JP6492320B2 - Digital rotor phase speed estimation device for AC motor - Google Patents

Description

The present invention provides an AC motor in which a rotor exhibits salient pole characteristics with respect to application of a high-frequency voltage having a frequency higher than the drive frequency (for example, a permanent magnet synchronous motor having a permanent magnet in the rotor, a wound synchronous motor, a synchronous reluctance motor, Uses a position speed sensor to detect the phase (synonymous with position) and speed of a rotor used in a drive controller for a hybrid field type synchronous motor or induction motor having a permanent magnet and field winding on the rotor. That is, the present invention relates to a digital rotor phase speed estimation device for performing estimation without using a sensor. In particular, the basic high frequency ωh of the applied high-frequency voltage is a digital rotation that has a critically high frequency that is about a fraction of the digital control frequency ωs of the stator current (ωs = 2π / Ts when the control period is Ts). The present invention relates to a child phase velocity estimation apparatus. In this specification, the unit of time such as a period is seconds (s), the unit of phase is radians (rad), and the unit of frequency (synonymous with angular frequency) is (rad / s). That is, the explanation is made on the assumption that the international unit (SI unit) is used. If the unit is self-explanatory, this description is omitted.

High-performance control of the AC motor can be achieved by a so-called vector control method. The vector control method requires information on the phase of the rotor or the speed that is a differential value thereof. Conventionally, a position speed sensor such as an encoder has been used to secure this. However, the use of this type of position / velocity sensor is not preferable in terms of reliability, axial volume, sensor cable routing, cost, and the like, so-called sensorless vector control method that does not require a position / velocity sensor. Research and development has been conducted for many years.

As a powerful sensorless vector control method, a high-frequency voltage having a frequency higher than the drive frequency is forcibly applied to the motor, and a corresponding high-frequency current (response high-frequency current) is processed to estimate the rotor phase (so-called high-frequency voltage) Various application methods have been developed and reported so far. In this specification, the constituent frequency components of the stator voltage and the stator current are divided into a low frequency component for driving (synonymous with driving frequency) and a high frequency component. The low-frequency component for driving is a component directly related to the rotational speed of the motor (that is, a frequency component comparable to the rotor speed (electrical speed)), and a high-frequency component (component of the basic high frequency ωh). Is a component of a frequency (frequency value ωh is known) much higher than the low frequency component for driving. In particular, in the present invention, the basic high frequency ωh is selected to be a critically high frequency which is about a fraction of the digital control frequency ωs of the stator current.

The rotor phase to be estimated may be determined at an arbitrary position of the rotor, but generally, either the negative salient pole phase or the positive salient pole phase of the rotor is selected as the rotor phase. As is well known to those skilled in the art, there is only an electrical phase deviation of ± π / 2 (rad) between the negative salient pole phase and the positive salient pole phase. The other phases are naturally found. Considering the above, hereinafter, the negative salient pole phase of the rotor will be referred to as the rotor phase unless otherwise specified. A biaxial orthogonal coordinate system composed of a d axis synchronized with the rotor phase and no phase deviation and a q axis orthogonal to the d axis is called a dq synchronous coordinate system (see FIG. 3).

In the high frequency voltage application method, estimated values of the rotor phase and speed are generated through processing of a high frequency current that is a response of the applied high frequency voltage. In general, the process of applying a high-frequency voltage is called “modulation”, and the process of processing a high-frequency current (a high-frequency component contained in the stator current) to generate a rotor phase estimation value is called “demodulation”.

Conventionally, as the fundamental high frequency of the high frequency voltage used for modulation, a frequency sufficiently higher than the low frequency component for driving and sufficiently lower than the control frequency has been employed (see Non-Patent Documents 1 and 2). A typical basic high frequency value employed is about 800π to 1000π (rad / s) (see Non-Patent Documents 1 and 2). From the viewpoint of separating the low frequency component for driving and the high frequency component included in the stator current, it is desirable that the frequency difference between the two components is as large as possible. However, as the fundamental high frequency of the applied high frequency voltage approaches the control frequency, the approximation accuracy by discrete time approximation of continuous time integration decreases, and as a result, π / 2 between the input and output of the continuous time integration element established by continuous time integration. The (rad / s) phase difference cannot be maintained by discrete time approximation. In the phase estimation problem, the phase difference error is a fatal defect. In order to avoid a fatal defect, that is, to maintain the discrete time approximation accuracy of the continuous time integration, a value sufficiently lower than the control frequency has been adopted as the basic high frequency of the high frequency voltage.

In recent years, efforts have been made to increase the basic high frequency of the high frequency voltage to the same level as the control frequency in order to improve the speed response of the phase estimation by the high frequency voltage application method. In the present specification, a high-frequency voltage application method that employs a modulation method as high as the control frequency is simply referred to as a “limit high-frequency voltage application method”. The present invention relates to a limit high-frequency voltage application method. As prior inventions of the limit high-frequency voltage application method, there are Patent Document 1 and Non-Patent Documents 3 to 5 listed in the Prior Art Document column. The main points of the prior invention will be described below with reference to the drawings of Patent Document 1.

FIG. 11A shows a drive control system for the AC motor shown in Patent Document 1. FIG. The “magnetic pole position estimation means 12” in FIG. 4 corresponds to the term “digital rotor phase speed estimation device” in this specification. FIG. 11B shows the detailed internal structure of the “magnetic pole position estimating means 12” in FIG. As shown in FIG. 11 (b), the stator current detected in discrete time is first subjected to discrete time difference processing (see processing in the vicinity of blocks 121 and 122). Subsequently, the current change amount calculation means 125 further performs discrete time difference processing on the discrete time difference processing value of the stator current. Thus, the technical feature of phase estimation in Patent Document 1 is that “phase estimation is performed through discrete time difference processing for the stator current”. The phase estimation according to other prior inventions dealing with the limit high-frequency voltage application method has similar characteristics (see Non-Patent Documents 3 to 5).

In the prior inventions such as Patent Document 1 and Non-Patent Documents 3 to 5, the problem of the phase difference error that is far from the previous integration process can be solved by the discrete time difference process that approximately realizes the continuous time differentiation process. On the other hand, it caused a new problem. The actual measurement signal includes noise to some extent. Discrete-time difference processing (approximate processing of continuous-time differentiation) of an actual measurement signal containing noise is extremely sensitive to noise, and the resulting phase estimation value causes noise-like pulsation. Consequently, the limit high-frequency voltage application method based on the discrete time difference processing cannot be applied to an AC motor that generates a stator current having a high noise level.

