Induction motor drive unit, motor drive system, and elevating system

[0025] (first embodiment)

[0026] As shown in Equation (1), when using lifting equipment such as a crane to hoist the cargo at an acceleration α, the required torque τm depends on the acceleration α and the weight m·g of the cargo (cargo). In addition, in formula (1), the cargo is hoisted by a pulley with radius r, J is the total moment of inertia including the motor or mechanical system (pulley, etc.), ω is the rotation speed of the motor, m is the mass of the cargo, and g is acceleration of gravity.

[0027] M·α·r=J·dω/dt=τm-m·g·r (1)

[0028] At this time, the maximum value of torque τm is determined by the capacity of the motor. When m·g·r is large, the acceleration α of formula (1) is negative or zero, and the cargo cannot be lifted. Furthermore, even if the acceleration α is positive, if the velocity change rate dω * /dt ratio (maximum torque τm max -m·g·r)/J is large, the motor cannot follow the instruction and the hoisting fails.

[0029] FIG. 1 is a configuration diagram of a motor drive system according to an embodiment of the present invention. The motor drive system of this embodiment includes an induction motor drive device 90 and an induction motor 16. The induction motor drive device 90 includes a control unit 91, a power conversion unit 15, and a current detection unit 18, and is a speedless motor that does not detect the rotational speed of the induction motor 16. Sensor control system. Here, the induction motor 16 rotates at a rotational speed ωr obtained by subtracting the slip frequency ωs from the frequency of the three-phase input voltage (primary frequency ω1). The power conversion unit 15 includes a plurality of switching elements such as IGBT (Insulated-Gate-Bipolar-Transistor), and drives the induction motor 16 based on a PWM (Pulse-Width-Modulation) signal generated by the control unit 91 . The current detection unit 18 is a current sensor that detects the motor current I flowing through the induction motor 16, and detects actual currents of the U-phase and W-phase. In addition, in FIG. 1 , the description of the three-phase AC power supply 92 and the AC current conversion unit 94 shown in FIGS. 10 and 11 is omitted.

[0030] The control unit 91 mainly includes: a speed command value calculation unit 1 , an excitation current command calculation unit 2 , a slip frequency calculation unit that is a slip frequency estimated value calculation unit 3 , a maximum torque generation slip calculation unit 4 , and an acceleration determination unit 5 , a speed change rate calculation unit 6, a speed change rate correction unit 7, a speed calculation unit that is a speed estimation calculation unit 8, a primary frequency command calculation unit 9, a d-axis current control unit 10, a q-axis current control unit 11, and a phase calculation unit 12 , a voltage command calculation unit 13, a coordinate conversion unit 14, a power conversion unit 15, a coordinate conversion unit 17 and a speed control unit 19, and their functions are realized by CPU, ROM, RAM and programs.

[0031] First, sensorless vector control of the induction motor 16 will be described. In addition, it is defined as the vector control with the motor flux axis as the d-axis, and the axis electrically perpendicular to the d-axis as the q-axis.

[0032] The speed command value calculating unit 1 outputs the speed command ωra which is the reference of the rotation speed of the induction motor 16 *. When accelerating, the speed command ωra * Change with time at the initially set time rate of change (acceleration rate). The field current command computing unit 2 calculates the field current command Id *.

[0033] The speed estimation calculation unit 8 calculates the speed estimation value ωr̂ based on the q-axis current Iq calculated based on the IqFB output from the coordinate conversion unit 17 and the like.

[0034] ωr^=1/(1+T1·s)·(L2 * /M * )·(1/φ2d * )·(r1 * ·I

[0035] q * +ω1 * ·(M * /L2 * )·φ2d * +ΔVq-(r1 * +r2' * +Lσ * ·s)·Iq)

[0036] The speed control unit 19 inputs the slave speed command ωr * The value after subtracting the estimated speed value ωr^, according to the estimated speed value ωr^ and the speed command ωr * Consistent way to calculate the torque current command Iq *. The d-axis current control unit 10 follows the Id detection value (IdFB) and Id * The d-axis voltage correction value Δd is calculated in a consistent manner. In addition, the q-axis current control unit 11 follows the Iq detection value (IqFB) and the Iq * The q-axis voltage correction value Δq is calculated in a consistent manner.

[0037] The primary frequency command calculation unit 9 compares the speed estimated value ωr^ and the speed command ωr * are added together to calculate the primary frequency ω1 of the voltage input to the induction motor 16 *. The phase calculation unit 12 performs a primary frequency ω1 * Integrate to calculate the phase command θ *.

