JPH07147800A - Variable speed equipment - Google Patents

Abstract

PURPOSE:To enable highly accurate measurement even with a load connected to an induction motor, by calculating at an operation unit the error in magnetic flux phase between actual magnetic flux axis and control axis from secondary time constants, voltages before and immediately after current cut-off and frequency components among stored measurement data obtained by a first and a second measuring means. CONSTITUTION:A measurement and storage unit 12 measures and stores output current detection values i1d, i1q in the ordinary operation state before a select switch 11 rapidly changes a current command to zero, voltage control signals V1d, V1q and angular frequency omega1. A measurement and storage unit 13 measures and stores output current detection values i1d, i1q in the transitional state immediately after the select switch 11 rapidly changes a current command to zero, voltage control signal V1d, V1q and variations in angular frequency omega1 on a time basis. A constant operation unit 15 calculates the secondary time constant tau2, magnetic flux phase error theta, exciting inductance M' and leakage reactance Lsigma of an induction motor 1. This enables highly accurate measurement of constants.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an induction motor variable speed device, and more particularly to a variable speed device having a constant control function of an induction motor having a current control system for vector control.

[0002]

2. Description of the Related Art As a constant measuring method for an induction motor, JE
There are standards such as C37. In this JEC37, the primary resistance is measured by the DC potential drop method, the leakage inductance and the secondary resistance are measured by the restraint test, and the exciting inductance is measured by the no-load test.

[0003]

In the conventional constant measuring method, the constraint test requires means for fixing the output shaft of the electric motor, and often requires a large-scale instrument for measurement.

Regarding the measurement conditions, since the measured voltage is low, the desired voltage accuracy is obtained by measuring at a frequency higher than the frequency component generated in the secondary circuit during actual operation.

Therefore, in a constraint test of an induction motor in which the secondary conductor has a special shape such as a double squirrel cage, due to the effect of the skin effect, the value of the leakage inductance measured at a frequency higher than the operating frequency and the actual operating time There is an error with the value of.

Therefore, when vector control using measurement constants by a restraint test such as JEC37 or high-performance control such as temperature fluctuation compensation of secondary resistance in vector control is performed, an error occurs in output torque, etc. High precision control becomes difficult.

As described above, in the conventional constant measurement, since the induction machine constant under the actual operating conditions cannot be accurately obtained,
In the vector control, the load torque characteristic test was actually conducted, and the induction machine constant was adjusted according to the operator's feeling so that the torque characteristic result and the operating characteristics such as the transient response characteristics would match the ideal characteristics.

The inventors of the present application have already proposed a method of measuring the constant of an induction machine with high accuracy by utilizing a transient phenomenon (1993 Annual Meeting of the Institute of Electrical Engineers, Proceedings [6], 22.
Page, 636, Induction machine constant measurement using an inverter).

In this measuring method, there is a measurement condition of no-load operation, but in an actual variable speed equipment, no complete no-load operation is possible after the induction machine is incorporated into a gear device, so that constant measurement is required. There may be some load torque and it may not be possible to measure under no-load operating conditions.

An object of the present invention is to provide a variable speed device which can perform highly accurate motor constant measurement even when a load is connected to the induction motor, and which facilitates constant measurement.

[0011]

In order to solve the above-mentioned problems, the present invention has a variable speed control system for controlling an induction machine having a current control system for controlling a current supplied from a semiconductor power converter to the induction machine. In the apparatus, first measuring means for measuring and storing the primary voltage, current and frequency components of the induction machine in a steady operation state, and induction after suddenly changing the current command of the current control system to zero in the steady operation state Second measuring means for measuring and storing the primary voltage, current and frequency components of the machine according to time, and an attenuation coefficient of the voltage measured value after the current command is suddenly changed to zero among the measured and stored data of both measuring means. The secondary time constant τ ₂ of the induction machine is calculated from this, and the magnetic flux phase error Δφ between the actual magnetic flux axis and the control axis is calculated from this secondary time constant τ ₂ and the voltage and frequency components before and immediately after the change of zero current. Variable speed equipment characterized by having an arithmetic unit .