Daigo Kaneko, Yoshihisa Iwaji, Kiyoshi Sakamoto, Tsunehiro Endo: “AC Motor Control Device and AC Motor System”, Japanese Patent Laid-Open No. 2005-20918 (2303-6-27), Correction (2306-6-15) )

Shinnaka Shinji: “Vector control technology of permanent magnet synchronous motor, second volume (the essence of sensorless drive control)”, Denpa Shimbun (2308-12) Shinnaka Shinji: “Vector Control of Permanent Magnet Synchronous Motor, (Sensorless Vector Control Technology)”, Tokyo Denki University Press (2313-9) Ryozo Masaki, Satoru Kaneko, Yoshimi Sakurai, Mitsuyuki Motobu: “Position sensorless control system of IPM motor based on voltage superposition synchronized with carrier wave”, IEEJ Transactions, Vol. 122, No. 1, pp. 37-43 (2302-2) Y. D. Yoon, S .; K. Sul, S.M. Morimoto, and K.M. Ide: “High-Bandwidth Sensorless Algorithm for AC Machines Based on Square-Wave-Type Voltage Injection”, IEEE Trans Ind. Appl. , Vol. 47, no. 3, pp. 1361-1370 (2311-5 / 6) S. Kim, J. et al. I. Ha, and S.H. K. Sul: “PWM Switching Frequency Signal Injection Sensor Method in IPMSM”, IEEE Trans Ind. Appl. , Vol. 48, no. 5, pp. 1576-1587 (2312-9 / 10)

The present invention has been made under the above background, and its purpose is to perform phase velocity estimation without performing discrete-time difference processing basically in phase velocity estimation based on the limit high-frequency voltage application method. An object of the present invention is to provide a digital rotor phase speed estimation device that is robust to noise contained in a stator current while utilizing the speed response of phase estimation, which is the maximum advantage of the limit high-frequency voltage application method.

時刻ｔ＝ｋ＊Ｔｓに離散時間的に検出された固定子電流ｉ１，ｋに対して、あるいは固定子電流ｉ１，ｋから抽出された高周波電流ｉ１ｈ，ｋに対して、位相（（θｈ，ｋ−１）＋（θｉ））あるいはこの近似位相をもつ高周波信号を用いた処理を施して、あるいはこれと相当な処理を施して、高周波電流の振幅を抽出する高周波電流振幅抽出手段と、抽出された高周波電流振幅を少なくとも用いて、回転子位相の推定値あるいは回転子位相と基本的に微積分関係にある回転子速度の推定値の少なくとも何れかの推定値を、周期Ｔｓごとに生成する回転子位相速度推定値生成手段と、を備えることを特徴とする（連続時間と離散時間の時間的関係に関しては、図２を参照）。In order to achieve the above object, the invention of claim 1 is directed to a power converter (inverter) and a stator for an AC motor in which the rotor exhibits salient pole characteristics when a high frequency voltage having a frequency higher than the drive frequency is applied. A digital rotor phase speed estimation device used in a digital drive control device having a stator current digital control function with a period Ts (s) using a current detector, wherein k is an integer, and Nh is When a positive integer equal to or greater than 2, a high-frequency voltage v1h, k−1 having a period Th = Nh * Ts, a phase θh, k−1, and a basic high frequency ωh = ± 2π / Th (rad / s) is obtained at time t = (K−1) * Ts and t = k * Ts, and the high-frequency voltage applying means and the correction phase θi are expressed as follows:

The phase ((θh, k) with respect to the stator current i1, k detected discretely at time t = k * Ts or with respect to the high-frequency current i1h, k extracted from the stator current i1, k. -1) + (θi)) or high-frequency current amplitude extracting means for extracting the amplitude of the high-frequency current by performing processing using a high-frequency signal having this approximate phase or by performing processing equivalent thereto. A rotor that generates at least one of the estimated value of the rotor phase and the estimated value of the rotor speed that is basically in a calculus relationship with the rotor phase for each period Ts using at least the high-frequency current amplitude. And a phase velocity estimation value generation means (refer to FIG. 2 for the temporal relationship between continuous time and discrete time).

The invention according to claim 2 is the digital rotor phase speed estimation device according to claim 1, and is composed of a γ-axis aimed at synchronization with a rotor phase with zero phase deviation and a δ-axis orthogonal thereto. The high-frequency voltage applying means is configured to apply a high-frequency voltage on the γδ quasi-synchronous coordinate system.

The invention according to claim 3 is the digital rotor phase speed estimation device according to claim 1, comprising a γ-axis aimed at synchronization with a rotor phase with zero phase deviation and a δ-axis orthogonal thereto. The high-frequency current amplitude extracting means is configured to process a stator current or a high-frequency current on the γδ quasi-synchronous coordinate system.

According to a fourth aspect of the present invention, there is provided the digital rotor phase speed estimation device according to the first aspect, wherein a phase locked loop or a feedback loop corresponding to the phase locked loop is configured, and the rotor phase speed estimated value is set. The generation means is configured.

The description of the effects of the present invention described below is a permanent magnet synchronous motor having a permanent magnet in the rotor if the rotor is an AC motor having salient pole characteristics with respect to application of a high frequency voltage higher than the drive frequency, The present invention is applicable to any AC motor such as a wound linear synchronous motor, a synchronous reluctance motor, a hybrid field type synchronous motor, and an induction motor. Embedded magnet type permanent magnet synchronous motors, synchronous reluctance motors, and the like exhibit salient pole characteristics with respect to driving voltage and current. These electric motors similarly exhibit salient pole characteristics even when a high frequency voltage is applied. On the other hand, a surface magnet type permanent magnet synchronous motor and an induction motor that do not show salient pole characteristics with respect to the driving voltage / current show salient pole characteristics when a high frequency voltage is applied. The hybrid field type synchronous motor has characteristics of a permanent magnet type and a wound type synchronous motor, and can exhibit salient pole characteristics with respect to application of a high frequency voltage. In particular, the self-excited hybrid field synchronous motor has a strong saliency.

As a first preparation for explaining the effects of the invention, consider the discrete time element of FIG. This element is a discrete-time integration element including a zero-order holder and a sampler before and after the continuous-time integration element 1 / s. It is assumed that the zero-order holder and the sampler in the discrete time integration element operate in synchronization with the period Ts (see FIG. 1B). In the figure, a discrete time signal at a synchronized discrete time t = k * Ts is simply expressed as uk and yk using a leg k. Hereinafter, the discrete time “t = k * Ts” is simply expressed as the time “k”. As clearly shown in FIG. 1B, the discrete time input uk at time k is maintained from time k to (k + 1) by the zero-order hold circuit and becomes continuous time input u. The discrete time output yk is a discrete time signal sampled by the sampler in synchronization with the discrete time input uk. The two discrete time signals uk and yk are both discrete time signals at the same k time (that is, at the synchronization time). The input / output signals u and y of the continuous time integration element 1 / s are naturally continuous time signals.

