JP4281054B2 - Speed sensorless vector control method and apparatus for induction motor - Google Patents

Description

  The present invention relates to a speed sensorless vector control method and apparatus for an induction motor using an adaptive secondary magnetic flux observer robust to primary resistance fluctuations.

  The vector control of the induction motor requires a practical means for accurately detecting the secondary flux linkage vector. If a mathematical model of an induction motor is assembled as a simulator on a computer or arithmetic circuit, and the same input (for example, primary voltage) as the actual input is added to the mathematical model, the secondary flux linkage vector can be estimated from the mathematical model. . At this time, if there are constants that make up the mathematical model variable such as the rotational speed, the mathematical model becomes incorrect as the rotational speed changes, so there is a deviation between the mathematical model output and the measured primary current. Arise. Here, if there is a speed sensor, the constant of the mathematical model is constantly corrected using the output of the speed sensor, so that the magnetic flux observer for detecting the magnetic flux vector can be correctly configured.

  However, if an attempt is made to construct a speed sensorless vector control system without using a speed sensor, deviation will occur even if the observer is constructed. Therefore, assuming that the cause of the deviation is that a certain constant (rotational speed) in the mathematical model is not correct, the constant (rotational speed) of the mathematical model is corrected by a function of the output deviation. This is an adaptive observer, and a speed adaptive secondary magnetic flux observer is one that estimates the secondary flux linkage while adaptively estimating the rotational speed (Non-patent Documents 1 and 2).

Nakano "Vector Control of AC Motor", Nikkan Kogyo Shimbun, pp. 98-110, 1996 Kubota, Matsuse, Nakano “Adaptive secondary magnetic flux observer induction motor speed sensorless direct vector control” D-111, 11, 954-960, 1991 Tamura, Kubota “A Study on Estimation of Primary Resistance of Speed Sensorless Vector Control Induction Motor in Low Speed Regeneration”, IEEJ National Conference, 4-135

  However, in the prior art, when a speed sensor is not used, a speed estimation error occurs due to a change in the primary resistance parameter due to a temperature change, so that there is a problem that stable speed control cannot be performed (Non-patent Document 3). In particular, operation has been difficult in a low speed range where the primary voltage is low.

  SUMMARY OF THE INVENTION An object of the present invention is to provide a speed sensorless vector control method and apparatus for an induction motor that can reduce speed estimation errors caused by primary resistance fluctuations.

A speed sensorless vector control method according to the present invention is a speed sensorless vector control method using an adaptive secondary magnetic flux observer that estimates the primary current and secondary magnetic flux of an induction motor. The speed sensorless vector control method includes the following steps <1> to <3>. Estimate the rotation speed. <1>. In an orthogonal coordinate system composed of the d-axis and the q-axis, the angle θ _Rs formed between the primary current estimation error vector e _iRs and the d-axis due to only the primary resistance setting error, and all errors. The coordinates (e _ids , e _iqs ) of the primary current estimation error vector e _is calculated. <2>. In the Cartesian coordinate system composed of the γ-axis and the δ-axis obtained by rotating the d-axis and q-axis by the angle θ _Rs around the origin, the coordinates (e _ids , e _iqs ) are Convert to coordinates (e _iγs , e _iδs ). <3> The rotational speed is estimated using the δ-axis direction component e _iδs . In the procedures <1> to <3>, for example, expressions (1) to (9) described later are used.

The primary current estimation error vector e _is caused by all errors _{is the} primary current estimation error vector e _iRs caused only by the setting error of the primary resistance, or a plurality of primary current estimation error vectors caused only by various other errors. This is a composite vector. Here, the γ-axis coincides with the direction of the primary current estimation error vector e _iRs , and the δ-axis is orthogonal. Therefore, the component resulting from the setting error of the primary resistance _{is included in} the γ-axis direction component e _iγs of the primary current estimation error vector e _is , but is not included in the δ-axis direction component e _iδs . Therefore, the rotational speed estimated using the δ-axis direction component e _iδs is not affected by fluctuations in the primary resistance.

