JPH0685638B2 - Induction motor speed sensorless speed estimation device - Google Patents

Description

Description: TECHNICAL FIELD The present invention relates to a speed estimation method suitable for speed control or speed measurement of an induction motor without a speed sensor.

(Prior Art) Described in a paper entitled "VVVF speed sensorless vector control of an induction motor applying the instantaneous space vector theory" by the present inventors, etc., on pages 13-15 of the 1989 annual conference of the Japan Society of Electrical Engineers. As described above, the rotational angular velocity ωm, which is proportional to the speed of the induction motor, has been conventionally estimated and calculated as a value obtained by subtracting the slip angular velocity ωs proportional to the output torque from the rotational angular velocity ω ₂ of the secondary interlinkage magnetic flux.

Fig. 2 shows PWM, which is common in variable speed systems of induction motors.
The block diagram which applied the said estimation calculation method to the inverter drive system is shown. Power is supplied to the 3-phase induction motor 3 from the DC power supply 1 via the 3-phase PWM inverter 2.

The voltage detection calculation unit 4 determines the input voltage Vu, V of each phase of the induction motor 3.
Detects v, Vw and formula Using Convert to. The current detection calculation unit 5 calculates the u-phase current iu and the v-phase current
iv and detect the expression Using Convert to.

The magnetic flux calculation unit 6 Is calculated. The calculation block diagram is shown in FIG. In the constant multiplication unit 61, the subtraction unit 62, and the integration unit 63, formula Is calculated more. That is, in the constant multiplication unit 61, the primary resistance value R ₁ And the output is subtracted by the subtractor 62. And the output of the subtractor 62 is integrated by the integrator 63, Is being calculated.

After that, the secondary conversion multiplication unit 64, the leakage constant multiplication unit 65, the subtraction unit 66
Then the formula By Is calculated. Where Ll is the leakage inductance,
It is expressed by the formula Ll = L ₁ L ₂ / M−M, where L ₁ is the primary self-inductance, L ₂ is the secondary self-inductance, and M is the mutual inductance. That is, in the secondary conversion multiplication unit 64 Is multiplied by L ₂ / M, and in the leakage constant multiplication unit 65 Is multiplied by Ll, and the subtraction unit 66 subtracts the output of the leakage constant multiplication unit 65 from the output of the quadratic conversion multiplication unit 64, Is being calculated.

The magnetic flux velocity calculation unit 7 in FIG. Then, the rotational angular velocity ω ₂ of the secondary flux linkage vector is given by Calculate more. Where φ ₂ d and φ ₂ q are respectively D and q axis components of.

The slip calculation unit 8 calculates the slip angular velocity ω ₅ of the rotor of the induction motor with respect to the secondary interlinkage magnetic flux by Calculate more. Where i ₁ d and i ₁ q are D and q axis components of R, and R ₂ is a secondary resistance value.

Each output of the magnetic flux velocity calculation unit 7 and the slip calculation unit 8 is subtracted by a subtraction unit
Then, the rotational angular velocity ωm of the rotor can be estimated and calculated by calculating the expression ωm = ω ₂ −ω ₅ .

(Problems to be Solved by the Invention) In the above-mentioned conventional technique, integral calculation is performed in the calculation of the primary interlinkage magnetic flux vector in the process of calculating the secondary interlinkage magnetic flux vector. If there is a drift in the integrator, it will be greatly amplified at the output of the integrator in the low speed range,
A calculation error occurs in the primary interlinkage magnetic flux vector, and the secondary interlinkage magnetic flux vector calculated from the primary interlinkage magnetic flux also includes a large error. Therefore, the rotational angular velocity ωm also obviously differs greatly from the actual rotational angular velocity at low speeds. Also,
The primary resistance value R ₁ is used in the calculation of the primary flux linkage vector, and even if R ₁ fluctuates due to temperature fluctuations of the electric motor, etc., a small error will occur in the input of the integrator. For this reason, the error of the rotational angular velocity ωm calculated at low speed increases.