離散時間領域の信号に作用する遅れ演算子をｚ−１とするならば、（２）式の関係は次の（３）式のように表現することもできる。なお、「ｚの逆数」は、離散時間信号に対して１制御周期Ｔｓの時間遅れを発生させる「遅延器」（演算的には、「遅れ演算子」）を意味する。本明細書では、他所でも、同様の意味で「ｚの逆数」を利用する。

FIG. 1B illustrates related signals (uk, u, y, yk) when the discrete-time input signal uk at time k is a single unit pulse function (delta function). In this example, the z-transform of the discrete-time output signal yk gives the transfer function Gi0 (z-1) of the discrete-time integral element (z-1 is the inverse of z). This is the following equation (2).

If the delay operator acting on the signal in the discrete time domain is z−1, the relationship of equation (2) can also be expressed as the following equation (3). “Reciprocal of z” means a “delay device” (in terms of operation, “delay operator”) that generates a time delay of one control period Ts with respect to a discrete-time signal. In the present specification, the “reciprocal of z” is used in the same way in other places.

（５）式に利用したω−（すなわち、ωバー）は、元来の周波数ω（単位はｒａｄ／ｓ）と離散時間周期Ｔｓ（単位はｓ、秒）とを用いた次の（６）式で定義された正規化周波数であり、その単位はｒａｄである。

正規化周波数は、当然のことながら、元来の周波数ωの正負の極性に応じて、正負の極性を持つ。正規化周波数を用いて記述された（５ｃ）式が、本発明で重要な役割を演ずる「補正位相」を示すものとなる（（１）式参照）。The following two points "(1) The above equation (3) showing the characteristics of the discrete time integration element does not contain any approximate elements", "(2) Discrete time integration under consideration in the present invention" Special attention should be paid to “the element is not an approximation of a continuous-time integral element”. In the discrete-time integration element of FIG. 1, the output signal at time k is dominated by the input signal before time (k−1), as clearly shown in equations (2) and (3). Considering this point, consider the transfer function Gi1 (z-1) between uk-1 and yk, and further consider this frequency response. These are the following equations (4) and (5), respectively.

The ω− (that is, ω bar) used in the equation (5) is the following (6) using the original frequency ω (unit: rad / s) and the discrete time period Ts (unit: s, second). The normalized frequency defined by the equation, the unit of which is rad.

Naturally, the normalized frequency has positive and negative polarities according to the positive and negative polarities of the original frequency ω. The expression (5c) described by using the normalized frequency indicates the “correction phase” that plays an important role in the present invention (see the expression (1)).

As a second preparation for explaining the effects of the invention, a signal obtained by processing a discrete-time two-phase signal with a discrete-time integration element is analyzed. Consider the following discrete-time two-phase signal constructed using a positive integer Nh of 2 or more.

The amplitude of the discrete-time two-phase signal in equation (7a) is 1. Further, the phase θk at time k varies discretely and has a cycle NhTs (see equation (7b)). Therefore, the basic high frequency ωh and the normalized basic high frequency ωh− (that is, ωh bar) of the discrete-time two-phase signal are respectively expressed by the following equations.

（９）式における振幅Ａｉ、位相θｉは、（５ｂ）、（５ｃ）式の通りである。また、位相θｉの関数である２行２列（以下、２×２と略記）行列Ｒは、次の（１０）式で定義されたベクトル回転器である。

（９）式は、「２×２離散時間積分要素のｋ時点の出力である離散時間二相信号ｙｋは、入力の離散時間二相信号と同様に平均速度ωｈで空間的に回転するが、（ｋ−１）時点の入力の離散時間二相信号ｕｋ−１に対し、振幅はＡｉ倍となり、空間的に位相θｉをもつ」ことを意味している。When the discrete-time two-phase signal of equation (7) is regarded as a 2 × 1 (hereinafter abbreviated as 2 × 1) vector in a two-dimensional space, the vector rotates spatially at an average speed ωh. Two discrete-time integration elements in FIG. 1A are arranged in parallel to form a 2-input 2-output (hereinafter abbreviated as 2 × 2) discrete-time integration element. A discrete-time two-phase signal of Equation (7) is input to the 2 × 2 element, and the spatial behavior of the continuous-time two-phase signal and the discrete-time two-phase signal that are the outputs is analyzed. An analytical solution of the output discrete-time two-phase signal is obtained as follows by applying the result of equation (5).

The amplitude Ai and phase θi in equation (9) are as in equations (5b) and (5c). A 2 × 2 (hereinafter abbreviated as 2 × 2) matrix R that is a function of the phase θi is a vector rotator defined by the following equation (10).

The expression (9) indicates that “the discrete-time two-phase signal yk, which is the output at the time point k of the 2 × 2 discrete-time integration element, spatially rotates at an average speed ωh, similarly to the input discrete-time two-phase signal. This means that the amplitude is Ai times that of the discrete-time two-phase signal uk-1 that is input at the time point (k-1) and has a phase θi in space.

As a third preparation for explaining the effects of the invention, a biaxial orthogonal coordinate system is defined. Consider a γδ general coordinate system that rotates at a coordinate system speed ωγ specified by the control designer, as shown in FIG. However, the maximum speed of the coordinate system speed ωγ is at most about the maximum speed of the rotor speed. When the coordinate system speed ωγ is zero and the phase θγ is the phase θα, the γδ general coordinate system is an αβ fixed coordinate system. When the coordinate system speed ωγ is the rotor speed true value ω2n and the phase θγ is zero, the γδ general coordinate system is a dq synchronous coordinate system. In addition, when the coordinate system speed ωγ is an estimated value of the rotor speed and the phase θγ is aimed at zero convergence, the γδ general coordinate system is γδ quasi-synchronous aiming at convergence with zero phase difference to the dq synchronous coordinate system. Coordinate system.

The rotation from the main axis (γ axis) to the sub axis (δ axis) is defined as the positive direction. Unless otherwise specified, the 2 × 1 vector signals expressing the physical quantities of the AC motor to be treated below are all defined on the γδ general coordinate system. In the following mathematical expression, the 2 × 1 vector signal is expressed using bold characters.

The effect of the invention of claim 1 will be described below based on the first to third preparations of the invention effect explanation. Consider applying a high-frequency voltage for phase estimation superimposed on a voltage for driving an electric motor. In this case, the stator voltage v1, current i1, and linkage flux φ1 can be expressed as a composite vector of two components as follows.

The leg marks f and h of the signal on the right side of the equation (11) indicate a low frequency component and a high frequency component for driving, respectively. In particular, the three signals v1h, i1h, and φ1h, which are the second terms of each of the three formulas (11), are closely related to the present invention, and each of the applied high-frequency voltage, the high-frequency current as a response, and the applied high-frequency voltage. The high frequency magnetic flux resulting from is shown. Note that the basic high frequency ωh of the high-frequency voltage superimposed and applied for phase estimation is sufficiently high to satisfy the relationship of the following equation (12).