Further, when the angle θ _ωr formed between the primary current estimation error vector e _iωr caused only by the rotational speed estimation error and the d axis is calculated, and the difference between the angle θ _ωr and the angle θ _Rs is less than a certain value The rotational speed may be estimated by adding the γ-axis direction component e _iγs to the δ-axis direction component e _iδs . In this case, the difference between the angle theta _.omega.r and the angle theta _Rs becomes larger the ratio of using the γ-axis direction component e _Aiganmaesu smaller may be.

Primary current estimation error vector e _IS Since mainly depends on the primary current estimation error vector e _Aiomegaaru, is close to the primary current estimation error vector _e iωr. Therefore, as the difference between the angle θ _ωr and the angle θ _Rs decreases, the γ-axis direction component e _iγs increases and the δ-axis direction component e _iδs decreases. Therefore, when the difference between the angle θ _ωr and the angle θ _Rs is small, the accuracy decreases if only the small δ-axis direction component e _iδs is used. The rotational speed is also estimated using the γ-axis direction component e _iγs . At this time, if the ratio using the γ-axis direction component e _iγs is gradually increased as the difference between the angle θ _ωr and the angle θ _Rs becomes smaller, the influence of fluctuations in the primary resistance is minimized.

A speed sensorless vector control apparatus according to the present invention is a speed sensorless vector control apparatus using an adaptive secondary magnetic flux observer for estimating a primary current and a secondary magnetic flux of an induction motor. The adaptive secondary magnetic flux observer includes the following means. ing. <1>. In an orthogonal coordinate system composed of the d-axis and the q-axis, the angle θ _Rs formed between the primary current estimation error vector e _iRs and the d-axis due to only the primary resistance setting error, and all errors. Primary current estimation error vector calculation means for calculating the coordinates (e _ids , e _iqs ) of the primary current estimation error vector e _is based on the primary voltage and the primary current. <2>. In the Cartesian coordinate system composed of the γ-axis and the δ-axis obtained by rotating the d-axis and q-axis by the angle θ _Rs around the origin, the coordinates (e _ids , e _iqs ) are Coordinate conversion means for converting to coordinates (e _iγs , e _iδs ). <3> Rotational speed estimation means for estimating the rotational speed using the δ-axis direction component e _iδs . That is, the speed sensorless vector control apparatus according to the present invention uses the speed sensorless vector control method according to the present invention.

Further, the primary current estimation error vector calculation means calculates an angle θ _ωr formed by a primary current estimation error vector e _iωr caused by only the rotation speed estimation error and the d axis, and the rotation speed estimation means When the difference between θ _ωr and the angle θ _Rs is less than a certain _value , the rotational speed may be estimated by adding the γ-axis direction component e _iγs to the δ-axis direction component e _iδs . At this time, the rotational speed estimation means may increase the ratio of using the γ-axis direction component e _iγs as the difference between the angle θ _ωr and the angle θ _Rs decreases.

  In other words, the present invention constitutes an observer for estimating the primary current and secondary magnetic flux of the induction motor, and the current estimation error caused only by the direction of the current estimation error vector and the speed estimation error caused only by the setting error of the primary resistance. The direction of the vector is grasped, and the speed is estimated based on the current error information in the direction not affected by the primary resistance error. That is, an observer that estimates the primary current and secondary magnetic flux of the induction motor is configured, and the direction of the current estimation error vector caused only by the primary resistance setting error and the direction of the current estimation error vector caused only by the speed estimation error are determined. To grasp. When both vector directions are not close, speed estimation is performed using a current component orthogonal to a current estimation error vector caused only by the primary resistance error. When both vector directions are close to each other, velocity is estimated by applying a horizontal weight to the orthogonal component.

The speed sensorless vector control method and apparatus according to the present invention uses the δ-axis direction component e _iδs not caused by the primary resistance setting error vector e _is caused by all the errors, to _reduce the rotation speed. By estimating, it is possible to obtain an accurate rotation speed that is not affected by fluctuations in the primary resistance.

  That is, it is possible to estimate the speed by reducing the influence of the setting error of the primary resistance as much as possible. As a result, the error between the command value and the actual measurement value can be reduced, so that the effects of inverter dead time and element voltage drop can also be reduced. Similarly, the influence of speed estimation error can be suppressed when estimating the primary resistance.