[0038] The voltage command calculation unit 13 adopts the primary frequency ω1 * Computing the d-axis voltage command Vd as shown in equation (2) and equation (3) with various current commands, the output of the d-axis current control unit 10, and the output of the q-axis current control unit 11 * , q-axis voltage command Vq *.

[0039] Vd * =r1 * ·Id * -ω1 * · Lσ * · Iq * +Δd (2)

[0040] wxya * =r1 * · Iq * +ω1 * · Lσ * ·Id * +ω1 * ·(M * /L2 * )·Φ2d * +Δq(3)

[0041] Here, r1 * , Lσ * , M * , L2 * 、Φ2d * They are the primary resistance setting value, (primary + secondary) leakage inductance setting value, mutual inductance setting value, secondary inductance setting value and secondary side d-axis magnetic flux command value of the induction motor respectively. Δd and Δq are outputs of the aforementioned d-axis current control unit 10 and q-axis current control unit 11 .

[0042] The coordinate conversion unit 14 adopts the d-axis voltage command Vd * , q-axis voltage command Vq * and the phase command θ * , converted into 3-phase AC voltage commands Vu, Vv, Vw controlled by PWM.

[0043] The coordinate conversion unit 17 adopts the AC current detection value and the phase command θ from the current detection unit 18 * , converted into 2-phase DC current IdFB, IqFB. As mentioned above, in the existing vector control, the specified ωr is tracked according to the actual speed * way to control current and voltage.

[0044] Slip frequency estimated value calculation unit 3 performs estimation calculation of slip frequency estimated value ωs ̂ of induction motor 16 . For example, the slip frequency ωs is estimated as follows.

[0045] Inside the induction motor 16, the relationship of Expression (4) and Expression (5) is established. Values ​​without an asterisk (*) are actual values ​​on the induction motor side, and Φ2q is the q-axis magnetic flux on the secondary side.

[0046] Vd=r1·Id-ω1·Lσ·Iq-ω1·(M/L2)·Φ2q (4)

[0047] Vq=r1·Iq+ω1·Lσ·Id+ω1·(M/L2)·Φ2d (5)

[0048] Through the current control system, it is controlled as Id=Id * , Iq=Iq * , in the low-speed region, the second term on the right side of formula (3) and formula (5) is smaller than the third term on the right side respectively, and can be ignored. In addition, r1, ω1, M, and L2 are assumed to be set values, since Vd is generally * = Vd, Vq * =Vq, so when the d-axis magnetic flux estimated value Φ2d^ and the q-axis magnetic flux estimated value Φ2q^ are obtained, it is shown in formula (6) and formula (7).

[0049] Φ2d^=(Δq+ω1 * ·(M * /L2 * )·Φ2d * )/(ω1 * ·(M * /L2 * ))(6)

[0050] Φ2q^=-Δd/(ω1 * ·(M * /L2 * )) (7)

[0051] In the manner described above, the slip frequency command ωs^ is obtained as follows. Generally, Φ2q and Φ2d satisfy formula (8) and formula (9). T2 is the secondary time constant of the motor, and s is the differential operator.

[0052] Φ2d=(M·Id+ωs·T2·Φ2q)/(1+T2·s) (8)

[0053] Φ2q=(M·Iq-ωs·T2·Φ2d)/(1+T2·s) (9)

[0054] Here, in order to generate a high torque at a low speed, when the d-axis current Id of the rated current (approximately twice or more than the rated excitation current) flows and is controlled so that Iq=0, the constant solution of ωs is expressed as Equation (9) As well as formula (10).

[0055] ωs=-1/T2·Φ2q/Φ2d (10)

[0056] Therefore, ωs is calculated according to the formula (6), the formula (7) and the formula (10), and the slip frequency command ωs^ is set.

[0057] The torque τm is shown in formula (11). In addition, P in the formula (11) is the number of poles of the induction motor 16 .

[0058] m=3·(P/2)·(M/L2)·(Φ2d·Iq-Φ2q·Id) (11)

[0059] Here, similarly to the above, when the d-axis current Id is increased more than normal, and the torque τm is expressed by the formula (12) when Iq=0 is controlled.

[0060] m=3·(P/2)·(M/L2)·(-Φ2q·Id) (12)

[0061] Since the d-axis current Id is a fixed value, from Equation (12), Φ2q is negative, and the maximum torque occurs when the absolute value is maximum. Furthermore, when a constant solution is introduced into Equation (8) and Equation (9) with respect to Φ2d and Φ2q, it becomes like Equation (13) and Equation (14).

