명의일실시예에따른유도전동기의자속추정방법을 시뮬레이션한결과를나타낸그래프이다.
[21] 도 10및도 11은본발명의일실시예에따른유도전동기의자속추정방법을 이용하여실험한결과를나타낸그래프이다.
발명의실시를위한형태
[22] 이하,본문서의다양한실시예가첨부된도면을참조하여기재된다.그러나 이는본문서에기재된기술을특정한실시형태에대해한정하려는것이 아니며,본문서의실시예의다양한변경 (modifications),균등물 (equivalents), 및 /또는대체물 (alternatives)을포함하는것으로이해되어야한다.도면의설명과 관련하여,유사한구성요소에대해서는유사한참조부호가사용될수있다.
[23] 또한,본문서에서사용된 "제 1," "제 2,"등의표현들은다양한구성요소들을, 순서및 /또는중요도에상관없이수식할수있고,한구성요소를다른 구성요소와구분하기위해사용될뿐해당구성요소들을한정하지않는다.예를 들면,제 1사용자기기와제 2사용자기기는,순서또는중요도와무관하게 , 서로다른사용자기기를나타낼수있다.예를들면,본문서에기재된권리 범위를벗어나지않으면서제 1구성요소는제 2구성요소로명명될수있고, 유사하게제 2구성요소도제 1구성요소로바꾸어명명될수있다.
[24] 본문서에서사용된용어들은단지특정한실시예를설명하기위해사용된 것으로,다른실시예의범위를한정하려는의도가아닐수있다.단수의표현은 문맥상명백하게다르게뜻하지않는한,복수의표현을포함할수있다.
기술적이거나과학적인용어를포함해서여기서사용되는용어들은본문서에 기재된기술분야에서통상의지식을가진자에의해일반적으로이해되는것과 동일한의미를가질수있다.본문서에사용된용어들중일반적인사전에 정의된용어들은,관련기술의문맥상가지는의미와동일또는유사한의미로 해석될수있으며,본문서에서명백하게정의되지않는한,이상적이거나 과도하게형식적인의미로해석되지않는다.경우에따라서,본문서에서정의된 용어일지라도본문서의실시예들을배제하도록해석될수없다.
[25] 도 1및도 2는본발명의일실시예에따른자속관측기 (100)의자속추정
방법을나타내는블록선도 (Block diagram)이다.
[26] 도 1및도 2를참조하면,유도전동기자속관측기 (100)는전류모델
자속관측기 (10),고피나스모델자속관측기 (30),제 1연산기 (110),제 2 연산기 (120)와제 1가산기 (130)를포함할수있다.
[27] 전류모델자속관측기 (10)는고정자전류를입력받아전류모델추정회전자 자속을산출할수있다.전압모델자속관측기 (20)는고정자전류및고정자 전압을입력받아전압모델추정회전자자속을산출할수있다.
[28] 고피나스모델자속관측기 (30)는상기전류모델자속관측기 (10)및전압모델 자속관측기 (20)를포함할수있으며,고정자전압과상기고정자전류를 입력받아고피나스모델추정회전자자속을산출할수있다.
[29] 도 1로부터고피나스모델자속관측기 (30)의고피나스모델추정회전자자속 식
Figure imgf000006_0001
[32] 상기식을살펴보면고피나스모델추정회전자자속은전압모델과전류모델 추정회전자자속에각각고주파통과필터 (high pass filter)와저주파통과 필터 (low pass filter)를거친형태로나타난다. [33] 고피나스모델추정회전자자속의직류성분 )은전압모델의직류 성분에상기고피나스모델추정회전자자속식의고주파통과필터(
)를거친형태와같다.따라서,정상상태에서아래의식과같이
Figure imgf000007_0001
직류성분이 0이되므로고피나스모델로전압모델의순수적분문제를해결할 수있다.
Figure imgf000007_0002
[35] 한편, F C 를고피나스모델의특성함수 로정의하면,
F g, ra( )= 2
s +Kps+Kj
고피나스모델추정회전자자속식은아래와같이정리된다.
[36] 八 ^ 八 ^
^ r_gm F g ( ) ^ r_v "^" ^ ^gmC )入 r_cm ^ + ( gm C ( ^ r_ym r_ctn) [37] 고피나스모델의특성함수를운전각속도 e)에서주파수웅답해석을하면, 다음과같다.
Figure imgf000007_0003