Further, the angle between the primary current estimation error vector e _IRS due only to the primary current estimation error vector e _Aiomegaaru a setting error of the primary resistance due only to the rotation speed estimation error | is constant below | theta _Rs - [theta] _.omega.r In this case, the rotational speed is also estimated using the γ-axis direction component e _iγs caused by the primary resistance setting error vector e _is , so that the δ-axis direction component e _iδs not caused by the primary resistance setting error is _estimated. Reduces accuracy when using. At this time, as the angle | θ _Rs −θ _ωr | becomes smaller, the ratio of using the γ-axis direction e _iγs due to the setting error of the primary resistance is increased, thereby minimizing the influence of the fluctuation of the primary resistance and _{reducing the} primary resistance. It is possible to suppress a decrease in accuracy when the δ-axis direction component e _iδs that is not caused by the setting error is used.

  FIG. 1 is a block diagram showing a first embodiment of a speed sensorless vector control apparatus according to the present invention. Hereinafter, description will be given based on this drawing.

The speed sensorless vector control device 10 according to the present embodiment includes a general configuration such as a speed regulator 11, a vector rotator 12, a current regulator 13, a voltage-type PWM inverter 14, a current sensor 15, and the like, in addition to the general configuration of the induction motor 20. A novel adaptive secondary magnetic flux observer 21 for estimating the primary current and the secondary magnetic flux is provided. The induction motor 20 has a cage shape and a three-phase. Vector rotator 12 is estimated secondary flux vector phi _r ^ current command value using the i _sM ^*, i _sT ^* current command value in the stator on the coordinates i _sa ^*, and coordinate transformation i _{S [beta} ^*, this Further, after two-phase / three-phase conversion, the current command value of each phase is output. The speed adjuster 11, the vector rotator 12, and the adaptive secondary magnetic flux observer 21 are realized by, for example, a DSP and its program. The current regulator 13 is a PI regulator, for example. The voltage type PWM inverter 14 is a switching circuit made of, for example, a power transistor.

The adaptive secondary magnetic flux observer 21 is characterized by calculating a rotational speed estimated value ω _r ^. The other operations of the speed sensorless vector control apparatus 10 are the same as those of the prior art, and thus the description thereof is omitted.

  FIG. 2 is a block diagram showing an adaptive secondary magnetic flux observer in the sensorless vector control apparatus of FIG. Hereinafter, description will be given based on this drawing.

The adaptive secondary magnetic flux observer 21 is characterized in that it has a novel speed adaptation mechanism 30 for calculating the rotational speed estimated value ω _r ^. The other operations of the adaptive secondary magnetic flux observer 21 are the same as those in the prior art, and the description thereof is omitted.

FIG. 3 is a block diagram showing a speed adaptation mechanism in the adaptive secondary magnetic flux observer of FIG. 4 is a graph showing the operation of the speed adaptation mechanism of FIG. 3. FIG. 4 [1] corresponds to the operation of the primary current estimation error vector calculation means in FIG. 3, and FIG. 4 [2] is the coordinate transformation in FIG. This corresponds to the operation of the means. Hereinafter, description will be given based on these drawings.

The speed adaptation mechanism 30 includes a primary current estimation error vector calculation means 31, a coordinate conversion means 32, a rotation speed estimation means 33, and the like. The primary current estimation error vector calculation means 31 has an angle θ _Rs formed between the primary current estimation error vector e _iRs and the d axis, which is caused only by the primary resistance setting error, in the orthogonal coordinate system including the d axis and the q axis. one due to an error primary current estimation error vector e _iS coordinates (e _ids, e _iqs) and is calculated on the basis of the primary voltage v _s and the primary current i _s (Fig. 4 [1]). The coordinate conversion means 32 coordinates coordinates (e _ids , e _iqs ) in an orthogonal coordinate system composed of γ-axis and δ-axis obtained by rotating the d-axis and q-axis by the angle θ _Rs around the origin. (E _iγs , e _iδs ) (FIG. 4 [2]) . The rotation speed estimation means 33 estimates the rotation speed ω _r using the δ-axis direction component e _iδs . Thus, the speed adaptation mechanism 30 uses the speed sensorless vector control method according to the present invention.

