JP6409313B2 - Periodic disturbance automatic suppression device - Google Patents

Description

  The present invention relates to a periodic disturbance automatic suppression device that automatically suppresses torque ripple of a rotating electric machine such as a rotating electric machine such as a motor, and particularly corrects an identification model of a periodic disturbance observer as needed during suppression control. The present invention relates to the periodic disturbance automatic suppression apparatus.

  Periodic disturbance generation suppression control includes power system control in power receiving / transforming equipment, robot positioning control, dynamometer system shaft torque resonance suppression, motor housing vibration suppression (related to riding comfort of electric vehicles, elevators, etc.) Etc.), and it is desired to suppress the periodic disturbance in each of these products with high accuracy.

  For example, a motor generates torque ripple in principle, causing various problems such as vibration, noise, adverse effects on riding comfort, and electrical / mechanical resonance. In particular, in an embedded magnet type PM motor, cogging torque ripple and reluctance torque ripple are generated in a composite manner. As a countermeasure, a periodic disturbance observer compensation method has been proposed as a control method for suppressing torque ripple.

FIG. 10 is a control block diagram related to the nth-order torque ripple frequency component of the periodic disturbance observer, which is known from Patent Literature 1 and Non-Patent Literature.
Reference numeral 1 denotes a torque ripple compensation value calculation unit. The difference between the control command rn (usually 0) for the sine wave / cosine wave and the estimated values dT _A ^ n and dT _B ^ n by the periodic disturbance observer 3 is a sine / cosine wave A torque ripple compensation command Tc * n is generated by multiplying the values and adding them, and is output to the controlled object 2. In the controlled object 2, a periodic disturbance (hereinafter referred to as a periodic disturbance dTn) may occur. For example, if the object to be controlled is a motor, a torque ripple that is a disturbance synchronized with the rotational speed of a cogging torque or the like corresponds to this, which causes vibration and noise.

The periodic disturbance observer 3 suppresses the periodic disturbance dTn. By using the inverse system model of the disturbance observer for the system identification model expressed by a complex vector for each frequency component, the disturbance of the frequency to be controlled is directly estimated. To compensate.
Thereby, although it is a comparatively simple control configuration, a high suppression effect can be obtained for the contrasted frequency regardless of the order.

Concerning the acquisition of the identification model P ^ n of the system, it performs a pre-system identification on the control target plant _{Pn (= P A n + jP} B n) prior to the control, in the form of a one-dimensional complex vector (1) Express as
_{P ^ n = P ^ A n} + jP ^ B n ... (1)
However, n subscript n-order component, the variable is a complex vector both of which are expressed as _{Xn = X A n + jX B} n.

For example, when the system identification result of 1 to 1000 Hz is expressed by a complex vector every 1 Hz, the system can be expressed by a table composed of 1000 one-dimensional complex vector elements. Alternatively, the system can be expressed by formulating the identification result. In any method, the system model can be expressed with a simple one-dimensional complex vector for a specific frequency component.
Incidentally, a complex vector P ^ n mentioned in this document is not limited to the system identification model, rn, dTn, dT ^ n , Tn also be expressed as _{Xn = X A n + jX B} n.

Since the torque ripple of the motor is a disturbance periodically generated according to the rotation phase θ [rad], the control of the periodic disturbance observer 3 uses an arbitrary order n (the electric rotation frequency of the electric rotation frequency) by using a torque pulsation frequency component extraction means. It is converted to cosine coefficient T _A n and sine coefficient T _B n. The strict measurement means of the frequency component includes Fourier transform, etc., but in FIG. 10, importance is placed on simplicity, and the low pass filter G _F (s) which simplifies the Fourier transform with respect to the plant output is passed through. A frequency component to be suppressed for the periodic disturbance dTn is extracted. This is multiplied by the inverse system expressed by the reciprocal number P ^ n ^-1 of the extracted system identification model, and the periodic disturbance dTn is estimated from the difference from the control command value passed through the low-pass filter G _F (s). , periodic disturbance estimated value _{dT ^ n (= dT ^ a} n + jdT ^ B n) as the torque ripple compensation value Sappii from the output to control instruction rn to the arithmetic unit 1 inhibits the cycle disturbance dTn.