In the present invention, the arithmetic unit approximates the primary voltage measurement value after the current command is suddenly changed to zero by an exponential function, and changes the current command to zero from the coefficient component of the approximate expression. Voltage initial value estimation calculation means for estimating the output voltage component immediately after the operation, speed electromotive force component calculation means for separately calculating the speed electromotive force component of the estimated voltage component, and the magnetic flux from the current component in the steady operation state. A magnetic flux axis current component calculating means for extracting an exciting current component corresponding to an actual magnetic flux axis due to a phase error, and an exciting inductance from an angular velocity when the speed electromotive force and the magnetic flux axis current component and the current command by each calculating means are suddenly changed to zero. M'calculates a leakage component from the voltage component orthogonal to the current in the voltage difference between the voltage measurement value during steady operation and the speed electromotive force, the current value during steady operation, and the angular velocity. And a calculating means for calculating the closet Lσ.

[0013]

In principle, the constant measurement of the induction machine using the variable speed device of the present invention will be described below.

(1) In the approximate vector control of the equivalent circuit, the iron loss component is represented by TI for simplification.
It is treated as an iron loss resistance in parallel with the exciting inductance M ′ of the mold equivalent circuit. This is an approximation that the iron loss current circuit and the secondary circuit are tightly coupled by the exciting inductance M ′ and there is no leakage component.

For simplification of control, this approximation condition is also applied to an equivalent circuit on two axes, and the dq axis circuit is separated and T is separated.
As shown in FIG. 2 which represents a -I-type equivalent circuit, treated as parallel resistance of 'the iron loss resistance R _m and' secondary resistance R _2.

(2) Basic circuit equation In the equivalent circuit of FIG. 2, the voltage / current equation on the biaxial coordinate synchronized with the power supply frequency ω1 can be expressed by the following equation (1).

[0017]

[Equation 1]

V _1d , V _1q : d, q-axis primary voltage i _1d , i _1q : d, q-axis primary current λ _2d , λ _2q : d, q-axis secondary interlinkage magnetic flux ω ₁ : primary angular frequency ω _S : slip angular frequency R _1: primary resistance R _2: T-I type equivalent rotor resistance Lσ: T-I type equivalent circuit primary leakage inductance M ': T-I type equivalent circuit primary magnetizing inductance R _m': T- I-type equivalent circuit iron loss resistance M: Exciting inductance of T-type equivalent circuit P: Differential operator R _2m '= 1 / (1 / R ₂ ' + 1 / R _m ') λ ₂ / M = i _1- (i ₂ '+ IRM') (3) State during steady operation In a vector control variable speed device having a current control system, the AC motor is supplied to the induction motor from a semiconductor power converter such as a voltage source inverter during steady operation. The current is controlled by a current control system that performs vector control calculation. The current command to the current control system is the torque current component I _1q and the exciting current component I _1d.
And give.

In steady operation with an arbitrary speed and load torque, the d-axis and q-axis current components are the exciting current in the d-axis component and the torque current component in the q-axis component, as shown in the following equation (2). There is an iron loss current component. Each component is given a () _-0 symbol during normal operation.

[0020]

(2) (i _1d ) _-0 = (λ _2d / M) _-0 , (λ _2q / M) _-0 = 0 (i _1q ) _-0 = (i _2q ') _-0 + (iRM _q ') _-0 …… (2) (i _1d ) _-0 : Steady state primary d-axis current component (i _1q ) _-0 : Steady state primary q-axis current component (λ _2d / M) _-0 , (λ _2q / M) _-0 : d, q-axis component obtained by dividing the secondary flux linkage by M (i _2q ') _-0 : secondary torque current component (iRM _q ') _-0 : secondary iron loss current component When applied to the equivalent circuit of FIG. 2, it becomes that shown in FIG. The actual magnetic flux axis is the d-axis, and the magnetic flux component in the steady operation state is (a) the magnetic flux component (λ _2d /
M) _-0 (b) Leakage flux component L that does not interlink with the secondary due to the exciting current L
σ · (i _1d ) ₋₀ (c) There are three types of leakage flux component Lσ · (i _1q ) _{−0 that} do not interlink to the secondary due to the torque current. (Λ _2q / M) ₋₀ is zero.