The primary current estimation error vector e _is caused by all errors _{is the} primary current estimation error vector e _iRs caused only by the setting error of the primary resistance, or a plurality of primary current estimation error vectors caused only by various other errors. This is a composite vector. Here, the γ-axis coincides with the direction of the primary current estimation error vector e _iRs , and the δ-axis is orthogonal. Therefore, the component resulting from the setting error of the primary resistance _{is included in} the γ-axis direction component e _iγs of the primary current estimation error vector e _is , but is not included in the δ-axis direction component e _iδs . Therefore, the rotational speed estimated using the δ-axis direction component e _iδs is not affected by fluctuations in the primary resistance.

Next, the operation of the adaptive secondary magnetic flux observer 21 including the speed adaptation mechanism 30 will be described in more detail.

<< 1 >>. 1. Introduction In this embodiment, in speed sensorless vector control of an induction motor using an adaptive observer, a rotational speed estimation error due to primary resistance fluctuation in a low speed region is compensated. In short, the vector direction that depends only on the primary resistance error is calculated, the current estimation error is coordinate-transformed, and the speed estimation is performed using only the current estimation error component in the vector direction shifted by 90 ° with respect to the axis. . As a result, the estimated rotational speed becomes insensitive to the primary resistance fluctuation.

<< 2 >>. Adaptive Secondary Flux Observer The state equation of the induction motor can be expressed as in Equation (1), where the state variables are the primary current and the secondary magnetic flux and the control input is the primary voltage in the rotating coordinate system. Note that ^e represents a rotating coordinate system.

ただし、
ｉ^e _s＝［ｉ^e _ds ｉ^e _qs］^T：一次電流
φ^e _r＝［φ^e _dr φ^e _qr］^T：二次磁束
ｖ^e _s＝［ｖ^e _ds ｖ^e _qs］^T：一次電圧
ｘ^e＝［ｉ^e _s φ^e _r］^T
Ａ₁₁＝−｛Ｒ_s／（σＬ_s）＋Ｒ_r（１−σ）／（σＬ_r）｝Ｉ＝ａ_r11Ｉ
Ａ₁₂＝Ｍ／（σＬ_sＬ_r）｛（Ｒ_r／Ｌ_r）Ｉ−ω_rＪ｝＝ａ_r12Ｉ＋ａ_i12Ｊ
Ａ₂₁＝（Ｍ／τ_r）Ｉ＝ａ_r21Ｉ
Ａ₂₂＝−（１／τ_r）Ｉ＋ω_rＪ＝ａ_r22Ｉ＋ａ_i22Ｊ
Ｂ₁＝１／（σＬ_s）Ｉ
Ｃ＝［Ｉ ０］
Ｉ＝［¹ ₀ ⁰ ₁］ Ｊ＝［⁰ ₁ ^-1 ₀］
Ｒ_s，Ｒ_r：一次及び二次抵抗
Ｌ_s，Ｌ_r：一次及び二次自己インダクタンス
Ｍ：相互インダクタンス
σ：漏れ係数（σ＝１−Ｍ²／（Ｌ_sＬ_r））
τ_r：二次時定数（τ_r＝Ｌ_r／Ｒ_r）
ω：電源角周波数
ω_r：電動機角速度（電気角換算）
式（１）より、状態オブザーバを式（２）のように構成する。

However,
^{_{i e s = [i e ds}} i e qs] T: primary current ^{_{φ e r = [φ e dr}} φ e qr] T: secondary magnetic flux ^{_{v e s = [v e ds}} v e qs] T: primary voltage x ^e = [i ^e _s φ ^e _r ] ^T
A ₁₁ = − {R _s / (σL _s ) + R _r (1−σ) / (σL _r )} I = a _r11 I
A ₁₂ = M / (σL _s L _r ) {(R _r / L _r ) I−ω _r J} = a _r12 I + a _i12 J
A ₂₁ = (M / τ _r ) I = _{ar 21} I
A ₂₂ = − (1 / τ _r ) I + ω _r J = a _r22 I + a _i22 J
B ₁ = 1 / (σL _s ) I
C = [I 0]
I = [ ¹ ₀ ⁰ ₁ ] J = [ ⁰ ₁ ^-1 ₀ ]
R _s , R _r : primary and secondary resistances L _s , L _r : primary and secondary self-inductance M: mutual inductance σ: leakage coefficient (σ = 1−M ² / (L _s L _r ))
τ _r : quadratic time constant (τ _r = L _r / R _r )
ω: Angular frequency of power supply ω _r : Angular speed of motor (electrical angle conversion)
From the equation (1), the state observer is configured as in the equation (2).