International Publication WO2010 / 024195A1

Torque ripple suppression control method for PM motor based on periodic disturbance observer using complex vector representation, IEEJ Transactions D, Vol.132, No.1.P.84-93 (2012)

  In the control based on the periodic disturbance observer, what constitutes the basis of the control and influences the control performance is the accuracy with respect to the true value of the system identification model P ^ n. In order to improve the periodic disturbance suppression performance, more accurate system identification is required. However, it is difficult to obtain an identification model with high accuracy, and it is necessary to take into consideration long-term fluctuations of the plant due to secular changes and short-term fluctuations due to sudden system fluctuations. The error of the identification model may increase the convergence time until the suppression is completed, and in the worst case, the suppression control itself becomes a disturbance, which may make the control unstable. For this reason, improvement in robustness against identification model errors is required.

  Conventionally, the adjustment of the gain section is determined by a sequence while monitoring the operation state. In this case, however, there are many adjustment parameters, and whether or not correction can be achieved depends on the skill of the adjustment / designer. In addition, it is impossible to correct the model for a part other than the sequence.

  An object of the present invention is to provide an automatic periodic disturbance suppression apparatus that reduces the adjustment parameters and does not depend on the skill of the adjustment / designer.

Claim 1 of the present invention inputs a plant disturbance generating periodic disturbance to a periodic disturbance observer, extracts a frequency component to be suppressed from the plant output, and obtains a periodic disturbance estimated value for the extracted frequency component. in those calculated, it controls the plant based on the difference between the calculated period estimated disturbance value d ^ _n and the control command r _n,
Wherein the control command r _n is a constant value, assumed to be constant value having no periodicity of _{r n = 0 [n> 0} ], with respect to the time difference between the times t _1, t _2, during operation of the plant Assuming that the fluctuation cycle of the generated plant fluctuation / disturbance fluctuation is much longer, based on this, the periodic disturbance and the plant are assumed to be constant regardless of time,
Calculate the time variation of the plant output value y _n and the estimated output y ^ _n of the identification model P ^ _n when the estimated periodic disturbance estimated value d ^ _n is the disturbance. Based on the calculated time change amount of the output value y _{n and} the time change amount of the estimated output y ^ _n , an error P ^ref _n with the identification model is calculated and output, or the plant and time change amount of the output value y _n, wherein calculating a time variation of the periodic disturbance estimated value d ^ _n, and time change amount of the output value y _n of the calculated, the periodic disturbance estimated value d ^ _n times Model correction means for calculating and outputting a system model P ^^ _n estimated from the time difference based on the amount of change;
A low-pass filter that filters the coefficients of the real / imaginary part of the complex vector with respect to the output of the model correction unit;
It said periodic disturbance observer is provided with a memory function for storing the computed identified model _P'n as the identification model P ^mem _n,
The periodic disturbance observer performs suppression control for each of a plurality of operating points within a range in which the plant operates, and stores the identification model at the time when suppression control is completed in a memory, and is calculated by the model correction unit, A system model P ^^ _n estimated from an error P ^ref _n or time difference from the identification model filtered by the low-pass filter; an identification model P ^mem _n stored in the memory function; Using the filter function LPF of the low-pass filter, the final identification model P ′ _n (= P ^mem _n · LPF (P ^ref _n )) or P ′ _n (= P ^mem _n · LPF (P ^^ _n )) And a process for estimating and outputting a final periodic disturbance estimated value d ^ _n using the obtained final identification model P ′ _n . Is.

  According to a second aspect of the present invention, an interpolation unit is provided in the model correction unit, and the identification model is interpolated to the periphery of the target frequency.