In other words, the conventional method can only accurately estimate the speed only in the high speed range, and if an attempt is made to control the rotation speed of the induction motor by using the estimated speed by this speed estimation method, the speed command in the low speed range will be used. The error between the value and the actual speed becomes large, and the high-accuracy speed control range cannot be widened. For the same reason, so-called four-quadrant operation in which the speed of the induction motor is continuously changed from normal rotation to reverse rotation is impossible.

The present invention has been made to solve such a problem.

(Means for Solving the Problem) As means for solving the problems of the conventional technique due to the drift of the magnetic flux calculation unit 6, instead of the magnetic flux calculation unit using only the magnetic flux calculation unit 6 in the conventional technique, it is shown in FIG. As described above, the voltage-system magnetic flux calculation unit 120 equivalent to the conventional technology, and the calculated rotational angular velocity ωm A composite magnetic flux calculating means including a current magnetic flux calculating unit 130 for calculating a magnetic flux vector is used.

Further, as means for solving the problems of the prior art caused by the fluctuation of the primary resistance value, together with the above-mentioned means, as shown in FIG. 5, a slip calculator 15 equivalent to the slip calculator 8 in the prior art, And a second slip calculation unit (10) for obtaining a slip angular velocity by inner product calculation with, and a slip correction unit (11) which outputs the both of them as inputs and outputs a stable slip angular velocity ωs with respect to fluctuations of the primary resistance value. Use means.

The voltage-based magnetic flux calculation unit 120 shown in FIG. Using Is calculated, where p is a differential operator. In other words, the voltage drop calculation unit 121 calculates the voltage drop due to the primary resistance and the leakage inductance, and the subtraction unit 122. The voltage drop is subtracted from the secondary conversion multiplying unit 123 L ₂ / M
After multiplying, it is integrated by the integrator 125 and Are seeking. However, the subtractor 124 is placed before the integrator 125 to subtract the output of the error amplifier 141 from the input signal of the integrator.

The current system magnetic flux calculation unit 130 is an expression that is a characteristic expression on the secondary side of the induction motor. Is used to calculate the secondary flux linkage vector, where Is a secondary interlinkage magnetic flux vector, but a letter i is added to distinguish it from the one obtained by the equation. That is, Is multiplied by (M / L ₂ ) R ₂ in the constant multiplication unit 134 and is the output of the integrator 136. R ₂ / L ₂ times with the quadratic constant multiplier 131, And the angular velocity ω _{2 of the} rotor (calculated value) are calculated by the multiplier 133, and the output thereof is advanced by 90 ° in the phase advancer 132. The outputs of the constant multiplication unit 134, the quadratic constant multiplier 131, and the phase advancer 132 are added or subtracted by the adder / subtractor 135 and input to the integrator 136. The output of the integrator 136 is the output of the current system magnetic flux calculation unit 130. Becomes This and the output of the voltage system magnetic flux calculator 120 And the error is calculated by the subtractor 142, and this error is calculated by the error amplifier 1
It is multiplied by K at 41 and is fed back to the integrator 125 of the voltage system magnetic flux calculation unit 120.

The slip calculating means shown in FIG. Using as input, the same formula as the slip calculating unit 8 in the conventional technique is calculated, and the slip angular velocity ωs in the formula is output as ωs ₁ . The second slip calculation unit 10 is the same And input Expression using inner product operation of and Is calculated and output as the second slip angular velocity ωs ₂ .

The slip correction unit 11 subtracts the difference between the slip angular velocity ωs which is the output for calculating the rotational angular velocity ωm of the rotor and the second slip angular velocity ωs ₂ which is the output of the second slip computing unit 10.
After the calculation is performed by 112 and the output thereof is integrated by the integrator 111, the addition unit 113 calculates the sum with the output ωs ₁ of the slip calculation unit 15 and outputs the slip angular velocity ωs.