ただし、＾は状態変数及びパラメータの推定値を表す。本実施形態では、計算を簡略化するため、オブザーバゲイン行列Ｇ₁を零とする。

Here, ^ represents the estimated values of the state variables and parameters. In this embodiment, the observer gain matrix G _{1 is set} to zero in order to simplify the calculation.

<< 3 >>. First, the vector direction derivation of the current estimation error e ^e _iRs that depends only on the primary resistance error is shown below.

となる。一方、前述のとおりオブザーバゲイン行列Ｇ ₁ は零であるから、
ΔＡ＝Ａ^−Ａとすると、式（１）−式（２）の右辺は、

となる。ここで、式（１）−式（２）は式（３ａ）＝式（３ｂ）と書き替えることができるから、

が得られる。したがって、式（１）から式（２）を減算すると、式（３’）から次の式（３）が得られる。ただし、ｅ＝ｘ ^e −ｘ^ ^e とする。 When subtracting equation (2) from equation (1),
Since x ^e = [i ^e _s φ ^e _r ] ^{T as shown} in the proviso of equation (1) ,
The left side of Formula (1) -Formula (2) is

It becomes. On the other hand, since the observer gain matrix G ₁ is zero as described above ,
If ΔA = A ^ −A, then the right side of Equation (1) -Equation (2) is

It becomes. Here, since Formula (1) -Formula (2) can be rewritten as Formula (3a) = Formula (3b),

Is obtained. Therefore, when the equation (2) is subtracted from the equation (1), the following equation (3) is obtained from the equation (3 ′) . However, e = x ^e −x ^ ^e .

ｅは回転座標系で直流量であるので、定常状態においてde/dtは零となる。

Since e is a DC amount in the rotating coordinate system, de / dt becomes zero in a steady state .

Here, when a value other than the primary resistance is a true value, the steady-state error is de / dt = 0, so the left side of Equation (3) becomes 0, and Equation (4) is obtained.

なお、式（１）から次式

である。上式中のＡ ₁₁ ，Ａ ₁₂ ，Ａ ₂₁ ，Ａ ₂₂ のうち、式（１）ただし書の定義から、一次抵抗Ｒｓを含むのはＡ ₁₁ だけである。したがって、一次抵抗Ｒｓ以外が真値であるから、
ΔＡ _Rs （＝Ａ _Rs ^−Ａ _Rs ）は上記式（４）ただし書のようになる。
式（４）より、一次抵抗誤差のみを考慮した電流推定誤差ｅ^e _iRsのベクトル方向は式（５）となる。

From equation (1), the following equation

It is. Of A ₁₁ , A ₁₂ , A ₂₁ , and A _{22 in} the above formula, only A ₁₁ includes the primary resistance Rs from the definition of the proviso in formula (1) . Therefore, since the values other than the primary resistance Rs are true values,
ΔA _Rs (= A _Rs ^ −A _Rs ) is as shown in the above equation (4).
From Equation (4), the vector direction of the current estimation error e ^e _iRs considering only the primary resistance error is Equation (5).

同様に、回転速度以外が真値の場合、定常誤差は、de/dt＝０であるから式（３）の左辺が０になって、式（６）のようになる。

Similarly, when the value other than the rotation speed is a true value, the steady-state error is de / dt = 0, so the left side of Equation (3) becomes 0, and Equation (6) is obtained.

今度は、Ａ ₁₁ ，Ａ ₁₂ ，Ａ ₂₁ ，Ａ ₂₂ のうち、式（１）ただし書の定義から、回転速度ωｒを含むのはＡ ₁₂ ，Ａ ₂₂ だけである。したがって、回転速度ωｒ以外が真値であるから、
ΔＡ _ωｒ （＝Ａ _ωｒ ^−Ａ _ωｒ ）は上記式（６）ただし書のようになる。
式（６）より、回転速度推定誤差のみを考慮した電流推定誤差ｅ^e _iωrのベクトル方向は式（７）となる。

Now, among the _{A 11, A 12, A 21} , A 22, from the definition of the formula (1) proviso, it contains the rotational speed ωr is only A _12, A _22. Therefore, since values other than the rotational speed ωr are true values,
ΔA _ωr (= A _ωr ^ −A _ωr ) is expressed by the above equation (6).
From Equation (6), the vector direction of the current estimation error e ^e _iωr considering only the rotational speed estimation error is Equation (7).