[39] 고피나스모델의특성함수의크기 씨) I는운전주파수가 Ό'이면 Ό'이되고, 운전주파수가 'co'이면 Τ이된다.그러므로저속에서는전류모델이
우세 (dominant)하며고속에서는전압모델이우세하다.또한,차단주파수는전압 모델과전류모델의경계가되는전이주파수 (transition frequency)로해석될수 있다.그러나중속에서는실제자속의위치를신뢰할수있는전류모델과전압 모델로부터자속을추정하기어려운문제점이 있고,고피나스모델은중속에서 성능이좋지않다.
[40] 고피나스모델의전이주파수를매우낮게설정한다면,고피나스모델은
대부분속도영역에서전압모델로구현될수있다.그러나고피나스모델의 전이주파수를낮게하여고피나스모델이전압모델과유사하게동작되더라도 과도상태 (transient-state)에서자속의직류성분이오랫동안남아있게되므로 정상상태도달시간이길어지는단점이있다.
[41] 도 1에따른본발명의일실시예에따른자속관측기 (100)는아래의식과같다.
[42] 최종추정회전자자속은
[43]
r_new ~^f ^ r_gm + ( 1 ~^/)人 r_cm
Figure imgf000007_0004
λ r_cm [45] 로나타나고,자속관측기 (100)의특성함수는다음과같은새로운특성함수 F„e =k/^ 로변경된다.여기서, 는다음과같이고피나스모델의정상 상태에서의특성함수인 의역수이다.
[46] 1 κ . 와같이제 1이득 ^가결정되면，새로운
Figure imgf000008_0001
자속관측기 (100)의특성함수는다음과같고,정상상태에서는 1이된다.
Figure imgf000008_0002
[49] 따라서 ,정상상태에서최종추정회전자자속은다음과같다.
Figure imgf000008_0003
[51] 도 1및도 2에따른본발명의유도전동기의자속관측기 (100)에의하면,결국 전류모델이상쇄되면서전압모델만을출력할수있다.추정자속의직류성분은 기존의고피나스모델과마찬가지로상쇄되고,순수적분의문제점을해결하는 전압모델이구현될수있다.
[52] 도 3은도 1및도 2의자속관측기 (100)가자속을추정하는방법에대해서
순서도를나타낸것이다.
[53] 도 3을참조하면,전류모델자속관측기 (10)가고정자전류를입력받아전류 모델추정회전자자속을산출할수있다 (S310).전류모델자속관측기 (10)를 포함하는고피나스모델자속관측기 (30)가고정자전압과상기고정자전류및 전류모델추정회전자자속을입력받아고피나스모델추정회전자자속을 산출할수있다. (S320)제 1연산기 (110)가산출된고피나스모델추정회전자 자속에제 1이득값을곱한수정된제 1자속을산출할수있다. (S330)제 2 연산기 (120)가산출된상기전류모델추정회전자자속에제 2이득값을곱한 수정된제 2자속을산출할수있다. (S340)제 1가산기 (130)가수정된제 1자속및 수정된제 2자속을더하여최종추정회전자자속을산출할수있다. (S350)
[54] 도 4는동기각속도를추정하는방법을나타내는순서도이다.
[55] 본발명의일실시예에따른자속관측기 (100)를구현하기위한제 1이득 (^)의 연산과정에는동기각속도 e 필요하다.동기각속도는회전자의동기 좌표계 (synchronous reference frame) q죽전압방정식인
^세 ^ e와회전자의쇄교자속식인
dt 로부터얻어질수있다.동기각속도의추정식은아래와같다.
Figure imgf000009_0001
[57] 상호인덕턴스 ( J와회전자인덕턴스 (Z )의비는약 1로상호인덕턴스 (Z
)의변동에는큰영향이없지만,회전자저항 CRr 오차에영향을끼친다.
[58] 이러한문제를해결하기위해자속관측기 (100)의출력에서자속각 ( ^ )을
미분하여동기각속도 ( 0^)가얻어질수있다.하지만실제전동기구동
시스템에서는고주파와노이즈의때문에미분을구현하는것은불가능하기 때문에도 4및도 5와같이구현하였다.