(Operation) First, the operation of the composite magnetic flux calculating means shown in FIG. 4 will be described. It is the output of the current system magnetic flux calculation unit 130. Is the output of the integrator 136, but its input through the quadratic constant multiplier 131 Is fed back negatively, the amplified error in the low frequency range due to the drift of the integrator, which was a problem of the prior art, is output. Not included in. But this Since the calculation of (1) includes the secondary resistance value R ₂ which is a fluctuating constant, it cannot be completely trusted. Therefore, this Taking advantage of the low drift in the low frequency range, the integrator 1 of the voltage system magnetic flux computing unit 120, which was a problem of the conventional technology,
It is used to correct the drift of 25. In other words, due to the drift If contains an error, Is generated, which is amplified and fed back to the input of the integrator 125, The drift of is suppressed. In general Is an alternating amount, and the higher the frequency, the longer the delay in drift correction and the lesser the effect of drift correction becomes. Against Means that the effect of will be lessened. That is, for the fluctuation of the secondary resistance value R ₂ , Will not compromise the traditional advantage of being unaffected.

Next, the operation of the slip calculation means shown in FIG. 5 will be described. The output slip angular velocity ω s ₁ of the slip calculation unit 15 can be calculated by setting the primary resistance value R ₁ of the composite magnetic flux calculation unit correctly if Is accurate, and is calculated as an accurate slip angular velocity both steadily and transiently, but if the primary resistance value R ₁ is not accurate, Has a drawback that the calculation error of the slip angular velocity becomes larger as the velocity becomes lower. However, the second slip calculation unit 10
Slip angular velocity .omega.s ₂ outputs the second of the reasons to be described later, includes an error in the setting of the primary resistance value R ₁ also , The slip calculation error due to the error is smaller than the output slip angular velocity ω s ₁ of the slip calculation unit 15. Particularly, if a certain condition is satisfied, there is an advantage that even if the setting of the primary resistance value R ₁ includes an error in the restrained state of the electric motor, no error occurs in the calculated value of the second slip angular velocity ω s ₂ . But,
Since the output second slip angular velocity ωs ₂ of the second slip calculation unit 10 is inaccurate in the transient state, the slip correction unit 11 outputs the slip angular velocity ωs ₁ of the slip calculation unit 15 which is the calculated value of the conventional slip angular velocity to this slip angular velocity ωs ₁ . By adding the output of the integrator 111 that corrects the error with the slip angular velocity ωs ₂ , the second slip angular velocity ωs ₂ becomes equal to the output slip angular velocity ωs of the slip correction unit 11 only in the steady state. By doing so, it is possible to steadily reduce the speed estimation error in the low speed range due to the fluctuation of the primary resistance value R ₁ which is a problem of the conventional technology, and does not impair the conventional characteristics even in the transient state. You can

The reason why the second slip angular velocity ω s ₂ which is the output of the second slip calculation unit 10 is not steadily influenced by the error of the primary resistance value R ₁ will be described below. In order to know the generation error of the second slip angular velocity ωs ₂ due to the setting error of the primary resistance R ₁ , the relationship between the actual slip angular velocity ωsr and the second slip angular velocity ωs _{2 in} the presence of the primary resistance setting error ΔR ₁ can be investigated. Good.

The input of the integrator 125 of the voltage system magnetic flux calculation unit 120 in FIG. It is represented by. In steady state, Can be assumed to be sine waves with angular frequency ω (= ω ₂ ), so both sides of the equation can be integrated. Becomes here, Represents the real secondary flux linkage vector and is equal to the right side of the equation. Expression Solving for Becomes Also, when the right side of the equation is calculated under the assumption of steady state as described above, Next, by considering the ωm = ω ₂ -ωs as ω = ω ₂ Solving for Becomes Where T ₂ = L ₂ / R ₂ . In the same way Also, when the actual slip angular velocity is represented by ω ₂ r, Can be written in. ， By substituting the expression into the expression, Becomes