Claim 3 of the present invention, the model correction means for calculating, model P ^^ _n estimated from the time difference, the time t _1, y _n .t ₁ the n-th output value of the plant at t _2, y _n .t ₂ and the periodic disturbance estimated values at times t ₁ and t ₂ as d ^ _n .t ₁ and d ^ _n .t ₂ are obtained by the following equations .
P _n = − (( y _n .t ₂ −y _n .t ₁ ) / (d _{n n} .t ₂ −d _n n.t ₁ ))
(= P ^^ _n )
However, P _n is a plant, and each value is a complex number. Claim 4 of the present invention is that the error P ^ref _n with the identification model calculated by the model correcting means is the n-th order output value of the plant at times t ₁ and t ₂ . Y _n .t ₁ and y _n .t _2, and the estimated output of the identification model P ^ _n at time t ₁ and t ₂ when the periodic disturbance estimated value d _n is a disturbance, is y _n . t ₁ and y ^ _n .t ₂ are obtained by the following equation .
P _n = − (( y _n .t ₂ −y _n .t ₁ ) / (y _{n n} .t ₂ −y _n n.t ₁ )) · P _n
= P ^ref _n・ P ^ _n
However, P _n is a plant and each value is a complex number. Claim 5 of the present invention is characterized in that a plurality of periodic disturbance observers having the model correcting means are connected in parallel to obtain a plurality of estimated periodic disturbance values. Is.

  As described above, according to the present invention, it is possible to reduce the adjustment parameters and perform highly accurate periodic disturbance suppression control without depending on the skill of the adjustment / designer. In addition, a memory function is provided in the periodic disturbance suppression control to learn the identification model correction amount, the identification model is saved, and the saved identification model is output via the periodic disturbance observer, so that a more accurate period can be obtained. Disturbance suppression control can be performed, and if the frequency is the same again or a frequency in the vicinity thereof, the correction operation is unnecessary or the correction time can be shortened.

The period disturbance observer control block diagram of this invention. Flow chart of erroneous learning prevention processing. FIG. 3 is a basic periodic disturbance observer control block diagram. The period disturbance observer control block diagram. The period disturbance observer control block diagram which shows other embodiment. The period disturbance observer control block diagram which shows other embodiment of this invention. Interpolation process explanatory drawing. Interpolation process explanatory drawing. Interpolation process explanatory drawing. The control block diagram of the whole period disturbance observer.

  The present invention inputs an output of a controlled object that generates a periodic disturbance to a periodic disturbance observer, calculates a periodic disturbance estimated value, and controls the controlled object based on a difference between the calculated periodic disturbance estimated value and a control command. The periodic disturbance observer is provided with a model correction means, and the model correction means corrects the system model based on the identification model error calculated using the time difference between the output value of the periodic disturbance observer and the detected value of the control target. The details will be described below with reference to the drawings.

FIG. 3 shows a control configuration by the periodic disturbance observer of the present invention. The symbols in FIG. 3 are as follows.
P _n : Plant (control target), d ^ _n : Periodic disturbance estimated value,
y _n : Control target output (corresponding to T _A n and T _B n in FIG. 10),
r _{n: n} The following control command, d _{n: n} Tsugigairan value,
d ^ _n .t: n-th order compensation command at time t (corresponding to dT _{A ^} n and dT _{B ^} n in FIG. 10),
P ^ _n : Identification model, y ^ _n : Estimated output of the plant of the identification model P ^ _n when the disturbance is d ^ _n .
PDO: Periodic disturbance observer, y _n .t: n-order output at time t, value in s region indicates time t after the comma of the subscript.