以降では、記号ｓは、これが連続時間信号に作用する場合には微分演算子ｄ／ｄｔを、Ｉは２×２単位行列を、Ｊは次式で定義された２×２交代行列を意味するものとする。

In the following, the symbol s denotes the differential operator d / dt if it acts on a continuous time signal, I denotes a 2 × 2 identity matrix, and J denotes a 2 × 2 alternating matrix defined by Shall.

When the equation (12) is established, the relationship of the following equations (14) and (15) is established for the high frequency component of the stator in the AC motor in which the rotor exhibits salient pole characteristics with respect to the application of the high frequency voltage. To do.

また、Ｌｉ、Ｌｍは固定子の同相インダクタンス、鏡相インダクタンスであり、ｄ軸、ｑ軸インダクタンスとは次の関係を有する。

なお、鏡相インダクタンスＬｍは、回転子位相として負突極位相を選定する場合には負となる。Here, the 2 × 2 matrix Q is defined as follows.

Li and Lm are the in-phase inductance and mirror phase inductance of the stator, and have the following relationship with the d-axis and q-axis inductances.

The mirror phase inductance Lm is negative when a negative salient pole phase is selected as the rotor phase.

（１５）式は、高周波磁束の応答たる高周波電流として、次の（１９ａ）式のように書き改められ、特に、（１２）式が成立している状況下では、（１９ｂ）式のように積分近似される。

ただし、

The expression (14) is rewritten as the following expression (18a) as a high-frequency magnetic flux as a response of the applied high-frequency voltage, and particularly in the situation where the expression (12) holds, the expression (18b) Integral approximation to

Equation (15) is rewritten as the following equation (19a) as a high-frequency current as a response of the high-frequency magnetic flux, and particularly in the situation where equation (12) is established, equation (19b) Integral approximation.

However,

高周波周期Ｔｈと離散時間周期Ｔｓに関しては、２以上の正の整数Ｎｈを用いた次の関係が成立している。

（２２）式の高周波周期Ｔｈの採用は、発明効果説明第２の準備の中で説明した（７ｂ）式、（８）式の採用を意味する。「電圧の脚符（ｋ−１）は時間ｔ＝（ｋ−１）＊Ｔｓ〜ｋ＊Ｔｓの間に、零次ホール回路としてモデル化されるＰＷＭ機能を備えた電力変換器を介して印加される電圧を意味する」ことを、重ねて注意しておく（図１、図２参照）。A high frequency voltage applying means which is the first means of the invention of claim 1 will be described. In the first aspect of the invention, when k is an integer and Nh is a positive integer of 2 or more, a high frequency voltage having a period Th = Nh * Ts, a phase θh, k−1, and a basic high frequency ωh = ± 2π / Th v1h, k-1 is applied between time t = (k-1) * Ts and t = k * Ts. Application between time t = (k−1) * Ts and t = k * Ts is performed via a power converter having a PWM function modeled as a zero-order Hall circuit. FIG. 2 shows an example of the applied high-frequency voltage on the γδ general coordinate system and the high-frequency current that is the response. In addition, the relationship between continuous time and discrete time is shown. As is clear from the comparison between the signal application / response example of FIG. 2 and the signal application / response example of FIG. 1, the power converter having the PWM function is modeled as a zero-order Hall circuit. A typical one of the above-described applied high-frequency voltages v1h and k-1 is the following constant elliptical high-frequency voltage.

Regarding the high frequency period Th and the discrete time period Ts, the following relationship is established using a positive integer Nh of 2 or more.

The adoption of the high frequency cycle Th in the equation (22) means the adoption of the equations (7b) and (8) described in the second preparation of the invention effect description. “Voltage mark (k−1) is applied during time t = (k−1) * Ts to k * Ts via a power converter with a PWM function modeled as a zero order Hall circuit. Note that “means the voltage to be applied” (see FIGS. 1 and 2).

The high-frequency voltage applying means provided in the digital rotor phase speed estimation apparatus of the present invention includes means for generating a voltage command value for applying a high-frequency voltage represented by equation (21) having the property of equation (22). (See FIGS. 4, 5, and 9 below). The high-frequency voltage is applied to the electric motor through the power converter according to the voltage command value (see FIGS. 4 and 9 described later).

Next, a high-frequency current amplitude extracting means which is the second means of the invention of claim 1 will be described. As a matter of course, the AC motor rotates in both positive and negative directions. Further, the basic high frequency ωh of the high-frequency voltage can also have positive and negative frequencies. In the description of the high-frequency current amplitude extracting means, it is assumed that the motor rotates in the positive direction including zero speed in order to ensure simplicity. Accordingly, the speed ωγ of the γδ quasi-synchronous coordinate system is also made non-negative. The basic high frequency ωh of the high frequency voltage is also positive. This premise is for distinguishing the high-frequency magnetic flux caused by the applied high-frequency voltage, and the positive-phase and reverse-phase components of the high-frequency current. When the direction of rotation or polarity of frequency is reversed, forward / reverse phase inversion may occur. This premise is to avoid descriptive confusion caused by forward / reverse phase inversion, so that the generality of the discussion is not lost.

上式では、脚符ｐ、ｎがおのおの正相、逆相を意味している。In accordance with the above assumptions, for the purpose of simplifying the description, the positive-phase and negative-phase unit vectors are defined as follows.

In the above formula, the foot marks p and n mean the normal phase and the reverse phase, respectively.

（２１ａ）式の一定楕円形高周波電圧は、上の正相、逆相の単位ベクトを用い、以下のように表現することができる。

The following relationship is established between the normal phase unit vector and the negative phase unit vector.

The constant elliptical high-frequency voltage of the equation (21a) can be expressed as follows using the above unit vectors of the positive phase and the negative phase.

（２６）式における振幅は次の（２７）、（２８）式のように定義されている。

(1) The discrete-time high-frequency voltage v1h, k-1 of the equation (21) is applied through a power converter modeled as a zero-order holder in FIG. (2) On the other hand, the applied discrete-time high-frequency voltage v1h, k-1 in equation (21) is analytically rewritten as in equation (25a). Based on the facts of (1) and (2), applying the equation (9), which is the characteristic equation of the discrete-time integration element accompanied by the zero-order holder and the sampler, to the continuous time integration relationship of the equation (19b) , (K−1) time discrete time high frequency current i1h, k corresponding to discrete time high frequency voltage v1h, k−1 in equation (21), positive phase component ihp, k and negative phase component ihn, k The following expression is obtained that is analyzed as the sum of.

The amplitude in the equation (26) is defined as the following equations (27) and (28).