[0062] Φ2d=M·Id/(1+(ωs·T2) 2 ) (13)

[0063] Φ2q=-ωs·T2·M·Id/(1+(ωs·T2) 2 ) (14)

[0064] Here, since the rated exciting current or more flows through Id, the magnetic flux increases in the magnetic flux saturation region, and the magnetic flux becomes γ·Φ0. In addition, Φ0 represents the rated magnetic flux of the motor, and γ is generally about 1.1 to 1.3.

[0065] At this time, Φ2d and Φ2q satisfy Expression (15), and have a relationship like Expression (16).

[0066] ( Φ 2 d 2 + Φ 2 q 2 ) = γ · Φ 0 - - - ( 15 )

[0067] M · Id / ( 1 + ( ωs · T 2 ) 2 ) = γ · Φ 0 - - - ( 16 )

[0068] According to formula (16) and formula (14), Φ2q is shown as formula (17).

[0069] Φ 2 q = - ωs · T 2 · γ · Φ 0 / ( 1 + ( ωs · T 2 ) 2 ) - - - ( 17 )

[0070] When the load is lifted by a lifting device such as a crane, the primary frequency ω1 is increased and the supplied d-axis current Id is above the rated value, Φ2q obeys equation (17) and increases monotonously with respect to ωs. However, when the primary frequency ω1 (equal to ωs from the static state of the motor to the start of rotation) continues to increase, the magnetic saturation is released, and Φ2q follows the formula (14).

[0071] figure 2 A graph showing the absolute value (normalized by a predetermined value) of Φ2q to ωs (normalized by a predetermined value) on the horizontal axis. The position where the curve of formula (14) intersects with the curve of formula (17) is saturated, and Φ2q, that is, the torque becomes the maximum. From Equation (17), the slip frequency ωsmax at which the torque becomes maximum is expressed by Equation (18), and the maximum torque τmmax at this time is expressed by Equation (19). In formula (18), Id can also adopt IdFB and Id * To obtain, γ can also adopt a value of about 1.1 to 1.3.

[0072] ωs max = ( ( M · Id / γ / Φ 0 ) 2 - 1 ) / T 2 - - - ( 18 )

[0073] τm max = 3 ( P / 2 ) · ( M / L 2 ) · γ · Φ 0 · ( Id 2 - ( γ · Φ 0 / M ) 2 ) - - - ( 19 )

[0074] The maximum torque generation slip calculating unit 4 calculates the slip frequency smax at which the induction motor 16 generates the maximum torque based on Equation (18). In addition, when the d-axis current Id is controlled at a fixed value equal to or higher than the rated excitation current, and the q-axis current Iq is controlled at a fixed value of 0, the current value is always fixed regardless of the size of the load. In addition, the fixed value refers to a value that is fixed within a range including noise components accompanying measurement and the like.

[0075] The acceleration determination unit 5 a shown in FIG. 3 is one form of the acceleration determination unit 5 ( FIG. 1 ), and includes a threshold calculation unit 30 , a comparison unit 31 , a timer 32 , and a limiter 33 .

[0076] The threshold calculation unit 30 calculates a threshold ωsmaxTH corresponding to the slip frequency smax (for example, ωsmax is multiplied by α (α≦1)). The comparator 31 compares the slip frequency ωs^ with the threshold ωsmaxTH, and outputs when ωs^>ωsmaxTH. The timer 32 receives a signal from the comparator 31, and outputs the elapsed time from that point. In addition, the timer 32 is reset (reset) when the comparison part 31 outputs ωs^≦ωsmaxTH. That is, when the state of ωŝ>ωsmaxTH continues and the output of the timer 32 exceeds a predetermined threshold, the acceleration determination unit 5 determines that the hoisting cannot be performed, and the limiter 33 outputs a signal.

[0077] In addition, as the threshold value of the limiter 33, the change in the slip rate is set to be smaller than the quadratic time constant of the motor reflected in the torque (for example, 0.15 seconds), and larger than the calculation cycle of the controller (for example, 0.0001 seconds). value. In addition, instead of the timer 32, other elapsed time measuring means such as an integral circuit may be employed.

[0078] Furthermore, the speed change rate calculation unit 6 receives the signal of the limiter 33 and outputs a negative predetermined value as the speed change rate. In addition, the signal from the limiter 33 is input to the alarm device 35, and an alarm is generated. For this alarm, an electric signal, sound, light, vibration, etc. are used. In addition, a braking device (not shown) may be used simultaneously to forcibly stop the rotation of the induction motor 16 .

[0079] Returning to Fig. 1 again, the speed change rate correction unit 7, for the input speed command value ωra *, correct along the output of the speed change rate calculation unit 6, and output the speed command value ωr *. For example, the speed change rate correcting unit 7 corrects the speed change rate to a negative value and decelerates the induction motor 16 when the output of the speed change rate calculation unit 6 becomes a negative value according to conditions.