Each state equation at time t ₁ is expressed by equations (2) and (3). It is assumed that the command value is constant at t = t _{0 to} t _n and is a steady value having no periodicity of r _n = 0 [n> 0]. Each value is a complex number.
(D _n −d ^ _n .t ₁ ) × P _n = y _n .t ₁ (2)
_{d ^ n .t 1 × P ^} n = y ^ n .t 1 ...... (3)
Similarly, the respective state equations at time t ₂ are represented by equations (4) and (5).
(D _n −d ^ _n .t ₂ ) × P _n = y _n .t ₂ (4)
d ^ _n .t ₂ × P ^ _n = y ^ _n .t ₂ (5)
The difference between the times is obtained from the equations (2) to (5), and are defined as equations (6) and (7).
Here, the disturbance d _n and plant P _n is assumed to be constant regardless of time, it can be excluded steady-state term of the disturbance Taking the time difference. In addition, regarding the acquisition time difference between the times t ₁ and t ₂ , a number [ms] is set as a target value. Even if plant fluctuation / disturbance fluctuation occurs during operation, it is assumed that this fluctuation cycle is very long with respect to the time difference between times t ₁ and t ₂ , and the above assumption is considered to be valid.
_{- (d ^ n .t 2 -d} ^ n .t 1) × P n = y n .t 2 -y n .t 1 ...... (6)
_{(D ^ n .t 2 -d ^} n .t 1) × P ^ n = y ^ n .t 2 -y ^ n .t 1 ...... (7)
Equation (8) is obtained from Equation (6), and Equation (9) is obtained by substituting Equation (7) into Equation (8).
Here, the system model was estimated from the difference result put P ^^ _n, the error of the identified model and P ^ref _n.
P _n = − (( y _n .t ₂ −y _n .t ₁ ) / (d _{n n} .t ₂ −d _n n.t ₁ ))
(= P ^^ _n ) (8)
P _n = − (( y _n .t ₂ −y _n .t ₁ ) / (y _{n n} .t ₂ −y _n n.t ₁ )) · P _n
= P ^ref _n・ P ^ _n (9)
From the above, it is possible to estimate the system model by using the difference between the states at times t ₁ and t ₂ . This is corrected by feeding back to the identification model P ^ _n , and finally, a highly accurate disturbance estimated value d ^ _n can be obtained.

FIG. 4 shows a control block diagram of a periodic disturbance observer provided with a model correcting unit, and the model correcting unit 30 performs a correction calculation according to the equation (9). That is, the model correction means 30 performs an operation corresponding to the Enter (9) the identification model P ^ by _n n The following plant output compensation value y ^ _n and the plant output y _n. Since the calculation of equation (9) is a differential calculation, noise is included. For this reason, the output of the model correction means 30 is passed through the low-pass filter 31 to remove noise. For other operations, the same basic operations as in FIG. 10 are performed.

  Therefore, since the periodic disturbance observer PDO can output correction command values for the phase error and the gain error due to the corrected identification model error to correct the identification model at any time during the torque ripple suppression control, the adjustment parameter is decreased, It is possible to suppress the periodic disturbance with higher accuracy without depending on the skill of the adjustment / designer.

In FIG. 4, the internal model of the periodic disturbance observer PDO is corrected by the model correction unit 30 having a calculation unit corresponding to the equation (9). In FIG. 5, the calculation using the system model P ^^ _n estimated from the difference result of the equation (8) is performed by the model correction means 30a. If P ^^ _n is used, disturbance can be suppressed even if the identification model P ^ _n is completely unknown. In this case, the model correction function is always enabled, and the internal model is controlled by using the system model P ^ _n by the model correction means 30a obtained by the equation (8).

Therefore, FIG. 5 has the same effect as FIG. 4 and is particularly effective when the system identification cannot be performed in advance because the system model can be estimated from the detected value of the time difference between times t ₁ and t _2. It will be. FIGS. 4 and 5 show the means for adaptively correcting the identification model of the system of periodic disturbance observer PDO at a certain frequency.

  In the present invention shown in FIG. 1, a memory function is further added, and a correction value obtained by the model correction means 30 (30a) shown in FIGS. 4 and 5 is fed back to the periodic disturbance observer PDO through the low-pass filter LPF to finalize the correction. The value is calculated and stored in the memory function according to the rotational frequency nω of the plant.