Here, the voltage / current components during steady operation are represented in a space vector diagram as shown in FIG.

(3) State immediately after sudden change in current zero When the current command to the current control system is suddenly changed to zero from the steady operation state, the response of the current control system is sufficiently fast, and the output voltage capacity of the current control system is high. If it is large enough, the primary current also instantly changes to zero. At this time, the magnetic energy of the leakage magnetic flux of the d and q axis components, which is linked to the leakage inductance Lσ, is also absorbed by the power source and becomes zero.

However, since the d-axis secondary interlinkage magnetic flux component does not interlink with the primary circuit even when the primary current becomes zero, it tries to maintain the magnetic flux component. Therefore, when the d-axis primary current (i _1d ) _-0 component becomes zero, it is commutated to the d-axis secondary circuit. Each component is added with () ₊₀ symbol at zero current operation.

[0024]

(3) (λ _2d / M) ₊₀ =-(i _2d ') ₊₀ = (i _1d ) _-0 (3) The state immediately after this sudden current change is shown in FIG. 5 and corresponds to FIG. The space vector is shown in FIG.

As described above, since the primary current is zero when the current suddenly changes, the magnetic flux component is only the secondary interlinkage magnetic flux. The current component has an exciting current component (i _2d ') ₊₀ and a core loss current component (iRM _q ') ₊₀ that are secondarily commutated. Here, the iron loss current component is sufficiently smaller than the exciting current component and can be ignored.

(4) Transient Phenomenon after Sudden Change in Current It is assumed that the primary current is kept abruptly changed to zero and the d and q control shafts are rotating in synchronization with the rotor of the induction machine. At this time, the primary current is maintained at zero, and the secondary circuit is not affected by the primary current.

For this reason, the secondary magnetic flux component and the secondary current component on the coordinate synchronized with the rotor are initialized by the condition of the equation (3), and energy is consumed by the secondary resistance R ₂ '. It changes as a DC transient phenomenon.

Therefore, the secondary interlinkage magnetic flux becomes an exponentially decaying equation such as the following equation (4).

[0029]

Λ _2d / M (t) = (λ _2d / M) ₋₀ · exp (−1 / τ ₂ · t) λ _2d / M (t) = 0 (4) τ ₂ : Secondary time constant _{(= M '/ R 2'} ) voltage generated in the primary circuit by the secondary magnetic flux, a d-axis voltage component (a transformer electromotive force due to a change in magnetic flux component itself) q
There is an axial voltage component (velocity electromotive force due to relative motion between the magnetic flux on the rotor and the stator), and each is given by the following equation (5).

[0030]

V _1d (t) = p {λ _2d / M) _{+ 0} · exp (−1 / τ ₂ · t)} = −1 / τ ₂ · M ′ · (λ _2d / M) _{+ 0} · exp (-1 / τ ₂ · t) (5-1) V _1q (t) = ω ₁ (t) · M ′ · (λ _2d / M) _{+ 0} · exp (−1 / τ ₂ · t) ) (5-2) FIG. 7 shows d *, q * of the control axis of the output voltage of the current control system when the current is suddenly changed to zero using the variable speed system.
The measurement waveform of an axis component is shown.

As is clear from this measurement waveform, the voltage of the exponential function of the equations (5-1) and (5-2) is generated on each axis. These d * and q * axis voltage components are measured and stored for about a quadratic time constant.