[59] 도 4를참조하면,감산기 (210)가상기최종추정회전자자속의제 1자속각과 적분기 (240)로부터피드백된제 2자속각을입력받아자속각의오차를산출할수 있다 (S410).비례적분제어기 (220)가산출된자속각의오차로부터보상각속도 ( Δω β)를산출할수있다 (S420).제 2가산기 (230)가산출된보상각속도와 유도전동기의파라미터를이용하여산출된전향보상각속도를더하여동기 각속도를산출할수있다 (S430).적분기 (240)가산출된동기각속도를입력받아 상기제 2자속각을산출할수있다 (S440).상기단계를거쳐동기각속도가 추정된다.
[60] 이렇게추정된동기각속도는자속관측기 (100)의자속추정방법단계에서,제 1 자속을산출하는단계는추정된동기각속도를입력받아추정된동기각속도를 기초로제 1이득값 ( ^)을산출하고,제 2자속을산출하는단계는추정된동기 각속도를입력받아추정된동기각속도를기초로제 2이득값 ( 1 _ )을산출한다.
[61] 이하에서,도 5를참조하여동기각속도추정기 (200)의구성을살펴본다.
[62] 도 5는동기각속도추정기 (200)를나타내는블록선도이다.
[63] 동기각속도추정기 (200)는감산기 (210),비례적분제어기 (220),제 2
가산기 (230)와적분기 (240)를포함할수있다.
[64] 감산기 (210)는최종추정회전자자속의제 1자속각과적분기 (240)로부터추정 동기각속도의적분된제 2자속각을피드백받아자속각의오차를구한다.
비례적분제어기 (220)는자속각의오차로부터보상각속도 (Δω β)를산출한다.
[65] 제 2가산기 (230)는유도전동기의파라미터를이용하여구한동기각속도(
)에서전향보상하여최종출력단의 (X니 r+
Figure imgf000009_0002
제 2자속각이제 1자속각 ( )으로수렴하는동안의과도상태를보상한다. 이러한과정을거쳐최종적으로는동기각속도 (； 'θ；)로출력된다.
e dt e [66] 도 6은경계각속도를기준으로동기각속도의변화에따른특성함수를 나타내는그래프이다.
[67] 그래프에서확인된것처럼,도 4및도 5에서구해진동기각속도 가기
설정된경계각속도 (ω¾)보다큰경우,상기최종추정회전자자속은
정상상태에서전압모델자속관측기 (20)의추정회전자자속으로나타내도록 설계가된다.
[68] 동기각속도 e)가기설정된경계각속도 보다작은경우에는전압모델과 전류모델의흔합모델로구현될수있다.
[69] 우선,동기각속도 e)가기설정된경계각속도 보다큰경우,제 1이득값 ( k;)은 κ. κ 로고피나스모델의특성함수의역수로 이루어져,최종추정회전자자속은전압모델추정회전자자속으로
이루어지게된다.이에대한설명은앞에서언급하였으므로생략한다.
[70] 동기각속도 e)가기설정된경계각속도 보다작은경우,제 1이득값
)은기설정된경계각속도 (ωέ; 1대한동기각속도 (^^(의비의제곱에고피나스 모델의정상상태의특성함수의역수의곱으로산출되어아래와같이표현된다.
Figure imgf000010_0001
[72] 상기제 1이득값을이용하여구해진정상상태에서의흔합모델의특성함수는 다음과같다.
[73]
Fnew (j'He) = {— )
[74] 최종적인정상상태의최종추정회전자자속은다음과같다.
[75]
λ；=(— ) λ； vm+ { 1 )2} λ； cm
[76] 이와같이설계된본발명의추정자속기의경우,경계주파수이상에서는추정 자속백터가전압모델에위치하며,경계주파수이하에서는흔합모델로서전류 모델과전압모델의중간지점에위치한다.특히, ω β 2와같이 특성함수는오직실수로설계되므로,본발명의일실시예에따른자속관측기 모델은기존의고피나스모델과달리중속에서도자속오차가작기때문에,보다 정확한자속이추정될수있다.
도 7은본발명의일실시예인유도전동기의자속추정방법을시뮬레이션하기 위한유도전동기의사양및파라미터를나타낸표이다. [78] 시뮬레이션에의해,본발명의일실시예에따른자속관측기 (100)가경계 주파수이상에서전압모델로구현되는지확인하고자한다.사용된유도 전동기의사양및파라미터는도 7에나타난표와같고,인버터의출력전압의 읍셋성분은정격의 1%인 1.3 V로설정하였다.
[79] 자속관측기의비교대상은기존의전압모델,전류모델,고피나스모델,그리고 본발명의일실시예에따른자속관측기 (100)의모델이다.
[80] 기존의전압모델은 † Γ ( 2 )와