In FIG. 1, SW is a switch means, 32 is a memory function, and a final identification model P ′ _n described later is stored in the memory function 32 in a state corresponding to the rotational frequency nω of the plant. The other configuration is the same as in FIG. 4 or FIG.
The periodic disturbance observer PDO uses the identification model error P ^ref _n (or P ^ _n ) corrected by the model correction means 30 (or 30a) and the identification model P ^mem _n stored in the memory function to obtain the equation (10). To calculate the final identification model P ′ _n ,
The periodic disturbance observer PDO uses this to estimate and output a final periodic disturbance estimated value d ^ _n .
P ′ _n = P ^mem _n · LPF (P ^ref _n ) (10)
Here, the initial value of P ^mem _n is P ^ _n , the LPF filters the coefficients of the real part / imaginary part of the complex vector, and the output is a complex vector by the real part / imaginary part after filtering. And

The switch means SW is turned on / off at the timing when the identification model P ′ _n calculated by the periodic disturbance observer PDO is stored in the memory function 32. The storage timing is the time when the final identification model P ′ _n is processed and the periodic disturbance is sufficiently suppressed, and a new identification model P ′ _n is obtained simultaneously with the completion of the suppression control. The identification model P ′ _n becomes valid for a certain time from the state where the model error exists, and the switch means SW is turned on with the model correction and the periodic disturbance suppression sufficiently converged, and the identification model P ′ _n at that time is stored in the memory function 32. And the memory value P ^mem _n is updated. When the memory storage is completed, the switch means SW is turned off.

In the case of storing in the memory function 32, the operation point is repeatedly executed in the entire range in which the plant operates or in a certain range. FIG. 2 shows a processing sequence for preventing erroneous learning. In step S1, the operating point is moved. When the operation is completed in step S2, the switch means SW is turned on in step S3 to perform suppression control. The identification model _P'n in the memory function 32 When suppression control is complete in step S4 and stored (S5), and terminates the suppression control by turning off the switch means SW in step S6, and ends when all operating points is completed .

  When a learning function is added as shown in FIG. 1, interpolation between the case where the learning point is a one-point frequency and a specific frequency range is conceivable. 7 to 9 are explanatory diagrams of the interpolation processing.

Assume that the frequency response graph relating to P _A (P _B ) as shown in FIG. 7 is owned as the identification model P ^ _n in the initial state. FIG. 8 shows an example of an identification model as a result of learning the identification model error in FIG. Since the periodic disturbance suppression control is performed only at a specific frequency point a or b point associated with model fluctuation, only a specific frequency at which a model error can be learned is obtained. For this reason, the learning result cannot be applied if even a slight shift from the learned frequency occurs. When the controlled object is a variable speed application such as a motor, relearning is required even with a minute speed (frequency) fluctuation.

  As a countermeasure, interpolation means is provided in the model correction means. Interpolation by the interpolation means is performed by any known method. For example, as shown in FIG. 9, the learning point a or b is connected to the frequency to be interpolated by a straight line and linear approximation is performed. As a result, the learning result at a specific frequency is affected to the frequency domain, so that the stability of the suppression control and the time required for learning can be shortened even when the frequency fluctuates.

FIG. 6 is constructed in parallel and simultaneously with respect to each order for suppressing the control system of the apparatus shown in FIG. That is, N periodic disturbance observers PDO with model error correction are provided, the identification model error is estimated for the n-th order periodic disturbance component to be estimated, and the system disturbance model is corrected based on the estimation error. the value _{d ^ n - 1 ~d ^ n} - n is output to the torque ripple compensation value calculation unit 1 (in FIG. 10).