(5) Induction machine constant calculation (5a) Estimating the secondary time constant and the voltage value immediately after the sudden current change by approximation of the exponential function Note that the actual magnetic flux axes d and q and the control system internal It does not always match the assumed magnetic flux axis d *. Actually, as shown in FIGS. 4 and 6, there may be an axis deviation Δφ between these axes. Therefore, the control axis is defined by the symbols d * and q *, and the measurement is performed using these d * and q *
Do it on the axis.

From the measured data when the current suddenly changes to zero, the d * and q * axis voltage measurements during transient are approximated by the exponential function of the following equation (6) to _obtain (V _1d *) ₊₀ , (V _From the coefficients of _1q *) ₊₀ and τ _2q *, find the initial value and the secondary time constant immediately after the sudden change of the current to zero.

[0034]

[Equation 6] V _1d * (t) = (V _1d *) _{+ 0} · exp (-1 / τ _2d * · t) V _1q * (t) = (V _1q *) _{+ 0} · exp (-1 / τ _2q * · t) (6) Here, usually the q-axis component (velocity electromotive force) rather than the d-axis component
Since the measured voltage is larger, the secondary time constant τ ₂ ≈ τ _2q * is used from the viewpoint of measurement accuracy. Also, (V _1d *) ₊₀
And (V _1q *) ₊₀ can be regarded as an approximate value of the initial value of the transient phenomenon when the current changes suddenly.

(5b) Calculating the axis shift phase between the control axis and the actual magnetic flux axis In the above equation (6), if the rotation speed (ω _r = ω ₁ ) is approximated to be constant, (5-1), (5- From the relationship of equation 2), d *,
The ratio of the q * axis voltage component is constant as in the following equation.

[0036]

[ _Equation 7] (V _1d *) ₊₀ / (V _1q *) ₊₀ = (-1 / τ ₂ ) / ω ₁ ≈ constant Therefore, if there is little attenuation of the rotation speed, the voltage of each axis at any time The ratio is constant. This is shown in FIG. 8 as an example. This figure shows the voltage component of each axis measured in FIG.
It is converted to the plane coordinates of the * and q * axes.

Although the voltage component changes abruptly when the current zero changes suddenly, after that, the voltage vector moves on the straight line of the following equation (7) passing through the origin with the passage of time.

[0038]

[ _Equation 8] V _1q (t) = (− τ ₂ · ω ₁ ) V _1d (t) (7) Conversely, the measured voltage is the q-axis speed electromotive force V
_1q and the component of the d-axis transformer electromotive force V _1d are combined V ₁ to d
Measured with the control axes of * and q * as the reference.

Here, the q-axis phase orthogonal to the actual magnetic flux axis is rotated from the phase of the measured voltage vector (V ₁ ) _{+0 by} the phase of the following equation (8) from the relationship of the above equation (7). It can be seen that it exists in the phase.

[0040]

[Equation 9] θ _a = tan ⁻¹ (τ ₂ '· ω ₁ ) (8) Further, the phase of the actual magnetic flux on the control axis is

[0041]

(Φ _V *) ₊₀ = tan ^-1 {(V _1q *) ₊₀ / (V _1d *) ₊₀ } Δφ = (φ _V *) ₊₀ −θ _a −90 ° ...... ( 9) Here, ω ₁ is known from the operating condition, τ ₂ can use the coefficient of the approximate expression of the above equation (6), and the axis deviation angle Δφ can be calculated from the measurement and calculation results up to now. is there.

(5c) Calculation of Excitation Inductance M ′ and Leakage Reactance Lσ If the axis shift phase component Δφ is obtained, the excitation inductance M ′ is orthogonal to the amplitude component of velocity electromotive force component (V _1q ) ₊₀ = E _2. It is possible to calculate from the exciting current component.

The velocity electromotive force component is

[0044]

[Equation 11] | E ₂ | = {(V _1d *) ₊₀ ² + (V _1q *) ₊₀ ² } ^1/2 · cos (θ _a ), and the exciting current component is

[0045]

(12) (I ₀ ) ₊₀ = | (i ₁ ) _-0 | · cos (φ _i −Δφ)}. From these, the magnetizing inductance M ′ is obtained from the following equation.