X r vm= v s- R sdt)- σ Lsis} σ= 1 -— ^ ~~―
"
같이연산하되 에읍셋전압이없도록이상적인것으로한것이며,본발명의 일실시예에따른자속관측기 (100)모델이정상상태에서전압모델로
구현되는지확인을위한것이다.고피나스모델과본발명의일실시예에따른 자속관측기 (100)모델의차단주파수는 10 Hz로설계하여비례제어기의이득을 같게하여조건이같도록하였다.본발명의일실시예에따른자속관측기 (100) 모델의흔합모델과전압모델의경계가되는경계주파수는 10 Hz로
설계되었다.그리고성능의평가가되는자속오차는실제회전자자속백터와 자속관측기의추정회전자자속백터의거리로정의되었다.최초무부하에서 운전중에반부하가인가되었다.파라미터오차는회전자파라미터인회전자 저항 과상호인덕턴스( „)에오차를주었다.
[81] 도 8내지도 9는본발명의일실시예인유도전동기의자속추정방법을
시뮬레이션한결과를나타낸그래프이다.
[82] 도 8을참조하면,유도전동기가 300 rpm으로운전하는것을나타낸것이다.
[83] 부하인가전과인가후모든경우의정상상태에서,본발명의일실시예에 따른자속관측기 (100)모델은전압모델로구현된다.비록부하가인가된과도 상태에서본발명의일실시예에따른모델의자속오차가일시적으로
상승하지만,기존의고피나스모델과전류모델보다는자속오차가훨씬낮다. 고피나스모델은전압모델을부분적으로이용하지만,특성함수의위상으로 인하여오히려전류모델보다자속오차가더크게나타난다.
[84] 도 9를참조하면,유도전동기가 1000 rpm운전하는것을나타낸것이다.
고피나스모델은전이주파수보다높은고속에서전압모델로상당수넘어가기 때문에전류모델보다자속오차가작다.본발명의일실시예에따른모델은 300 rpm과마찬가지로전압모델로구현된다.정상상태뿐만아니라과도
상태에서도본발명의일실시예에따른자속관측기 (100)모델은전압모델과 거의동일하게동작한다.
[85] 도 10내지도 11은본발명의일실시예에따른유도전동기의자속추정방법을 이용하여실험한결과를나타낸그래프이다.
[86] 실험의조건은시뮬레이션의조건과동일하다. [87] 시뮬레이션과마찬가지로비교대상은전류모델,고피나스모델과본발명의 일실시예에따른자속관측기 (100)모델이다.
[88] 도 10은회전자저항의오차가있는경우이다.무부하에서는차이가없으나 부하를인가한이후에차이가나타난다. 300 rpm과 lOOO rpm에서본발명의일 실시예에따른자속관측기 (100)모델이정상상태에서전압모델로구현되면서 자속오차가 0 Wb이다.본발명의일실시예에따른자속관측기 (100)모델은 중속과고속에서고피나스모델전류모델보다작은자속의오차를가진다.
[89] 흔합모델로구현되는 100 rpm에서도본발명의일실시예에따른모델은
기존의모델보다자속오차가낮다.기존의고피나스모델과달리특성함수의 위상이없도록설계를했기때문이다.
[90] 도 11은상호인덕턴스의오차가있는경우이다. 300 rpm과 1000 rpm에서본 발명의일실시예에따른자속관측기 (100)모델이전압모델로구현되므로 무부하와반부하모두자속오차가거의 0 Wb에가깝다.흔합모델로구현되는 100 rpm에서도본발명의일실시예에따른자속관측기 (100)모델의자속오차가 기존의모델보다낮다.
[91] 상기시뮬레이션과실험결과를살펴보면,고피나스모델은특성함수의성질에 따라중속에서는추정자속의오차가크게발생된다.그러나본발명의일 실시예에따른자속관측기 (100)모델은추정자속의오차가중속과고속에서는 전압모델을이용하게되므로오차가정상상태에서는거의발생하지않고, 저속에서도흔합모델을이용하므로,고피나스모델보다오차가적게발생된다. 따라서,본발명의일실시예에따른자속관측기 (100)를이용하면,전속도 영역에서추정자속의오차가거의발생하지않음을알수있다.따라서 ,다양한 유도전동기의회전자파라미터오차에도강인한특성을가지는
자속관측기 (100)가제공된다.
[92] 이상과같이,본발명은비록한정된실시 예와도면에의해설명되었으나,본 발명은이것에의해한정되지않으며,본발명이속하는기술분야에서통상의 지식을가진자에의해본발명의기술사상과아래에기재될특허청구범위의 균등범위내에서다양한수정및변형가능함은물론이다.

