JPS62271102A - Servo constant setting system for servocontrol system - Google Patents

Abstract

PURPOSE:To automatically set an optimum servo constant in a servo controlling section by providing a function that generates signals for measuring, makes servocontrol operation, observes desired variable in servocontrol operation and decides characteristic of a servo system in a digital servocontrol section. CONSTITUTION:Basically, a servocontrol section is constituted of an operating means using a microprocessor etc., and a digital servocontrol section 4 that makes servocontrol by operation is used. In such a digital servocontrol section 4, the servo constant is held as a value in a memory, and its change is made only by change of the value. Deciding of servo characteristic can be executed by operation utilizing variables of operation point by servocontrol operation, and accordingly, a servo characteristic deciding function 4b, open loop characteristic can be measured by BS/ES, closed loop characteristic can be measured by PS/IS and error characteristic can be measured by ES/IS.

Description

[Detailed Description of the Invention] 3. Detailed Description of the Invention [Table of Contents] Overview Industrial Field of Application Conventional Technology Problems to be Solved by the Invention Means for Solving the Problems (Fig. 1) Working Examples (a) Explanation of the configuration of one embodiment (FIGS. 2 and 3) (b) Explanation of the operation of one embodiment (FIGS. 4 and 5) (C) Explanation of other embodiments Effects of the invention [Summary] In a servo control system in which a servo control unit that servo-controls a controlled object is configured with a digital servo control unit, a measurement signal is generated to the digital servo control unit, a servo water control calculation is performed, and the desired result in the servo control calculation is By providing a function to observe variables and determine the characteristics of the servo system, the optimum servo constants can be automatically set within the servo control section.

[Industrial application field]

In a servo control system in which a controlled object is servo-controlled by a servo control unit, the present invention uses a digital servo control unit in the servo control unit that performs servo control using servo control calculations using servo constants to automatically determine the optimal servo constants. Regarding the servo constant setting method.

Servo control technology is widely used in head positioning control of magnetic disk drives and the like.

In such a servo control system, it is necessary to set appropriate servo constants according to the servo characteristics in order to stabilize the system.

[Conventional technology]

For example, the π system in the servo system used to control the head positioning of a magnetic disk is, as shown in FIG. A servo control unit 2 is provided for a controlled object l having a demodulator 1d.

The servo control unit 2 includes an error generation unit 2a that takes the difference between the command amount Is and the state quantity BS from the controlled object 1 (servo signal from the demodulator 1d), and an error generation unit 2a that calculates the error 31Es of the error generation unit 2a.
, and a gain setting section 2C that multiplies the output PS of the phase compensation section 2b by a set gain and outputs a control amount U.

In such a servo control system, the characteristics of the servo system are determined by the servo constants of the phase compensation section 2b and the gain setting section 2c.

Conventionally, in setting this servo constant, the servo control section 2 was configured with analog elements such as an operational amplifier, a resistor, and a capacitor, and the servo constant was adjusted by the values of various variable resistors.

At this time, using a dedicated external measuring instrument such as the servo analyzer 3, the noise signal NS for measurement is given from the servo analyzer 3 to the servo control section 2, and the signals of each section due to the operation of the servo control section 2 are observed. The servo characteristics are determined using , Fourier transform, etc., and the variable resistor is manually adjusted to match the design value to obtain the desired servo characteristics.

[Problem that the invention seeks to solve]

Optimum setting of such servo constants uses an external measuring instrument, so it is usually done in the product testing process or adjustment process. However, since the adjustment is done manually using an external measuring instrument, productivity can be improved. There was a problem that it was not possible to automate the process.

In addition, even if the product has already been shipped, if it is used for a long period of time, the servo characteristics itself may change due to various mechanical factors such as temperature conditions and deterioration due to use, and in this case, the servo constant It is difficult to readjust the servo characteristics, for example, it may not be possible to determine until a failure occurs in the equipment, and when the failure occurs, maintenance personnel will have to readjust it for the first time using the external measuring device mentioned above. There is also the problem that unsatisfactory operation of the servo system due to a change in the current level continues until a failure occurs, and furthermore, in this case, readjustment must be performed manually.

Furthermore, since the characteristics of each device are not completely uniform and differ slightly, there is also the problem that adjustment and readjustment of these servo constants must be done manually for each device.
Optimal automatic setting technology for servo constants has been desired.

In view of the above-mentioned points, it is an object of the present invention to provide a servo constant setting method for a servo control system that can automatically set the servo constants of the servo control system.

[Means for solving problems]

FIG. 1 is a diagram explaining the principle of the present invention.

In the figure, the same parts as those shown in FIG. 6 are indicated by the same symbols, and 4 is a digital servo control section, which is a servo control calculation function section 4 that performs the above-mentioned servo control by calculation.
a, and a servo characteristic determination function part 4b, and the servo control calculation function part 4a has an error occurrence calculation 40b for obtaining an error BS between the command quantity Is and the state quantity BS, and a servo constant (phase A phase compensation calculation 40C for compensating the phase with a constant) and a gain multiplication calculation 40d for multiplying the phase compensated signal PS by a servo constant (gain) are performed by calculation processing, and the measurement signal (noise etc.) NS is A noise injection addition calculation 40a is provided to introduce the command amount Is into the servo calculation loop.