[0046]

[Equation 13] M ′ = | E ₂ | / {ω ₁ · | (i ₁ ) ₋₀ | · cos (φ _i −Δφ)} (10) The leakage reactance Lσ is also the voltage (V ₁ ) A component orthogonal to the current of the voltage difference between _-0 and E ₂ (δ axis)
Then, it can be obtained from the current component at the time of steady operation by the following equation (11).

[0047]

[Formula 13] ΔV ₁ δ = | (V ₁ ) ₋₀ | · sin {(φ _V *) ₋₀ − (φ _i *) ₋₀ } − | E2 | · sin {(φ _V *) ₊₀ − (Φ _i *) ₋₀ −θ _a } Lσ = ΔV ₁ δ / {ω ₁ · | (i ₁ ) ₋₀ |} (11) (5d) Excitation inductance M ′ and leakage reactance under no load condition Calculation of Lσ The axis deviation can be ignored under no load.

[0048]

[Number 14] _{(i 1) -0 ≒ (I} 1d *) -0, θ a ≒ 0, | E 2 | ≒ (V 1q *) +0, ΔVδ ≒ (V 1q *) -0 - (V 1q *) ₊₀ holds, and the following simplified formula is obtained.

[0049]

[Equation 15] M ′ = (V _1q *) ₊₀ / {ω ₁ · (I _1d *) ₋₀ } Lσ = {(V _1q *) ₋₀ − (V _1q *) ₊₀ } / {ω ₁・ (I _1d *) _-0 } (12) From the above, according to the present invention, the induction in which the load torque exists from the measured current / voltage value during steady operation and the measured current / voltage value during sudden current zero change. Calculate the machine constant.

[0050]

FIG. 1 is a block diagram showing an embodiment of the present invention. The induction motor 1 is supplied with a vector-controlled AC voltage having a PWM waveform from the semiconductor power converter 2. The rotor speed ω _r of the induction motor 1 is detected by the speed detector 3. The primary current of the induction motor 1 is detected by the current detector 4 by separating it into a torque current component and an exciting current component.

The current control systems 5 and 6 use the exciting current command i _1d *
And the torque current command i _1q * are compared with the respective detected current components, and the exciting voltage control signal V _1d and the torque voltage control signal V _1q are given to the semiconductor power converter 2 so that the current is generated according to the command current.

The exciting current calculator 7 receives the speed detection signal ω _r as an input and outputs an exciting current command (or exciting magnetic flux) i _1d . The torque current calculator 8 obtains a torque current component orthogonal to the magnetic flux axis from the torque command from the speed control system and the exciting current command, and outputs the torque current command i _1q .

The slip frequency calculating section 9 determines the exciting current command i
_The slip frequency ω _s is obtained from _1d and the torque current command i _1q . The frequency / phase calculator 10 adds the slip frequency ω _s and the rotor speed detection value ω _r to find the frequency of the AC voltage supplied to the induction motor 1, and integrates the frequency to find the phase θ of the magnetic flux axis. .

The vector control variable speed device configured as described above is provided with the calculation / measurement elements 11 to 15 as constant measuring means of the induction motor 1 in this embodiment.

The changeover switch 11 gives the current commands i _1d *, i _1q * from the exciting current calculator 7 and the torque current calculator 8 to the current control systems 5 and 6 in the steady operation state, and the current command i is switched by the changeover. _Suddenly change _1d * and _i1q * to zero.

The measuring / storing unit 12 stores the output current detection values i _1d , i _1q (or the current commands i _1d *, i _1q *) in the steady state before switching, in which the changeover switch 11 suddenly changes the current command to zero.
And voltage control signals V _1d and V _1q (or voltage applied to the induction motor 1) and angular frequency ω ₁ are measured and stored.

The measuring / storing unit 13 stores the detected output current values i _1d , i _1q (or current commands i _1d *, i _1q *) in the transitional state after the changeover switch 11 suddenly changes the current command to zero.
And the changes in the voltage control signals V _1d and V _1q (or the voltage applied to the induction motor 1) and the angular frequency ω ₁ are measured and stored by time.