（２９）式における振幅は以下のように定義されている。

(1) The discrete-time high-frequency voltage v1h, k-1 of the equation (21) is applied through a power converter modeled as a zero-order holder in FIG. (2) On the other hand, the applied discrete-time high-frequency voltage v1h, k-1 in the equation (21) is analytically rewritten as in the equation (25b). Based on the facts of (1) and (2), applying the formula (9), which is the characteristic formula of the discrete time integration element accompanied by the zero-order holder and the sampler, to the relationship of the continuous time integration of the formula (19b) The following analytical expression is obtained as the discrete-time high-frequency current i1h, k at time k corresponding to the discrete-time high-frequency voltage v1h, k-1 at time (k-1) in equation (21).

The amplitude in the equation (29) is defined as follows.

“The symbol (k−1) indicating the discrete time of the voltage means the voltage applied during the continuous time t = (k−1) * Ts to k * Ts, whereas the symbol k of the current. Note again that “means the discrete time detection time at t = k * Ts” (see FIGS. 1 and 2).

The role of the high frequency current amplitude extracting means is to extract a signal (cp, sp, cn, sn) or (cγ, sγ, cδ, sδ) corresponding to the equation (28) or (30), which means the amplitude of the high frequency current. It is to be. The amplitude of the high-frequency current is the amplitude of the high-frequency current having a phase ((θh, k−1) + (θi)), as clearly shown in the equation (26) or (29). In order to extract the amplitude of the high-frequency current, the phase information of the high-frequency current is essential. According to the first aspect of the present invention, the high-frequency current amplitude extracting means uses a high-frequency signal having the same phase ((θh, k−1) + (θi)) as the high-frequency current. Therefore, the amplitude of the high frequency current can be appropriately extracted. A specific extraction method will be described in detail later using an embodiment (see FIGS. 6 and 7).

The rotor phase speed estimated value generating means, which is the third means of the invention of claim 1, will be described. The amplitudes (cp, sp, cn, sn) and (cγ, sγ, cδ, sδ) of the high-frequency current have a phase θγ as shown in the equations (28) and (30). By processing this, it is possible to obtain a phase θγ or a positive correlation signal (a signal having a positive correlation with the phase θγ) corresponding to the phase θγ. These relationships are on the γδ general coordinate system. Therefore, this relationship can also be applied to signals on the αβ fixed coordinate system, which is a special case of the γδ general coordinate system. More specifically, if the condition of the αβ fixed coordinate system is applied to the invention of claim 1, an estimated value of the rotor phase θα viewed from the α axis of the αβ fixed coordinate system can be obtained (see FIG. 3). ). This relational expression can also be applied to signals on the γδ quasi-synchronous coordinate system, which is a special case of the γδ general coordinate system. More specifically, if the condition of the γδ quasi-synchronous coordinate system is applied to the invention of claim 1, an estimated value of the rotor phase θγ viewed from the γ-axis of the γδ quasi-synchronous coordinate system can be obtained (see FIG. 3). Note that there is the following mutual conversion relationship between the two types of amplitudes (cp, sp, cn, sn) and (cγ, sγ, cδ, sδ) of the high-frequency current, and one can be calculated from the other. .

If the rotor phase can be generated, the rotor speed can be easily generated by processing the generated rotor phase by the integral feedback speed estimation method and the generalized integral PLL method. Details of this are well known to those skilled in the art through Non-Patent Document 1, Non-Patent Document 2, and the like, and will be omitted.

As is apparent from the above description, according to the invention of claim 1, in the phase velocity estimation based on the limit high frequency voltage application method, the phase velocity estimation is basically performed without performing the discrete time difference processing, and as a result The phase and speed can be estimated in a form that is robust to noise contained in the stator current while taking advantage of the speed response of phase estimation, which is the greatest feature of the limit high-frequency voltage application method (speed estimation value generation) Specific examples of these are shown through examples using FIGS. 5, 8, and 10 described later).

Then, the effect by the invention of Claim 2 is demonstrated. As clearly shown in the equation (11), the stator voltage includes two components, a low frequency component for driving and a high frequency component. In order to obtain a good estimation value, it is desirable to make the difference between the frequencies of both components as large as possible. In the invention of claim 2, a high frequency voltage is applied on the γδ quasi-synchronous coordinate system. On the other hand, on the γδ quasi-synchronous coordinate system, the low frequency of the driving component is substantially zero. Therefore, if the high-frequency voltage applying means is configured according to the invention of claim 2, it can be said that the fundamental high frequency of the applied high-frequency voltage can be made the difference between the frequencies of both components, and that the maximum frequency difference can be obtained. An effect is obtained. As a result, the effect of claim 1 can be further enhanced.

Then, the effect by the invention of Claim 3 is demonstrated. As clearly shown in the equation (11), the stator current includes two components, a low frequency component for driving and a high frequency component. It is the high frequency component of the two components, that is, the high frequency current, that holds the rotor phase information. In the present invention, the low frequency component for driving included in the stator current is separated through the low-pass filter processing having the low-pass characteristic instead of performing the discrete-time difference processing (see FIG. 6 described later). (See the embodiment using FIG. 7), (discrete time integration processing also has a low-pass characteristic). According to the high frequency current amplitude extracting means based on the invention of claim 3, the processing of the stator current is performed on the γδ quasi-synchronous coordinate system. On the γδ quasi-synchronous coordinate system, even when the motor rotates at high speed, the frequency of the low frequency component for driving the stator current is substantially zero, and the high frequency current amplitude extracting means according to the invention of claim 3 In this case, the driving low-frequency component is separated in an ideal state. As is apparent from the above description, according to the high frequency current amplitude extracting means of the invention of claim 3, the driving low frequency component is separated from the stator current in an ideal state in a wide speed range from zero speed to high speed. As a result, the effect that the invention of claim 1 can be applied in a wide speed range from zero speed to high speed is obtained. In other words, the effect that the effect of claim 1 can be further enhanced is obtained.

Then, the effect by the invention of Claim 4 is demonstrated. As shown by the 2 × 2 matrix Q in the equation (16), the salient pole characteristic has an ambiguity of ± π from the viewpoint of the rotor phase. In rotor phase estimation, this ambiguity must always be dealt with and eliminated. A phase-locked loop (PLL) or a feedback loop equivalent to a phase-locked loop serves to follow the phase estimate to a phase true value that changes from time to time. According to the fourth aspect of the present invention, the rotor phase speed estimation value generating means is configured so as to configure the phase locked loop or a feedback loop equivalent to the phase locked loop. An effect is obtained that the phase estimation value can be obtained in the excluded form. Furthermore, the effect of obtaining the speed estimated value in a form corresponding to the rotor phase estimated value is obtained. In other words, the effect that the effect of claim 1 can be further enhanced is obtained.