これらのベクトル方向θ_Rs，θ_ωrの位相差が接近していなければ、一次抵抗誤差に依存しない速度推定が可能となる。ただし、次の（ｅ^e _ids，ｅ^e _iqs）は電流推定誤差ｅ^e _isの成分である。

If the phase differences between the vector directions θ _Rs and θ _ωr are not close, speed estimation independent of the primary resistance error is possible. However, the following (e ^e _ids, e ^e _iqs ) _is a component of the current estimation error e ^e _is .

式（８）は、電流推定誤差をθ_Rs回転させることを示す。以上の関係を図４に示す。式（１）〜（７）及び図４［１］は一次電流推定誤差ベクトル算出手段３１（図３）の動作に対応し、式（８）及び図４［２］は座標変換手段３２（図３）の動作に対応する。

Equation (8) indicates that the current estimation error is rotated by θ _Rs . The above relationship is shown in FIG. Expressions (1) to (7) and FIG. 4 [1] correspond to the operation of the primary current estimation error vector calculation means 31 (FIG. 3), and Expressions (8) and 4 [2] correspond to the coordinate conversion means 32 (FIG. This corresponds to the operation 3).

Thus, since e ^e _iδs is not affected by the primary resistance error, the speed is estimated by Equation (9) using only e ^e _iδs .

ただし、Ｋｐ，Ｋｉは定数とする。式（９）は、回転速度推定手段３３（図３）の動作に対応する。

However, Kp and Ki are constants. Equation (9) corresponds to the operation of the rotation speed estimation means 33 (FIG. 3).

  FIG. 5 is a graph showing a second embodiment of the speed sensorless vector control device according to the present invention. Hereinafter, description will be given based on this drawing. However, the description of the same part as the first embodiment is omitted.

In the present embodiment, the primary current estimation error vector calculation means calculates an angle θ _ωr formed by the primary current estimation error vector e _iωr caused by only the rotation speed estimation error and the d axis. The rotational speed estimating means, the angle between the primary current estimation error vector e _IRS and primary current estimation error vector _e iωr | θ _Rs -θ ωr | if is less than a predetermined value theta _th, [delta] axial component e _The rotational speed is estimated by adding the γ-axis direction component e _iγs to _iδs . FIG. 5 [1] shows a case where | θ _Rs −θ _ωr |> θ _th , and FIG. 5 [2] shows a case where | θ _Rs −θ _ωr | ≦ θ _th .

That is,
When | θ _Rs −θ _ωr |> θ _th , the rotational speed is estimated based only on e _iδs .
When | θ _Rs −θ _ωr | ≦ θ _th , the rotational speed is estimated based on K ₁ e _iδs + K ₂ e _iγs . However, K ₁ and K ₂ are constants.

At this time, as the angle | θ _Rs −θ _ωr | decreases, the ratio using the γ-axis direction component e _iγs may be increased. For example, K ₁ = | θ _Rs −θ _ωr | / θ _th , K ₂ = 1−K ₁ .

Primary current estimation error vector e _IS (not shown), so relies mainly on primary current estimation error vector e _Aiomegaaru, is close to the primary current estimation error vector _e iωr. Therefore, as shown in FIG. 5, the smaller the angle | θ _Rs −θ _ωr |, the larger the γ-axis direction component e _iγs and the smaller the δ-axis direction component e _iδs . Therefore, when the angle | θ _Rs −θ _ωr | is small, the accuracy decreases if only the small δ-axis direction component e _iδs is used. The rotational speed is estimated using the direction component e _iγs . At this time, if the ratio using the γ-axis direction component e _iγs is gradually increased as the angle | θ _Rs −θ _ωr | becomes smaller, the influence of fluctuations in the primary resistance is minimized.

  Next, simulation results of an example that further embodies the first embodiment will be described. The rating of the induction motor used is shown in Table 1 below.