Claims

청구범위
전류모델자속관측기가고정자전류를입력받아전류모델추정회전자 자속 (스 )을산출하는단계 ;
상기전류모델자속관측기를포함하는고피나스모델자속관측기가 고정자전압과상기고정자전류를입력받아고피나스모델추정회전자 자속으 )을산출하는단계 ; 제 1연산기가산출된상기고피나스모델추정회전자자속에제 1이득값 ( 을곱한수정된제 1자속을산출하는단계；
제 2연산기가산출된상기전류모델추정회전자자속에제 2이득값 ( 1 을곱한수정된제 2자속을산출하는단계 ;및
제 1가산기가상기수정된제 1자속및상기수정된제 2자속을더하여 최종추정회전자자속 ( 1 ^ )을산출하는단계；를포함하는
유도전동기의자속추정방법 .
제 1항에 있어서,
상기최종추정회전자자속 ( )은,
동기각속도 가기설정된경계각속도 (∞ 보다큰경우,
정상상태에서전압모델자속관측기의추정회전자자속 ( ^ ^ )인, 유도전동기의자속추정방법 .
제 2항에 있어서,
상기제 1이득값 은,
1 에의해산출되고,상기 F ( 는고피나스모델의 kf " , . ᅵ지 e 특성함수인 인,유도전동기의자속추정방법.
fr 제 1항에 있어서,
상기최종추정회전자자속 ( )은,
동기각속도 가기설정된경계각속도 (∞ 보다작은경우,전압 모델자속관측기와상기전류모델자속관측기의흔합된추정회전자 자속인,유도전동기의자속추정방법.
제 1항에 있어서,
상기제 1이득값 은,
동기각속도 e 기설정된경계각속도 (ωέ)보다작은경우,기설정된 경계각속도 j에대한동기각속도 j의비의제곱을이용하여,
1 으로산출되고,상기 F ( 는고피나스모델의
Figure imgf000014_0001
특성함수인 인,유도전동기의자속추정방법
F예, ) =
s +K s+Kf
[청구항 6] 제 2항내지제 5항중어느한항에 있어서,
감산기가상기최종추정회전자자속 ( ^ )의제 1자속각과
적분기로부터피드백된제 2자속각을입력받아자속각의오차를 산출하는단계 ;
비례적분제어기가산출된상기자속각의오차로부터보상동기각속도 ( Δω β)를산출하는단계 ;
제 2가산기가상기산출된보상동기각속도와상기유도전동기의 파라미터를이용하여산출된전향보상각속도를더하여상기동기 각속도를추정하는단계；및
상기적분기가상기산출된동기각속도를입력받아상기제 2자속각을 산출하는단계 ;를더포함하고,
상기제 1자속을산출하는단계는상기추정된동기각속도를입력받아 상기제 1이득값 ( )를산출하고,
상기제 2자속을산출하는단계는상기추정된동기각속도를입력받아 상기제 2이득값 ( u 를산출하는,유도전동기의자속추정방법.
[청구항 7] 고정자전류를입력받아전류모델추정회전자자속 ( ^ )을
산출하는전류모델자속관측기 ;
상기전류모델자속관측기를포함하며,고정자전압과상기고정자 전류를입력받아고피나스모델추정회전자자속 ( - )을산출하는 고피나스모델자속관측기 ;
산출된상기고피나스모델추정회전자자속 ( - )에제 1이득값 Ot,
)을곱한수정된제 1자속을산출하는제 1연산기；
산출된상기전류모델추정회전자자속 ( )에제 2이득값 ( 1 - 을 곱한수정된제 2자속을산출하는제 2연산기 ;및
상기수정된제 1자속및상기수정된제 2자속을더하여최종추정 회전자자속 ( - )을산출하는제 1가산기；를포함하는유도전동기의 자속관측기.
PCT/IB2017/056323 2016-10-12 2017-10-12 유도전동기의 자속관측기 및 자속 추정 방법 WO2018069865A2 (ko)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
KR10-2016-0132091 2016-10-12
KR1020160132091A KR101845412B1 (ko) 2016-10-12 2016-10-12 유도전동기의 자속관측기 및 자속 추정 방법