  "Figures showing the relationship between continuous-time and discrete-time integral elements and examples of temporal relationships between continuous-time and discrete-time"   "Figure showing examples of temporal relationships between applied discrete-time high-frequency voltage, applied continuous-time high-frequency voltage, response continuous-time high-frequency current, and response discrete-time high-frequency current"   "Figure showing an example of the relationship between the three coordinate systems and the rotor phase"   "Block diagram showing drive control system for one embodiment"   “Block diagram showing basic configuration of phase velocity estimator in one embodiment”   "Block diagram showing basic configuration of amplitude extractor in one embodiment"   "Block diagram showing basic configuration of amplitude extractor in one embodiment"   "Block diagram showing basic configuration of estimated value generator in one embodiment"   "Block diagram showing drive control system for one embodiment"   "Block diagram showing basic configuration of estimated value generator in one embodiment"   “Block diagram showing a configuration example of a conventional digital drive control device”

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. FIG. 4 shows an example of a digital drive control device provided with the digital rotor phase speed estimation device of the present invention for a permanent magnet synchronous motor which is a typical AC motor. 1 is an AC motor (synchronous motor), 2 is a power converter (inverter), 3 is a discrete time current detector, 4a and 4b are 3 phase 2 phase converters, 2 phase 3 phase converters and 5a, respectively. 5b is a vector rotator, 6 is a current controller, 7 is a command converter, 8 is a speed controller, 9 is a high-frequency component removal filter (digital filter for removing high-frequency components), 10 Denotes a phase velocity estimator (another name for a digital rotor phase velocity estimator), 11 denotes a coefficient unit, and 12 denotes a cosine sine signal generator. In FIG. 4, various devices from 2 to 12 except for one electric motor constitute a digital drive control device. In this figure, a 2 × 1 vector signal is represented by one thick signal line to ensure simplicity. The following block diagram expression follows this.

The three-phase stator current detected by the discrete-time current detector 3 is converted into a two-phase current on the αβ fixed coordinate system by the three-phase two-phase converter 4a, and then the rotor phase ( It is converted into a two-phase current in the γδ quasi-synchronous coordinate system aiming for phase synchronization with zero phase deviation to the same phase as in the dq synchronous coordinate system (see FIG. 3). A high frequency component contained in the stator current is removed from the converted current through the high frequency component removal filter 9 (in other words, a low frequency component for driving is extracted), and this is sent to the current controller 6. The current controller 6 indicates that the two-phase driving current on the γδ quasi-synchronous coordinate system is the current command value of each phase (the initial * of the current command value means the command value. The command value of the signal is expressed by adding a prefix * to the related signal) to generate a driving two-phase voltage command value on the γδ quasi-synchronous coordinate system. Here, the command value of the high-frequency voltage received from the phase velocity estimator 10 is superimposed on the driving two-phase voltage command value, and the superimposed two-phase voltage command value is sent to the vector rotator 5b. In 5b, the voltage command value for superposition and synthesis on the γδ quasi-synchronous coordinate system is converted into a two-phase voltage command value in the αβ fixed coordinate system, and sent to the two-phase three-phase converter 4b. In 4 b, the two-phase voltage command value is converted into a three-phase voltage command value and output as the final voltage command value to the power converter 2. The power converter 2 generates a voltage corresponding to the final voltage command value, applies it to the AC motor (synchronous motor) 1 and drives it.

The phase speed estimator 10 receives as input the stator current on the γδ quasi-synchronous coordinate system that is the output signal of the vector rotator 5a, and receives the rotor phase estimated value θα ^, the rotor (electrical) speed estimated value ω2n ^, And a high-frequency voltage command value v1h * are output. The rotor phase estimation value is converted into a cosine / sine signal by the cosine sine signal generator 12, and then passed to the vector rotators 5a and 5b that determine the γδ quasi-synchronous coordinate system. This means that the rotor phase estimation value is the phase of the γδ quasi-synchronous coordinate system (equivalent to the phase of the γ axis).

As is well known to those skilled in the art, the two-phase current command value on the γδ quasi-synchronous coordinate system is obtained by converting the torque command value through the command converter 7. In the speed controller 8, the rotor speed estimated value (rotor electrical speed estimated value), which is one of the output signals from the phase speed estimator 10, is a constant value, and the coefficient unit 11 And then converted to a machine speed estimate and sent. Since the example of FIG. 4 shows an example in which the speed control system is configured, a torque command value is obtained as the output of the speed controller 8. As is well known to those skilled in the art, the speed controller 8 is unnecessary when the control purpose is torque control and the speed control system is not configured. In this case, the torque command value is directly applied from the outside.

The core of the present invention resides in a phase speed estimator 10 that is synonymous with a digital rotor phase speed estimation device. In any of the speed control and the torque control, the phase speed estimator 10 does not require any change. Further, even when the driven motor is another synchronous motor or induction motor, the phase speed estimator 10 does not require any change. In the following, various embodiments of the phase speed estimator 10 will be described without losing generality with respect to control modes such as speed control and torque control, and without losing generality with respect to the AC motor to be driven. To do.

As is apparent from the above description, the digital rotor phase speed estimation apparatus of FIG. 4 is an example in which the phase speed estimator 10 which is the main object of the present invention is configured on the γδ quasi-synchronous coordinate system. This means that “the high-frequency voltage applying means and the high-frequency current amplitude extracting means are configured on the γδ quasi-synchronous coordinate system based on the inventions of claims 2 and 3 (details will be described later). ).

FIG. 5 shows a typical internal configuration example of the phase velocity estimator 10. The phase speed estimator includes a high frequency voltage command device 10-1 that realizes a high frequency voltage application means, an amplitude extractor 10-2 that realizes a high frequency current amplitude extraction means, and an estimated value generation that realizes a rotor phase speed estimated value generation means. It is comprised from the container 10-3. FIG. 5 shows a state of the control period t = k * Ts to (k + 1) * Ts (see FIG. 2).

The high frequency voltage command device 10-1 generates a high frequency voltage command value and outputs it to the outside of the phase velocity estimator 10. The high-frequency voltage commander 10-1 in the control period t = k * Ts to (k + 1) * Ts is applied to the motor via the power converter in the next control period t = (k + 1) * Ts to (k + 2) * Ts. The synthesized high frequency voltage command value v1h * is synthesized. The combined high-frequency voltage command value v1h * is, for example, a constant elliptical high-frequency voltage type of equation (21). When the power converter operates normally, the basic high frequency component of the applied high frequency voltage is the same as the high frequency voltage command value. A high-frequency voltage phase that is different for each control period is output to the amplitude extractor. In this example, the high-frequency voltage phase that is different for each control period is output in the form of a cosine sine value represented by the following equation.

The amplitude extractor 10-2 is different for each control period from the stator currents i1 and k detected for discrete time by the discrete time current detector for each control period Ts or the high frequency current i1h and k extracted from the stator current. Using the high frequency voltage phase (θi, k−1), the amplitude of the high frequency current i1h, k is extracted and output. The extracted high-frequency current amplitude is, for example, (cp, sp, cn, sn) or (cγ, sγ, cδ, sδ) or a part thereof. In FIG. 5, (cp, sp, cn, sn) is mainly shown as the amplitude of the high-frequency current i1h, k, and (cγ, sγ, cδ, sδ) is shown in parentheses.