The initial value of the speed command is set to 40 [rpm], a load of 7 [Nm] is applied, the speed command is ramped down by 0.5 [rpm / s] from 30 seconds later, and after 50 seconds, the speed command is 30 [rpm].

  Here, a simulation was performed by giving a + 30% error to the primary resistance value and using the conventional technique and the present embodiment. However, in the prior art, the rotational speed estimation formula shown in Non-Patent Document 3 is used. The primary resistance is not estimated.

  As shown in FIG. 6 [1], in the prior art, the speed estimation error increases as the speed decreases. On the other hand, as shown in FIG. 6 [2], in this embodiment, convergence is achieved without a speed error, and stable speed control can be performed even if the command speed is reduced.

Table 1. Induction motor rating Rated output 3.7 [kW]
Rated voltage 200 [V]
Rated current 15 [A]
Rated frequency 50 [Hz]
Rated rotational speed 1420 [rpm]
Rated torque 24.882 [Nm]
Number of poles 4
Stator resistance 0.332 [Ω]
Rotor resistance 0.234 [Ω]
Stator inductance 52.46 [mH]
Rotor inductance 52.46 [mH]
Mutual inductance 50.66 [mH]

It is a block diagram which shows 1st embodiment of the speed sensorless vector control apparatus which concerns on this invention. It is a block diagram which shows the adaptive secondary magnetic flux observer in the sensorless vector control apparatus of FIG. It is a block diagram which shows the speed adaptation mechanism in the adaptive secondary magnetic flux observer of FIG. 4 is a graph showing the operation of the speed adaptation mechanism of FIG. 3, FIG. 4 [1] corresponds to the operation of the primary current estimation error vector calculation means in FIG. 3, and FIG. 4 [2] is the operation of the coordinate conversion means in FIG. Corresponding to FIG. 5 is a graph showing a second embodiment of the speed sensorless vector control device according to the present invention, in which FIG. 5 [1] is | θ _Rs −θ _ωr |> θ _th and FIG. 5 [2] is | θ _Rs. This is the case when −θ _ωr | ≦ θ _th . It is a graph which shows the Example of this invention, FIG. 6 [1] is a case of a prior art, FIG. 6 [2] is a case of a present Example.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Speed sensorless vector control apparatus 11 Speed regulator 12 Vector rotator 13 Current regulator 14 Voltage type PWM inverter 15 Current sensor 20 Induction motor 21 Adaptive secondary magnetic flux observer 30 Speed adaptive mechanism 31 Primary current estimation error vector calculation means 32 Coordinate conversion Means 33 Rotational speed estimation means

Claims (7)

In a speed sensorless vector control method using an adaptive secondary magnetic flux observer for estimating the primary current and secondary magnetic flux of an induction motor,
In a Cartesian coordinate system comprising a d-axis and a q-axis, a primary current estimation error vector e _iRs caused only by a primary resistance setting error and an angle θ _Rs formed by the d-axis and a primary current estimation error caused by all errors Calculate the coordinates of the vector e _is (e _ids , e _iqs ) based on the primary voltage and primary current,
In the Cartesian coordinate system consisting of γ-axis and δ-axis obtained by rotating by each of the angle theta _Rs the d-axis and q-axis around its origin, the coordinate (e _ids, e _iqs) coordinates _(e iγs , E _iδs )
A rotational speed is estimated using the δ-axis direction component e _iδs .
A speed sensorless vector control method for an induction motor. Calculating an angle θ _ωr formed by a primary current estimation error vector e _iωr caused by only the rotational speed estimation error and the d axis;
When the difference between the angle θ _ωr and the angle θ _Rs is equal to or less than a certain _value , the rotational speed is estimated by adding the γ-axis direction component e _iγs to the δ-axis direction component e _iδs .
The speed sensorless vector control method of the induction motor according to claim 1. The ratio of using the γ-axis direction component e _iγs is increased as the difference between the angle θ _ωr and the angle θ _Rs decreases.
The speed sensorless vector control method of the induction motor according to claim 2. In a speed sensorless vector control method using an adaptive secondary magnetic flux observer for estimating the primary current and secondary magnetic flux of an induction motor,
A speed sensorless vector control method for an induction motor, wherein the rotational speed is estimated according to the following steps <1> to <9>.