here, When it is calculated and arranged to be decomposed into Where A = ωM ² (ω + ωsr T ₂ K) (1 + ωs ² T ₂ ² ) −KL ₂ ΔR ₁ (1 + ωs ² T ₂ ² ) (1 + ωsr ² T ₂ ² ) + KM ² (K−ωωs T ₂ ) (1 + ωsr ^{2 _{T 2 2) B = ωM 2}} (K-ωωsrT 2) (1 + ωs 2 T 2 2) + ωL 2 ΔR 1 (1 + ωs 2 T 2 2) (1 + ωsr 2 + T 2 2) -KM 2 (ω + ωsT 2 K) (1 + ωsr 2 T ₂ ² ) C = M (K ² + ω ² ) (1 + ωs ² T ₂ ² ) (1 + ωsr ² T ₂ ² ) That is, Substituting this into the numerator and denominator parts of the equation, φ ₂ d · i ₁ q − φ ₂ q · i ₁ d = {(A / C) i ₁ d + (B / C) i ₁ q} _{i 1 q - {(A /} C) i 1 q + (B / C) i 1 d} i 1 d = - (B / A) (i 1 d 2 + i 1 q 2) = - (B / A) i ₁ ² φ ₂ d ・ i ₁ d + φ ₂ q ・ i ₁ q = {(A / C) i ₁ d- (B / C) i ₁ q} i ₁ d + {(A / C) i ₁ q + (B / C) i ₁ d} i ₁ q = (A / C) (i ₁ d ² + i ₁ q ² ) = (A / C) i ₁ ² , so the formula is Becomes In the steady state, the slip correction unit 11 causes ωs =
Since ωs ₂ is obtained, rearranging ωs on the right side of the equation as ωs ₂ , it becomes a cubic equation related to ωs _{2. In} this equation, (1 + ω
Since it contains a factor of s ₂ T ₂ ) ² , we can get only one real solution of

Here, D and E are respectively D = M ² ωωsrT ₂ (ω + ωsrT ₂ K) −L ₂ ΔR ₁ ω (1 + ωsr ² T ₂ ² ) E = M ² ω (ω + ωsrT ₂ K) −L ₂ ΔR ₁ K (1 + ωsr ² T ₂ ² ). The formula is the actual slip angular velocity ωsr and the second slip calculation unit 10
It shows the relationship with the second slip angular velocity ωs ₂ which is the output of, and there is no primary resistance value setting error, and if ΔR ₁ = 0, naturally ω ₂ s = ωsr. Also in the restrained state ω = ω
Since sr + ωm holds, ω = ωsr in the restrained state, and ωs ₂ = ωsr if K = 1 / T ₂ . That is, in the state where the induction motor is restrained, even if there is a setting error ΔR _{1 in the} primary resistance value, the calculated value of the second slip angular velocity ωs ₂ does not include an error.

Since the rotational angular velocity ω ₂ is calculated so as not to be affected by the constant variation, the rotational angular velocity ωm of the rotor calculated by the difference between the rotational angular velocity ω ₂ and the slip angular velocity ωs can obtain an accurate value.

(Embodiment) FIG. 1 is a block diagram of an embodiment in which a speed sensorless speed estimation method for an induction motor according to the present invention is used in an inverter drive system, and the same symbols as those in the second embodiment indicate the same parts. Therefore, the description of the same parts as those in FIG. 2 will be omitted, and only the blocks 8 ', 10, 11, 15, 16 different from those in FIG. 2 will be described in detail.

A block 16 is the composite magnetic flux calculation unit described in FIG. 4, and is an output of the voltage detection calculation unit 4. It is the output of the current detection calculation unit 5. And the calculated value ωm of the rotational angular velocity of the rotor, Is output.

Block 8'is the slip calculation unit described in FIG.
It is the output of the current detection calculation unit 5. It is the output of the composite magnetic flux calculator 16. Is input to calculate and output the slip angular velocity ωs. As described in FIG. 5, the contents are as follows: The slip calculation unit 15 that outputs the slip angular velocity ω s ₁ calculated by the formula as in the conventional slip calculation, and the formula that calculates the second slip angular velocity ω S ₂ Second slip calculation unit 10 and slip correction unit
It is composed of 11 and.

The rotational angular velocity ωm of the rotor is calculated in the subtraction unit 9 as the rotational angular velocity ω ₂ of the magnetic flux output from the magnetic flux velocity calculation unit 7 minus the slip angular velocity ωs output from the slip calculation unit 8 ′.