The measurement commanding unit 14 sends a data acquisition command to the measurement / storage unit 12 in the steady operation state during constant measurement,
At the same time when the changeover switch 11 is switched, a data acquisition command is issued to the measurement / storage unit 13.

The constant calculation unit 15 includes the two measurement / calculation units 12,
Induction motor 1 reads the data measured and stored by 13
Each constant of is calculated. This calculation is performed by calculating the secondary time constant τ ₂ by the equation (6), calculating the magnetic flux phase error Δφ by the equation (9), calculating the exciting inductance M ′ by the equation (10), and (11) ) The leak reactance Lσ is calculated by the equation.

As an experiment in this embodiment, the current control systems 5 and 6 were configured to be PI-controlled on the rotating coordinates, and 22 KW was used.
The test induction motor of -6P, 325V-1180 rpm was operated at 1000 rpm to measure the constants.

[0061] Figure 7 is a no-load voltage V when the sudden change of the current to zero during operation _1d *, V _1q * indicates a waveform, 17 points samples (about tau _2/2 hours) to 20ms for each of the voltage waveform Then, each constant was obtained. As a result, the axis deviation obtained by approximating with the above equation (8) is Δφ = 0.3 °, and each constant was calculated using the equation (12) under the condition that there is no axis deviation. The results are shown in Table 1 below.

[0062]

[Table 1]

In order to verify the accuracy of this constant measurement result,
The torque characteristics were measured by applying it to a vector-controlled variable speed device. This characteristic is shown in FIG. Here, the iron loss current compensation value is set to zero and is not corrected. Further, due to the convenience of the load device, measurement was performed up to 75% of the rated torque, and the load region beyond that was extrapolated and indicated by the dotted line.

From this characteristic, when the secondary resistance R ₂ is not compensated, a torque error of about 10% occurs at the time of rating, and the secondary time constant τ ₂ is measured to be lower than the actual machine from the torque characteristic. It is estimated to be. This is considered to be due to the iron loss resistance existing in parallel with the secondary resistance R ₂ and the magnetic saturation which changes a little.

However, when the secondary resistance compensation using the model and the output error voltage is added, the torque error is about 3%.
It was found that when the measured values of the exciting inductance M ′ and the leak reactance Lσ used in the model are about 1000 rpm, sufficient accuracy can be obtained for practical use.

The secondary time constant itself changes depending on the temperature during operation, and if the torque accuracy is actually maintained during operation, feedback control such as secondary resistance compensation will be applied.

In other words, the secondary time constant changes during operation, accuracy of initial setting is not so required, and a rough value may be set.

On the other hand, the components of M ′ and Lσ directly affect the accuracy of the secondary resistance compensation control. Therefore, when performing secondary resistance compensation or the like in order to maintain torque accuracy, an accurate induction machine constant is required.

The constant measurement according to the present invention does not become unstable even without secondary resistance compensation, and it can be said that the measured value of R ₂ 'has sufficient accuracy as an initial setting value.

Further, it was found that the torque accuracy of the components of M ′ and Lσ was also improved as compared with the torque characteristics when the secondary resistance compensation control was performed, and the accuracy which can be sufficiently put into practical use was obtained.

[0071]

As described above, according to the present invention, by using the variable speed device for controlling the induction machine, the measured values of the primary voltage, the current and the frequency component of the induction machine in the steady operation state and the steady operation state can be obtained. Since the constants of the induction machine are calculated from the measured values of the primary voltage, current and frequency components of the induction machine when the current command of the current control system is suddenly changed to zero, the following effects are obtained.

(1) By suddenly changing the current command to zero, a secondary current is generated by utilizing the commutation phenomenon, whereby a reliable secondary current can be generated even at a high frequency. In addition, the velocity electromotive force component increases due to the high frequency, and the voltage detection error is reduced, so that highly accurate constant measurement can be performed.

