JP2663387B2 - Stabilization feedback control method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial Application Field] The present invention particularly relates to a technique for applying a robust state estimation of a feedback control system and correcting the directly detected state quantity obtained from a sensor or the like. The present invention relates to a stabilization feedback control method for performing stabilization control.

[Conventional technology]

This will be described with reference to FIG. 3 showing an example of a general motor speed control system.

FIG. 3 shows a conventional example of a stabilizing feedback control device, wherein 1 is a PI control device, and 2 is a control object. Here, R and Y are set inputs and state quantities, respectively. Therefore, in the example of the speed control system, R is a speed command and Y is a speed detection output. K _T is a torque generation coefficient.

That is, the speed control system is stabilized by applying the setting input R and the deviation e of the state quantity Y to the control target through the PI compensation element. As the general stabilization adjustments, by increasing the proportional gain K _P as the inertia J is large, the integral gain K ₁ accordingly also achieved by increasing.

Thus, as shown in the figure, normally, a PI compensation element is inserted in series, and the proportional gain K _P and the integral gain K ₁ according to the control target are set.
Is adjusted to achieve stabilization.

[Problems to be solved by the invention]

In this kind of conventional technology, the speed detection method as a response amount is generally a pulse generator (PG) which is a sensor.
The pulse signals from the above are counted within a certain period of time, and the speed detection value is obtained from the period of time.

Therefore, such a conventional technique has the following problems.

First, for example, when the speed becomes extremely low, the pulse signal interval becomes long, and it becomes impossible to calculate and update the speed detection value within a certain period of time. As a result, the speed information becomes information having a long sampling time, and the control performance is deteriorated. Come.

Secondly, if the pulse signal does not show a normal speed signal due to some mechanical or electrical situation, the control system becomes abnormal because it is used as feedback, giving a shock to the controlled object, and in some cases, Overspeed can be very dangerous.

As a countermeasure against this, configure an observer using the mathematical model of the control target together with the state quantities directly detected by sensors such as PG, and calculate and use the estimated state quantities similar to the directly detected state quantities. Thereby, the first and second problems can be solved. However, a general observer lacks so-called robustness because the parameter to be controlled is strongly dependent on the parameter, and the calculated estimated state quantity often differs greatly from the true value.

[Means and actions for solving the problem]

The present invention has an equivalent disturbance compensation function capable of minimizing the effects of parameter fluctuations and load fluctuations of the controlled object even when they occur, that is, the input amount of the controlled object and the state amount of the controlled object between the input and the state. By returning the difference from the quantity multiplied by the inverse function of the mathematical model to the input of the controlled object as an equivalent disturbance, all parameters of the controlled object can be described as nominal values. Is calculated based on the basic state of the art which always minimizes the difference from the directly detected state quantity. This will be further described with reference to the drawings.

Here, the present applicant has previously proposed Japanese Patent Application No. 1-159863, "Multifunctional control device".

In this proposal, the equivalent disturbance compensator is described as an example of a speed control system.

That is, FIG. 4 shows an equivalent disturbance compensator according to the present invention, and 3 is an equivalent disturbance compensator.

FIG. 5 shows a portion to be controlled after equivalent disturbance compensation.

In FIG. 4, an equivalent disturbance is calculated by utilizing information of a command T ^* (torque command in the case of speed control) and a speed detection output Y, and is added to the command T ^* . Here, T _L (S) is a load disturbance, and ω (S) is a rotation speed.

Then, by performing compensation by the equivalent disturbance compensator 3 shown in the drawing, K _T , the control target 2 and the _TL (S) shown in FIG.
As shown in FIG. 5, all parameters can be represented by nominal values, and T _L (S) can be eliminated. Where is the nominal value of the inertia of the controlled object, is the nominal value of the viscosity coefficient of the controlled object, and _T is the nominal value of the torque generation coefficient of the controlled object.

Therefore, the following equation (1) is obtained from FIG. _T T ^* (S) = Sω (S) + ω (S) (1) Further, when this is differentiated, Expression (2) is obtained.

In addition, when the backward difference is obtained, Expression (3) is obtained.

Where T _S is the CPU calculation sample time T ^* _(n) is the torque command of the current sample ω _(n) is the rotation speed at the current sample (unknown) ω _(n-1) is the rotation speed at the previous sample ( ₍ Known) ω _{(n + 1)} is the rotation speed at the next sample (estimated value) Here, since all of Expression (3) is constituted by known information, the rotation speed can be calculated by this expression.

Moreover, this rotational speed is a robust estimated value which is not affected by equivalent disturbances such as parameter fluctuations and load fluctuations, and is a state quantity of the stabilizing feedback amount shown in claim 1 of the appended claims. This is used as a feedback amount or as a part of the feedback amount.

Further, a specific example provided for practical use based on the above equation will be described below.

By the way, if the DSP (Digital Signal Processor) is used for the CPU, the CPU operation sample time T _S is usually 100 (μS).
Equation (3) is calculated every 100 (μS) and the second
And the rotational speed directly detected by the PG is the first state variable (actual speed), which is normally calculated about every 1 (mS) at high speed.

Thus, the actual speed by the PG is calculated and output every time the estimated speed by the formula (3) is calculated about 10 times, and ω in the formula (3) is calculated.
_{(n + 1)} or ω _(n-1), and any differences will be corrected through an appropriate weight function (usually a constant or first-order lag).

Further, at low speeds, the pulse signal interval from the PG becomes longer, so that the calculation of the actual speed becomes longer than the above-mentioned 1 (mS), and the correction interval of the estimated speed becomes longer by that amount, but this is not a problem.