DESCRIPTION OF SYMBOLS 1 ... Torque ripple compensation value calculating part 2 ( _Pn ) ... Control object (plant)
3 (PDO) ... cycle disturbance observer 30 ... model correcting means 31 ... low-pass filter 32 ... memory function d ^ _n ... period disturbance estimated value P ^ _n ... identification model y _n ... control object output

Claims (5)

The output of the plant that generates periodic disturbance is input to the periodic disturbance observer, the frequency component to be suppressed is extracted from the output of the plant, the estimated periodic disturbance is calculated for the extracted frequency component, and the calculated periodic disturbance in controls the plant based on the difference between the estimated value d ^ _n and the control command r _n,
Wherein the control command r _n is a constant value, assumed to be constant value having no periodicity of _{r n = 0 [n> 0} ], with respect to the time difference between the times t _1, t _2, during operation of the plant Assuming that the fluctuation cycle of the generated plant fluctuation / disturbance fluctuation is much longer, based on this, the periodic disturbance and the plant are assumed to be constant regardless of time,
Calculate the time variation of the plant output value y _n and the estimated output y ^ _n of the identification model P ^ _n when the estimated periodic disturbance estimated value d ^ _n is the disturbance. Based on the calculated time change amount of the output value y _{n and} the time change amount of the estimated output y ^ _n , an error P ^ref _n with the identification model is calculated and output, or the plant and time change amount of the output value y _n, wherein calculating a time variation of the periodic disturbance estimated value d ^ _n, and time change amount of the output value y _n of the calculated, the periodic disturbance estimated value d ^ _n times Model correction means for calculating and outputting a system model P ^^ _n estimated from the time difference based on the amount of change;
A low-pass filter that filters the coefficients of the real / imaginary part of the complex vector with respect to the output of the model correction unit;
It said periodic disturbance observer is provided with a memory function for storing the computed identified model _P'n as the identification model P ^mem _n,
The periodic disturbance observer performs suppression control for each of a plurality of operating points within a range in which the plant operates, and stores the identification model at the time when suppression control is completed in a memory, and is calculated by the model correction unit, A system model P ^^ _n estimated from an error P ^ref _n or time difference from the identification model filtered by the low-pass filter; an identification model P ^mem _n stored in the memory function; Using the filter function LPF of the low-pass filter, the final identification model P ′ _n (= P ^mem _n · LPF (P ^ref _n )) or P ′ _n (= P ^mem _n · LPF (P ^^ _n )) And a process for estimating and outputting a final periodic disturbance estimated value d ^ _n using the obtained final identification model P ′ _n . Periodic disturbance automatic suppression device.   2. The periodic disturbance automatic suppression device according to claim 1, wherein an interpolation unit is provided in the model correction unit, and the identification model is interpolated to the periphery of the target frequency. The model P ^ _n calculated from the time difference calculated by the model correcting means uses the plant n-order output values at times t ₁ and t ₂ as y _n .t ₁ and y _n .t ₂ , and the time t ₁ The periodic disturbance automatic suppression device according to claim ₁ , wherein the estimated periodic disturbance values at t ₂ and t ₂ are determined as d ^ _n .t ₁ and d ^ _n .t _{2 according} to the following equation.
P _n = − ((y _n .t ₂ −y _n .t ₁ ) / (d _n n.t ₂ −d _n n.t ₁ ))
(= P ^^ _n )
Where P _n is the plant and each value is a complex number The model correction means for calculating error P ^ref _n the identification model, the time t _1, the n-th output value of the plant at t ₂ and _{y n .t 1, y n .t} 2, the time t _1, t in _2, the estimated output of the plant of the identification model P ^ _n when a disturbance of the periodic disturbance estimated value d ^ _n as _{y ^ n .t 1, y ^} n .t 2, and characterized in that calculated by the following equation The periodic disturbance automatic suppression device according to claim 1 or 2.
P _n = − ((y _n .t ₂ −y _n .t ₁ ) / (y _n n.t ₂ −y _n n.t ₁ )) · P _n
= P ^ref _n・ P ^ _n
Where P _n is the plant and each value is a complex number   The periodic disturbance automatic suppression device according to any one of claims 1 to 4, wherein a plurality of periodic disturbance observers having the model correcting means are connected in parallel to obtain a plurality of periodic disturbance estimated values.

Images