The estimated value generator 10-3 generates and outputs the rotor phase and the rotor speed in a discrete time every period Ts by using all or part of the amplitude of the high-frequency current i1h, k.

FIG. 6 shows an embodiment of the amplitude extractor 10-2. The embodiment of FIG. 6 is based on the equation (26) that analytically shows that the high-frequency current contains a normal phase component and a reverse phase component. The amplitude extraction principle based on the equation (26) is as follows.

The symbol <·> in the equation (33) means processing by the digital low-pass filter F1 (z-1) that shows zero attenuation at the normalized frequency zero and sufficient attenuation at the normalized frequency (2ωh−). The 2 × 2 matrix R is a vector rotator defined by equation (10). When the digital low-pass filter F1 (z-1) exhibits sufficient attenuation characteristics at the normalized frequency (ωh−) in addition to the normalized frequency (2ωh−), the stator current containing the high-frequency current is reduced. Even for this, the above equation (33) is established. In consideration of this point, the input of the phase velocity estimator in FIG. 5 and, in turn, the input of the amplitude extractor in FIG. 6 is the stator current itself, and is a high-frequency current as a secondary (expressed in parentheses). A block F1 (z-1) in FIG. 6 means a digital low-pass filter having the above characteristics. When the input of the phase velocity estimator is a high frequency current, a high frequency current extraction block is disposed at the input end of the phase velocity estimator.

If the control period Ts and the high-frequency voltage period Th are determined, the correction phase θi is constant as shown in the equation (1). In consideration of the uniformity of the correction phase, FIG. 6 shows a configuration example in which the constant correction phase and the phases θh and k−1 that change with each control period are separated. In other words, FIG. 6 is an example faithfully based on the third expression of the expressions (33a) and (33b). You may comprise according to 2nd Formula of (33a) and (33b) Formula, and you may comprise according to 1st Formula (namely, left side). In particular, in the case of following the first expression of the expressions (33a) and (33b), an embodiment in which the cosine sine value of the phase ((θh, k−1) + (θi)) is directly used for the vector rotator and Become. The definition of the vector rotator is as shown in equation (10).

When extracting some of the four amplitudes (cp, sp, cn, sn) of the high-frequency current, processing equivalent to the processing by the vector rotator can be performed in a simplified manner. .

FIG. 7 shows an embodiment of the amplitude extractor 10-2. The embodiment of FIG. 7 is based on the equation (29) that analyzes that the amplitude of the γ-axis element and the δ-axis element of the high-frequency current has phase information. The amplitude extraction principle based on the equation (29) when Nh is greater than 3 is expressed by the following equation.

The meaning of the symbol <•> in the equation (34) is the same as that in the equation (33). The input of the phase velocity estimator in FIG. 5 and, in turn, the input of the amplitude extractor in FIG. 7 is the stator current itself, and it is a high-frequency current as a secondary (expressed in parentheses). (Refer to the explanation immediately after the equation (33)).

当然のことながら、位相（（θｈ，ｋ−１）＋（θｉ））をもつ余弦正弦値の高周波信号を直接生成するようにしてもよい。If the control period Ts and the high-frequency voltage period Th are determined, the correction phase θi is constant as shown in the equation (1). In consideration of the constancy of the correction phase, FIG. 7 shows a configuration example in which the constant correction phase and the phases θh and k−1 that change with each control period are separated. That is, in the embodiment of FIG. 7, an example is shown in which a high frequency signal having a phase ((θh, k−1) + (θi)) in the equation (34) is generated according to the following equation using a vector rotator.

As a matter of course, a high-frequency signal having a cosine sine value having a phase ((θh, k−1) + (θi)) may be directly generated.

It should be pointed out that the processing can be simplified when some of these are extracted instead of the four amplitudes of the high-frequency amplitudes (cγ, sγ, cδ, sδ). One example of this type of simplification is inevitably occurring when Nh is 2. The amplitude extraction principle based on the equation (29) when Nh is 2 is simplified as follows.

Next, an example of the estimated value generator 10-3 will be described. FIG. 8 shows a rotor using all or part of four amplitudes (cp, sp, cn, sn) (or all or part of four amplitudes (cγ, sγ, cδ, sδ)). The example which produces | generates a phase estimated value and a speed estimated value is shown. The estimated value generator 10-3 is mainly composed of a correlation signal synthesizer 10-3A and a phase synchronizer 10-3B.

当然のことながら、利用する振幅（ｃｐ、ｓｐ、ｃｎ、ｓｎ）、（ｃγ、ｓγ、ｃδ、ｓδ）に依存して、関数ｆｐは異なる。換言するなれば、正相関信号ｐｃを合成するための相関信号合成器は異なる。正相関信号ｐｃの具体的合成法（すなわち、相関信号合成器）は、非特許文献１及び２を通じ、当業者には公知であるので、これ以上の説明は省略する。The correlation signal synthesizer 10-3A uses all or any of the four amplitudes (cp, sp, cn, sn) of the high-frequency current (or all or any of the four amplitudes (cγ, sγ, cδ, sδ)). Thus, the positive correlation signal pc which is the rotor phase equivalent value is synthesized. The positive correlation signal is expressed by Expression (37a) (or Expression (38a)), and has a characteristic of Expression (37b) (or Expression (38b)) in a region where θγ is small.

As a matter of course, the function fp differs depending on the amplitudes (cp, sp, cn, sn) and (cγ, sγ, cδ, sδ) to be used. In other words, the correlation signal synthesizer for synthesizing the positive correlation signal pc is different. Since a specific method for synthesizing the positive correlation signal pc (that is, a correlation signal synthesizer) is known to those skilled in the art through Non-Patent Documents 1 and 2, further description thereof is omitted.

回転子速度推定値ω２ｎ＾は、γδ準同期座標系の速度ωγをそのまま用いてもよいし、ローパスフィルタＦ１ωで処理して用いてもよい。（３９ｃ）式には、直接利用の場合と１次ローパスフィルタを用いる場合の２例を示した。速度推定値生成にはローパスフィルタの使用・不使用の２例がある点を考慮し、図８では、本ローパスフィルタＦ１ωを破線ブロックで示している。According to the invention of claim 4, the phase synchronizer 10-3B forms a phase locked loop (PLL) to generate the estimated values of the rotor phase and the speed. This is described by the following equation (39) using the positive correlation signal pc.

As the rotor speed estimated value ω2n ^, the speed ωγ of the γδ quasi-synchronous coordinate system may be used as it is, or may be used after being processed by the low-pass filter F1ω. Equation (39c) shows two examples of direct use and a case where a primary low-pass filter is used. In consideration of the fact that there are two examples of the use of the low-pass filter and the non-use of the speed estimation value generation, in FIG. 8, the low-pass filter F1ω is indicated by a broken line block.