(Effects of the Invention) As described in detail above, in the speed sensorless speed estimation method for an induction motor, the secondary interlinkage magnetic flux vector calculation means according to the related art uses the secondary interlinkage magnetic flux due to the drift of the integrator in the low speed range. The vector was not accurately calculated, and the rotational angular velocity of the rotor calculated by the vector contained a large error. However, as in the present invention, by adding the current system magnetic flux calculation unit to the secondary linkage magnetic flux calculation unit so as to suppress the integral drift of the voltage system magnetic flux calculation unit, there is no integral drift even in the low speed range and it is accurate. The secondary flux linkage vector is calculated, and therefore the rotational angular velocity of the rotor is also calculated accurately.

Also, the secondary chain magnetic flux vector cannot be accurately calculated in the low speed range due to the fluctuation of the primary resistance value or the setting error, but the slip calculation in the newly installed second slip calculation section has the effect of the conventional slip calculation method. Since it can be made smaller than the above, it is possible to realize a speed calculation that is hardly influenced by the fluctuation of the primary resistance value and its setting error. In addition, by smoothly switching the slip calculation method between the steady state and the transient state in the slip correction unit, accurate slip calculation can be performed in any state and highly accurate speed estimation is possible.

[Brief description of drawings]

FIG. 1 is a block diagram of an embodiment in which a speed sensorless speed estimation method for an induction motor according to the present invention is used in an inverter drive system, and FIG. 2 is an example of a conventional speed sensorless speed estimation method for an induction motor in a PWM inverter drive system. FIG. 3, FIG. 3 is a block diagram of the magnetic flux calculation unit of FIG. 2, FIG. 4 is a block diagram of the magnetic flux calculation unit of the present invention, and FIG. 5 is a block diagram of the slip calculation unit of the present invention. 1 ... DC power supply, 2 ... 3-phase PWM inverter 3 ... 3-phase induction motor, 4 ... Voltage detection calculation unit 5 ... Current detection calculation unit, 6 ... Flux calculation unit 7 ... Flux speed calculation unit, 8,8 '…… Slip calculation unit 9,62,66,112,122 …… Subtraction unit 10 …… Second slip calculation unit 11 …… Slip correction unit, 15 …… Slip calculation unit 16 …… Complex magnetic flux calculation unit, 61,134… … Constant multiplication unit 63 …… Integration unit 64,123 …… Secondary conversion multiplication unit 65 …… Leakage constant multiplication unit, 111,125,136 …… Integrator 113 …… Addition unit, 120 …… Voltage system magnetic flux calculation unit 121 …… Voltage drop calculation Section, 124, 142 …… Subtractor 130 …… Current system magnetic flux calculation section 131 …… Secondary constant multiplier, 132 …… Phase advancer 133 …… Multiplier, 135 …… Adder / subtractor 141 …… Error amplifier

Claims (1)

[Claims] 1. A primary voltage and a primary current of an induction motor are detected. And a secondary interlinkage magnetic flux vector from the primary voltage vector and the primary current vector A first magnetic flux calculating means for calculating By combining the output of the second magnetic flux calculating means for calculating And a magnetic flux velocity calculating means for calculating the rotational angular velocity ω ₂ of the secondary linkage magnetic flux vector from the secondary linkage magnetic flux vector from the magnetic flux calculation means, and the primary current from the detection calculation means. Slip angular velocity by calculating the squared value of the magnitude of the secondary flux linkage vector from the vector and the secondary flux flux vector from the composite flux computation means. And the slip angular velocity by calculating the inner product of the primary current vector and the secondary interlinkage magnetic flux vector. And a second slip calculating means for calculating the slip angle angular velocity ωs
A rotor for calculating the rotational angular velocity of the rotor from the difference between the rotational angular velocity ω ₂ of the secondary interlinkage magnetic flux vector from the magnetic flux velocity computing device and the slip angular velocity ω s from the correcting device. A speed sensorless speed estimation device for an induction motor, comprising an angular speed calculation means.