Publications (2)

Publication Number Publication Date
WO2018069865A2 true WO2018069865A2 (ko) 2018-04-19
WO2018069865A3 WO2018069865A3 (ko) 2018-06-07

Family

ID=61906187

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/IB2017/056323 WO2018069865A2 (ko) 2016-10-12 2017-10-12 유도전동기의 자속관측기 및 자속 추정 방법

Country Status (2)

Country Link
KR (1) KR101845412B1 (ko)
WO (1) WO2018069865A2 (ko)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113395028A (zh) * 2021-07-28 2021-09-14 华北电力大学 基于电压电流混合模型的磁链观测器的形成方法

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10840841B2 (en) * 2018-09-27 2020-11-17 Tmeic Corporation Control device for power conversion device, control method, and motor drive system
KR102494504B1 (ko) * 2020-06-01 2023-02-02 경북대학교 산학협력단 센서리스 제어 자속 추정기 및 자속 추정 방법

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20060014541A (ko) * 2004-08-11 2006-02-16 학교법인 울산공업학원 유도전동기 직접 벡터제어를 위한 회전자 시정수 추정장치 및 이를 이용한 유도 전동기 제어 시스템
KR101376389B1 (ko) * 2010-11-30 2014-03-20 엘에스산전 주식회사 유도전동기용 자속 제어장치
KR101759371B1 (ko) * 2013-04-04 2017-07-18 엘에스산전 주식회사 유도전동기의 센서리스 벡터 제어 장치

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113395028A (zh) * 2021-07-28 2021-09-14 华北电力大学 基于电压电流混合模型的磁链观测器的形成方法
CN113395028B (zh) * 2021-07-28 2022-06-28 华北电力大学 基于电压电流混合模型的磁链观测器的形成方法

Also Published As

Publication number Publication date
KR101845412B1 (ko) 2018-04-04
WO2018069865A3 (ko) 2018-06-07

Similar Documents

Publication Publication Date Title
Yoo et al. Design of flux observer robust to interior permanent-magnet synchronous motor flux variation
CN103997272B (zh) 永磁同步电机的负载扰动补偿装置及方法
JP6462241B2 (ja) ファンモータ駆動装置及びブロア
JP5492192B2 (ja) 交流モータの制御装置
CN1741367B (zh) 电动机的无位置传感器控制电路
TW200524265A (en) Controller for synchronous motor, electric appliance and module
JP2011176953A (ja) 回転電気機械のトルクリプル抑制制御装置および制御方法
TWI476409B (zh) Motor speed estimation method
Abo‐Khalil et al. Sensorless control for PMSM using model reference adaptive system
KR101485989B1 (ko) 모터 제어 장치
Lin et al. An improved flux observer for sensorless permanent magnet synchronous motor drives with parameter identification
WO2018069865A2 (ko) 유도전동기의 자속관측기 및 자속 추정 방법
JP4402600B2 (ja) 同期電動機の駆動システム及び同期電動機の駆動方法
WO2020105204A1 (ja) 電力変換装置
Comanescu Implementation of time-varying observers used in direct field orientation of motor drives by trapezoidal integration
Yeo et al. Adaptive feedforward control to compensate cogging torque and current measurement errors for PMSMs
JP6241331B2 (ja) 電動機の制御装置
KR20120106449A (ko) 매입형 영구 자석 동기 전동기의 센서리스 제어를 위한 온라인 상수 보정 방법
Kadjoudj et al. Modified direct torque control of permanent magnet synchronous motor drives
WO2022054357A1 (ja) 電力変換装置
JP2011067066A (ja) 永久磁石形同期電動機の制御装置
WO2022029911A1 (ja) モータ鉄損演算装置およびそれを備えたモータ制御装置
JP2015173546A (ja) 電動機の制御装置
Zhou et al. High-order terminal sliding-mode observer for speed estimation of induction motors
Noguchi et al. Study of parameter variations compensation in sensorless control of PMSM

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 17860116

Country of ref document: EP

Kind code of ref document: A2

NENP Non-entry into the national phase in:

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 17860116

Country of ref document: EP

Kind code of ref document: A2