(2) Since the current command is set to zero and the output current is maintained at zero, a voltage equal to the induced electromotive force of the induction machine can be generated as the output of the current control system. This makes it possible to measure the output voltage of the current control system and obtain a measured value equivalent to the terminal voltage of the induction machine. That is, it is possible to eliminate the need for an external measuring device such as a voltmeter, and measure the variable speed device alone.

(3) Since the attenuation phenomenon of the secondary circuit is used, the frequency component of the magnetic flux change can be regarded as a secondary time constant. Therefore, the leakage reactance component Lσ
The skin effect that affects the can be measured under conditions close to the actual operating condition, and the measurement accuracy can be improved.

(4) Since the variable speed device can be provided with a small amount of measuring means to measure constants with high accuracy, (a) constants can be measured for each induction machine having variations in the manufacturing process, and optimum operation can be performed. It becomes possible, (b) the conventional manual adjustment becomes unnecessary, and the adjustment man-hours can be reduced. (C) The optimum constant setting can be made even for induction machines whose constants are unknown, such as existing induction machines. What can be done, (d) there is an effect that the constant of the induction machine is regularly measured and the fluctuation of the constant is checked to judge the abnormality of the induction machine.

[Brief description of drawings]

FIG. 1 is a block diagram showing an embodiment of the present invention.

FIG. 2 is a TI type equivalent circuit diagram of an induction machine in which d and q axis circuits are separated.

FIG. 3 is a diagram showing current and magnetic flux in the primary and secondary circuits in a steady operation state.

FIG. 4 is a voltage vector diagram of FIG.

FIG. 5 is a diagram showing the current and magnetic flux of the primary and secondary circuits when the current suddenly changes to zero.

FIG. 6 is a voltage vector diagram of FIG.

FIG. 7 is a transient voltage waveform diagram at the time of sudden change in current.

FIG. 8 is a voltage vector locus.

FIG. 9 is a torque characteristic diagram during vector control.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 ... Induction machine 2 ... Semiconductor power converter 5, 6 ... Current control system 7 ... Excitation current calculation part 8 ... Torque current calculation part 11 ... Changeover switch 12, 13 ... Measurement / storage part 14 ... Measurement command part 15 ... Constant calculation Department

Claims (2)

[Claims] 1. A variable speed device having a current control system for controlling a current supplied from a semiconductor power converter to an induction machine, the primary speed and current of the induction machine in a steady operation state, and A first measuring means for measuring and storing a frequency component; and a primary voltage, a current and a frequency component of the induction machine after the current command of the current control system is suddenly changed to zero in the steady operation state and measured and stored by time. The second time constant τ ₂ of the induction machine is calculated from the attenuation coefficient of the voltage measurement value after the current command is suddenly changed to zero among the measurement and storage data of the second measurement means and the two measurement means. A variable speed device comprising: a calculation unit that calculates a magnetic flux phase error Δφ between an actual magnetic flux axis and a control axis based on a voltage and a frequency component before and immediately after a change of the next time constant τ ₂ and zero current. 2. The calculation unit approximates the primary voltage measurement value after the current command is suddenly changed to zero by an exponential function, and outputs an output immediately after changing the current command to zero from a coefficient component of the approximate expression. A voltage initial value estimation calculation means for estimating a voltage component, a speed electromotive force component calculation means for separately calculating a speed electromotive force component of the estimated voltage component, and a magnetic flux phase error from a current component in the steady operation state. A magnetic flux axis current component calculating means for extracting an exciting current component corresponding to the magnetic flux axis, and an exciting inductance M ′ is calculated from the velocity electromotive force by each calculating means, the magnetic flux axis current component, and the angular velocity when the current command is suddenly changed to zero. And a leakage inductance Lσ from the voltage component orthogonal to the current in the voltage difference between the voltage measurement value during steady operation and the speed electromotive force, the current value during steady operation, and the angular velocity. Variable speed device according to claim 1, further comprising a calculating means for calculation for.