FIG. 9 shows a second configuration example of the phase speed estimator (digital rotor phase speed estimation apparatus) of the present invention in the digital drive control apparatus. The difference from the embodiment of FIG. 4 is the arrangement position of the phase velocity estimator. FIG. 9 shows an example in which the phase velocity estimator is configured on an αβ fixed coordinate system. The stator current as an input vector signal to the phase velocity estimator and the high-frequency voltage command value as an output vector signal are both vector signals defined on the αβ fixed coordinate system. This configuration means that the high-frequency voltage applying means (high-frequency voltage command device) and the high-frequency current amplitude extracting means (amplitude extractor) are configured on the αβ fixed coordinate system.

As an example of the internal configuration of the phase velocity estimator 10 of FIG. 9, the configuration of FIG. 5 can be adopted. The outline of the internal configuration of the phase velocity estimator 10 is basically the same whether it is configured on the γδ quasi-synchronous coordinate system or on the αβ fixed coordinate system.

As is clear from the fact that the phase velocity estimator 10 is configured on the αβ fixed coordinate system, the high-frequency voltage command device 10-1 included therein is configured on the αβ fixed coordinate system. The configuration method of the high-frequency voltage command device on the αβ fixed coordinate system is basically the same as the configuration method on the γδ quasi-synchronous coordinate system.

As apparent from the fact that the phase velocity estimator 10 is configured on the αβ fixed coordinate system, the amplitude extractor 10-2 included therein is configured on the αβ fixed coordinate system. The configuration method of the amplitude extractor on the αβ fixed coordinate system is basically the same as the configuration method on the γδ quasi-synchronous coordinate system.

As apparent from the fact that the phase velocity estimator 10 is configured on the αβ fixed coordinate system, the estimated value generator 10-3 included therein is configured on the αβ fixed coordinate system. As for the estimated value generator 10-3, the one shown in FIG. 8 cannot be used as it is. An example of the estimated value generator 10-3 that can be used in FIGS. 9, 5, 6, and 7 is rendered as shown in FIG. However, although the generation of the positive correlation signal pc follows the equation (37) or (38), the condition of Kθ = 1 is added. That is, it is necessary to generate the positive correlation signal pc in a form that satisfies Kθ = 1.

An example of a phase synchronizer corresponding to FIGS. 9, 5, 6, and 7 and further configured according to claim 4 is configured as shown in FIG. That is, the phase synchronizer in FIG. 10 generates a deviation signal between the positive correlation signal and the final phase estimation value, and modularly processes the generated deviation signal between −π / 2 and + π / 2 to obtain a modular processing signal. On the other hand, processing similar to that shown in FIG. 8 (that is, processing similar to equation (39)) is performed to generate final phase estimation values and velocity estimation values.

In general, the phase velocity estimator configured on the αβ fixed coordinate system can be used without substantial correction even when the high-frequency voltage command value is superimposed on the drive voltage command value on the uvw coordinate system. In this case, the high-frequency voltage command value on the αβ fixed coordinate system may be converted into the high-frequency voltage command value on the uvw coordinate system via a two-phase / three-phase converter and superimposed on this.

The digital rotor phase speed estimation apparatus (phase speed estimator) according to the present invention has been described in detail with reference to nine specific embodiments. It should be pointed out that the configuration method of the phase velocity estimator is not limited to the above nine examples, and there are various configurations other than the introduction examples according to the present invention.

For the convenience of specific description of the digital rotor phase speed estimation device, the synchronous motor is used as the driving motor and the drive control device related thereto is taken up. However, the digital rotor phase speed estimation device according to the present invention is a synchronous motor. It is pointed out repeatedly that it is not limited to. The digital rotor phase speed estimation device in the drive control device using another AC motor as the drive motor can be used as it is in the specific embodiment described in detail. Regarding the general arrangement of a drive phase control device and a drive control device in which the drive motor is a synchronous reluctance motor or an induction motor, refer to, for example, Patent Literature (Shinji Shinnaka: “Vector control method of AC motor” And the same device ", Japanese Patent No. 4120775 (1902-3-18)) and the like. Therefore, this description is omitted.

The present invention is suitable for applications that require phase estimation performance with high responsiveness in a speed range including zero speed, among applications in which an AC motor is sensorlessly driven.

1 AC motor (synchronous motor)
DESCRIPTION OF SYMBOLS 2 Power converter 3 Discrete time current detector 4a 3 phase 2 phase converter 4b 2 phase 3 phase converter 5a Vector rotator 5b Vector rotator 6 Current controller 7 Command converter 8 Speed controller 9 High frequency component removal filter 10 Phase velocity estimator 10-1 High frequency voltage command device 10-2 Amplitude extractor 10-3 Estimated value generator 10-3A Correlation signal synthesizer 10-3B Phase synchronizer 11 Coefficient unit 12 Cosine sine signal generator

Claims (4)

A stator current digital control function with a period Ts using a power converter and a stator current detector is provided for an AC motor whose rotor exhibits salient pole characteristics when a high frequency voltage having a frequency higher than the drive frequency is applied. A digital rotor phase speed estimation device used in a digital drive control device,
When k is an integer and Nh is a positive integer of 2 or more, high-frequency voltages v1h and k-1 having a period Th = Nh * Ts, a phase θh, k−1, and a basic high frequency ωh = ± 2π / Th are obtained. High-frequency voltage applying means for applying between time t = (k−1) * Ts and t = k * Ts;
When the correction phase θi is expressed as

The phase ((θh, k) with respect to the stator current i1, k detected discretely at time t = k * Ts or with respect to the high-frequency current i1h, k extracted from the stator current i1, k. -1) + (θi)) or high frequency current amplitude extracting means for extracting the amplitude of the high frequency current by performing processing using a high frequency signal having this approximate phase;
Using at least the extracted high-frequency current amplitude, at least one of an estimated value of the rotor phase or an estimated value of the rotor speed basically having a calculus relationship with the rotor phase is generated for each period Ts. A rotor phase speed estimation value generating means;
A digital rotor phase speed estimation apparatus comprising: The high-frequency voltage applying means is configured to apply a high-frequency voltage on a γδ quasi-synchronous coordinate system composed of a γ-axis aimed at synchronization with a rotor phase with zero phase deviation and a δ-axis orthogonal thereto. The digital rotor phase speed estimation apparatus according to claim 1. A stator current or a high-frequency current on a γδ quasi-synchronous coordinate system composed of a γ-axis aimed at synchronization with a rotor phase with zero phase deviation and a δ-axis orthogonal thereto is processed, and the high-frequency current amplitude 2. The digital rotor phase velocity estimation apparatus according to claim 1, wherein the extraction means is configured. 2. The digital rotor phase speed estimation apparatus according to claim 1, wherein the rotor phase speed estimation value generating means is configured to constitute a phase locked loop.

Images