Furthermore, if a slow general-purpose CPU is used,
Although the calculation sample time may be longer than 1 (mS), as long as the actual speed by the PG can be calculated at a high speed by another means, the actual speed may be always used for feedback.

By performing the equivalent disturbance compensation in this manner, the control target has a robust form that is not affected by equivalent disturbances such as parameter fluctuations and load fluctuations as shown in FIG. ) Is derived, which is equivalent to the state quantity of the stabilizing feedback device, and is used for feedback.

Further, the present invention will be described in detail with reference to the accompanying drawings.

〔Example〕

FIG. 1 is similar to FIG. 3 for easy understanding of the technical idea of the present invention, where γ is a final command output.
In the drawings, those having the same reference numerals as those in FIGS. 3 and 4 indicate portions having the same functions.

FIG. 2 shows an example of a main hardware configuration for the main blocks shown in FIG.

That is, in FIG. 1 and FIG. 2, assuming that the setting input R is changed by, for example, operating the command setting device to obtain the final command output γ, the final command output γ and a part of the state quantity are obtained. Of the output Y is subjected to stabilization compensation by the PI controller 1 or the like, and the output is T ^* (S).
Becomes Its T ^{* (S)} by passage through a power actuator, over the torque generating coefficient K _T gain, applied to the control object 2. On the other hand, in the equivalent disturbance compensation, T
^* (S) and Y are input and the operation result is similarly added to the output of the stabilization compensation.

In this way, such a practical example has the same function as the above-mentioned solution means, and obtains a special operation and effect.

In the case where the first state quantity and the second state quantity are used as feedback amounts, the feedback amount is monitored, for example, a method of capturing as a time amount or providing a limiter can be performed by using a conventional technique. It is clear that abnormality can be detected, and from this it can be detected if the feedback amount is appropriately determined as a sensor abnormality.

[Effects of the Invention] As described above, according to the present invention, although the detection of the state quantity is based on the direct detection value by the sensor,
Estimated state quantity is used as compensation for deterioration of control performance due to sensor circuit installation, and equivalent disturbance compensation is performed as described above for robustness not compensated using an observer, and parameter fluctuations and load fluctuations are performed. Even if a disturbance occurs, the estimated state quantity is hardly affected, the robustness of the estimated state quantity is ensured, and the optimal stabilizing feedback control that compensates for the decrease in the reliability of the state quantity directly detected by the sensor is performed. Device can be provided.

[Brief description of the drawings]

FIGS. 1 and 2 are a block diagram of a main configuration and a main hardware configuration diagram for facilitating the understanding of the technical idea of the present invention, and FIG. 3 is a block diagram showing a conventional example of a stabilizing feedback control device. 4 and 5 are a block diagram showing an equivalent disturbance compensator according to the present invention and a block diagram showing a control target portion after the compensation. 1 ... PI control device, 2,2 '... Control target, 3 ... Equivalent disturbance compensator.

Claims (5)

(57) [Claims] A speed (Y = ω (S)), which is one of the state quantities of a controlled object, is fed back to a speed control command (r),
A control method in which the deviation amount is input to the controlled object through stabilizing compensation means including P (proportional) and I (integrated), and a torque command (T ^* (S ^* )) And the speed (Y = ω (S)), which is one of the state quantities of the control object, between the input quantity (T ^* (S)) and the state quantity (Y = ω (S)) (control object). In the stabilized feedback control method in which an equivalent disturbance is estimated using an amount obtained by multiplying the inverse function of the mathematical model of the mathematical model and the estimated amount of the equivalent disturbance is fed back to the input side of the controlled object, (T ^* (S)) and a speed model (Y = ω (S)), which is one of the state variables, are represented in a nominal expression by using the nominally represented mathematical model. The speed (Y which is one of the state quantities of the control object
= Ω (S)), and the estimated state quantity is used as the feedback quantity or as a part of the feedback quantity. 2. A speed (Y = Y) which is one of the state quantities of the controlled object.
ω (S)) is calculated and determined by a microprocessor or the like based on the amount detected by the sensor, and this is used as a first state quantity. In addition, this is used as a second state quantity, and when the first state quantity is calculated and updated within a fixed time, the value is used as a feedback quantity, and during that time, the second state quantity is set to the first state quantity. If the amount is corrected as positive and the first state amount is not updated within a certain period of time, the second state amount is used as a feedback amount, and the first state amount is updated after a certain period of time. 2. The stabilized feedback control method according to claim 1, wherein the second state quantity is corrected when the value is positive. 3. When the first state quantity is calculated and updated within a predetermined time, the absolute value of the first state quantity or the total value of the change per unit time is calculated as the second state quantity. When it becomes larger than a certain value, the second state quantity is preferentially used as the feedback amount,
When the period continues for a certain time or more, it is determined that the sensor is abnormal, and when the first state quantity is not calculated and updated within a fixed time, the second state quantity is used as a feedback quantity, and 3. The stabilizing feedback control method according to claim 2, wherein when the first state quantity is not updated even after a certain amount of time has elapsed since the state quantity of (2) has become equal to or more than a predetermined amount, it is determined that the sensor is abnormal. 4. The method according to claim 2, wherein the second state quantity is calculated by a microprocessor or the like for each calculation sample time, and the time required for calculating and updating the first state quantity is less than the calculation sample time. 3. The method according to claim 2, wherein the second state quantity is used as feedback when the value is long, and the second state quantity is corrected to be positive when the first state quantity is updated. Stabilization feedback control method. 5. A sensor according to claim 4, wherein said sensor is determined to be abnormal when said first state variable is not updated after a certain period of time when said second state variable is abnormal. A stabilization feedback control method as described.