[0080] In addition, in a state where the signal from the limiter 33 ( FIG. 3 ) does not exist, the speed change rate calculation unit 6 outputs the initially set speed change rate value. Therefore, the ωra performed by the speed change rate correction unit 7 * Correction of the rate of change is not performed.

[0081] As described above, in the present embodiment, the acceleration determination unit 5 detects that the state in which the estimated slip value ωs^ exceeds the slip frequency ωsmaxTH at which the maximum torque is generated continues for a predetermined time. At this time, the speed change rate calculation unit 6 Set the speed command value ωr * The rate of change of is corrected to be negative, and the induction motor 16 is decelerated. That is, the motor drive system can judge in advance that the hoisting acceleration is impossible without malfunctioning, and if the load can be taken off stably and slowly, it can be stopped.

[0082] In addition, in lifting equipment such as a crane, there are cases where a load of a certain weight is lifted, put down at another location, and then a load of a different weight is lifted. Even when the maximum load can be assumed, the load of the induction motor 16 varies every time it increases, and even when an unassumed torque is applied, the acceleration rate corresponding to the load can be set, and the difficulty of acceleration can be avoided. Happening. In addition, the hoisting efficiency can be maximized by using the maximum torque generation slip as a reference.

[0083] (second embodiment)

[0084] In the first embodiment, the occurrence of the maximum torque is detected based on the slip frequency ωs, but if figure 2 As shown, the determination that the slip frequency ωs exceeds ωsmax can also be determined based on the secondary side q-axis magnetic flux φ2q. As shown in FIG. 4 , the acceleration determination unit 5b of the second embodiment corresponds to the acceleration determination unit 5 ( FIG. 1 ), and includes a threshold calculation unit 42 and a comparison unit 43, and inputs the φq estimation calculation unit 40 as a q-axis magnetic flux calculation unit. and the output signal of the maximum torque generation φq calculation unit 41 as the maximum torque generation q-axis magnetic flux calculation unit.

[0085] A q-axis magnetic flux estimated value φ2q^ corresponding to ωs^ is calculated in the φq estimation calculation unit 40 , and a q-axis magnetic flux φ2qmax corresponding to ωsmax is calculated in the maximum torque generation φq calculation unit 41 . The threshold calculation unit 42 calculates a threshold φ2qmaxTH corresponding to φ2qmax (for example, φ2qmax is multiplied by α (α≦1)). At this time, φ2q^, φ2qmax, and φ2qmaxTH are negative. Then, when φ2q^ is smaller than a predetermined value φ2qmaxTH or the absolute value of φ2q^ exceeds the absolute value of φ2qmaxTH, the comparison unit 43 determines that acceleration is impossible and outputs a signal. Furthermore, similarly to the first embodiment, deceleration is performed by the speed change rate calculation unit 6 according to the signal output from the comparison unit 43 , and an alarm command by the alarm device 35 is performed.

[0086] As described above, in this embodiment, by using the q-axis magnetic flux calculation value to detect the overslip, it is possible to prevent the load from falling, similarly to the first embodiment.

[0087] (third embodiment)

[0088] Differences between the third embodiment and the other embodiments will be described.

[0089] The acceleration determination unit 5c shown in FIG. 5 corresponds to the acceleration determination unit 5 (FIG. 1), but the q-axis magnetic flux estimation value φ2q^ may be calculated in the q-axis magnetic flux calculation unit, that is, the φq estimation calculation unit 40 instead. ωs^, in the maximum torque generation q-axis magnetic flux calculation unit 41 that is the maximum torque generation φq calculation unit, the q-axis magnetic flux φ2qmax corresponding to ωsmax is calculated.

[0090] The comparison unit 53 outputs a signal when φ2q^ is smaller than the output φ2qmaxTH of the threshold calculation unit 42 . The timer 54 outputs the time from the time when the above-mentioned signal was received. That is, when the state of 0>φ2qmaxTH>φ2q^ continues and the output of the timer 54 exceeds a predetermined threshold, the acceleration determination unit 5c determines that acceleration is impossible, and the comparison unit 53 outputs a signal. Furthermore, deceleration is performed by the speed change rate calculation part 6 according to the same signal as that of the first embodiment, and an alarm command from the alarm device 35 is performed.

[0091] As described above, in this embodiment, the overslip is detected using the calculated value of the q-axis magnetic flux, and the falling of the cargo with fewer malfunctions can be prevented by detecting the falling of the cargo based on the time of the over-slip state.