<1> In the rotating coordinate system, the state equation of the induction motor when the state variables are the primary current and the secondary magnetic flux and the control input is the primary voltage is expressed as the following equation (1).

However,
^“e” indicates a rotating coordinate system.
_{^{i e s = [i e ds}} i e qs] T: primary current _{^{φ e r = [φ e dr}} φ e qr] T: secondary magnetic flux _{^{v e s = [v e ds}} v e qs] T: primary voltage x ^e = [i ^e _s φ ^e _r ] ^T
A ₁₁ = − {R _s / (σL _s ) + R _r (1−σ) / (σL _r )} I = a _r11 I
A ₁₂ = M / (σL _s L _r ) {(R _r / L _r ) I−ω _r J} = a _r12 I + a _i12 J
A ₂₁ = (M / τ _r ) I = _{ar 21} I
A ₂₂ = − (1 / τ _r ) I + ω _r J = a _r22 I + a _i22 J
B ₁ = 1 / (σL _s ) I
C = [I 0]
I = [ ¹ ₀ ⁰ ₁ ] J = [ ⁰ ₁ ^-1 ₀ ]
R _s , R _r : primary and secondary resistances L _s , L _r : primary and secondary self-inductance M: mutual inductance σ: leakage coefficient (σ = 1−M ² / (L _s L _r ))
τ _r : quadratic time constant (τ _r = L _r / R _r )
ω: Angular frequency of power supply ω _r : Angular speed of motor (electrical angle conversion)

<2>. From the equation (1), the state observer is configured as the following equation (2).

Here, ^ represents an estimated value of the state variable and parameter, and the observer gain matrix G ₁ is set to zero.

<3> By subtracting the equation (2) from the equation (1), the following equation (3) is obtained.

^{However, e = x e -x ^ e} , de / dt = 0, and ΔA = A ^ -A.

<4>. By setting a value other than the primary resistance to a true value, the following expression (4) is obtained from the expression (3).

<5>. The vector direction of the current estimation error e ^e _iRs considering only the primary resistance error is obtained from the above equation (4) as the following equation (5).

<6>. The following equation (6) is obtained from equation (3) by setting values other than the rotation speed to true values.

<7> The vector direction of the current estimation error e ^e _iωr considering only the rotational speed estimation error is obtained from the above equation (6) as the following equation (7).

<8>. If the phase difference between these vector directions θ _Rs and θ _ωr is equal to or greater than a certain value, e ^e _iδs is _obtained by the following equation (8). However, the following (e ^e _ids, e ^e _iqs ) _is a component of the current estimation error e ^e _is .

<9>. The rotational speed is estimated by the e ^e _iδs and the following (9).

However, Kp and Ki are constants. In a speed sensorless vector controller using an adaptive secondary magnetic flux observer that estimates the primary current and secondary magnetic flux of an induction motor,
The adaptive secondary magnetic flux observer is
In a Cartesian coordinate system comprising a d-axis and a q-axis, a primary current estimation error vector e _iRs caused only by a primary resistance setting error and an angle θ _Rs formed by the d-axis and a primary current estimation error caused by all errors Primary current estimation error vector calculation means for calculating the coordinates (e _ids , e _iqs ) of the vector e _is based on the primary voltage and the primary current;
In the Cartesian coordinate system consisting of γ-axis and δ-axis obtained by rotating by each of the angle theta _Rs the d-axis and q-axis around its origin, the coordinate (e _ids, e _iqs) coordinates _(e iγs , E _iδs ),
Rotation speed estimation means for estimating the rotation speed using the δ-axis direction component e _iδs ,
A speed sensorless vector control device for an induction motor. The primary current estimation error vector calculation means calculates an angle θ _ωr formed by a primary current estimation error vector e _iωr caused only by a rotation speed estimation error and the d axis,
When the difference between the angle θ _ωr and the angle θ _Rs is equal to or less than a certain _value , the rotational speed estimation means _{calculates the} rotational speed by adding the γ-axis direction component e _iγs to the δ-axis direction component e _iδs. presume,
The speed sensorless vector control device for an induction motor according to claim 5. The rotational speed estimation means increases the ratio of using the γ-axis direction component e _iγs as the difference between the angle θ _ωr and the angle θ _Rs decreases.
The speed sensorless vector control apparatus of the induction motor according to claim 6.

Images