On the other hand, the servo characteristic determination function section 4b uses state variables (for example, command amount Is, state amount B) in the servo control calculation function 4a.
S, error amount ES, phase compensation output amount PS),
This is used to determine the servo characteristics through calculation and determine the optimal servo constant.

Note that 5a is a digital/analog converter, and 5b is an analog/digital converter, each of which is connected to the servo control section 4.
Digital system? This is for converting nl into an analog control amount and giving it to the controlled object 1, and converting the analog Lit of the controlled object 1 to a digital state at and giving it to the servo control section 4.

[Effect]

In the present invention, a digital servo control section 4 is used in which the servo control section is basically constituted by a calculation means such as a microprocessor, and the servo control is performed by calculation processing.

In such a digital servo control section 4, the servo constants are stored in the memory as values, and the changes can be made by simply changing the values.

Also 1. Since the servo characteristics can be determined by calculation using the variables of each calculation point based on the servo control calculation, the digital servo control section 4 can incorporate the servo characteristics determination function 4b for determining servo constants.

For example, in the servo characteristic determination function 4b, the open loop characteristic can be measured by BS/ES, the closed loop characteristic can be measured by PS/Is, and the error characteristic can be measured by BS/Is.

Therefore, it is possible to incorporate a servo characteristic judgment function into the device, and to have the servo constants adjusted within the servo control unit 4. Conventional tests and adjustments can be performed automatically, and changes and differences in servo characteristics can be detected. The corresponding servo constants are automatically adjusted.

〔Example〕

(a) Description of the configuration of one embodiment FIG. 2 is a configuration diagram of an embodiment of the present invention.

In the figure, the same components as those shown in FIG. 6 and FIG. 4B is a processor for measuring servo characteristics, which performs Fourier transform on variables obtained by servo control calculations, converts them into a frequency series, and calculates a transfer function as servo characteristics; 4C; is a memory for data transfer, and is a memory for receiving and transmitting variables resulting from servo control calculations and calculated transfer functions between the control processor 4A and the servo characteristic measurement processor (hereinafter referred to as measurement processor) 4B. It is provided as a buffer.

40 is a control processor microprocessor (hereinafter M
41 is a read-only memory (hereinafter referred to as ROM) that is necessary for the calculation operation of the MPU 40. 42 is a random access memory (hereinafter referred to as R
43d is an MPU 40 that stores variables and servo constants for servo calculations, which will be described later.
43a is a data bus used by the MPU 40 to send and receive data; 43a is an address bus used by the MPU 4
0 is used for sending and receiving addresses, 44 is a microprocessor for measurement (hereinafter referred to as MPU)
45 is a read-only memory (hereinafter referred to as ROM) which performs Fourier transform on the variables from the MPU 40 of the data transfer memory 4C, performs frequency conversion, and then calculates the gain and phase at each frequency of the transfer function. ,M
46d is a data bus that stores the programs necessary for the arithmetic processing of the PLI 40, and 46a is an address bus that the MPU 44 uses to send and receive addresses. , 47 is a common RAM (random access memory) in which variables are written by the MPU 40, the written variables are read by the MPtJ 44, transfer functions are written by the MPtJ 44, and the written transfer functions are read by the MPU.
40, 48a and 548b are buffers, respectively connected to the data buses 43d and 46d, and both connected to the data bus 47d of the i1 RAM 47, for reading to the common RAM 47 of the MPU 40.44. /Used for writing, 49a and 49b are latches, respectively, and are connected to address buses 43a and 46a, respectively, and also to address bus 47a of the common RAM 47, and are used to specify addresses to the common RAM 47 of the MPU 40.44. , O8 is the MPU interlock key control line,
C3A and C3B are chip select lines that exchange status with MPU40.44, respectively.
Each MPU 40 operates a buffer 48a, a latch 49a, an M
This is a control line for the PU 44 to select the buffer 48b and latch 49b.

FIG. 3 is an explanatory diagram of servo arithmetic processing by the control processor, FIG. 3(A) is a block diagram of the servo control system, and FIG.
Figure CB) is a block diagram showing the control system of Figure 3(A) in digital calculation form.

In FIG. 3(A), the deviation BS between the command quantity Is and the state quantity (current position) BS is filtered by a phase lead/lag element filter 4.
0C, multiplied by a gain K, and outputs the control amount US, and includes an adder 40a for injecting noise NS.

If this is expressed by digital calculation, it will be as shown in FIG. 3(B), and if the target position is CS, then the following will be obtained.

On the other hand, the phase lead/lag filter converts the transfer function to (1+T2
S)/ (1+T, S) (where T, , T2 are the time constants of the phase lead/lag filter), as shown in Figure 3 (B), the proportional amount RAS, the differential amount EFS1 the integral amount PBS, P S = E A S + E F S - P B
S -------(2), and further, where Z indicates a Z transformation (ie, Z-1 is a delay element).

Here, the aforementioned 80, al, b+ are the coefficients (servo constants) of the phase lead/lag filter, and are expressed as.

These a@, al, I)1 and K are the servo constants in this example.

Therefore, the RAM 42 of the measurement processor 4A contains the error IEs, the Z-converted value EDS of the error amount, the proportional amount RAS, the differential amount EFS, the integral amount PBS,
The Z-converted value PIS of the output PS, the output PS, the control amount US, the state amount BS, the command amount S, the target position C8, and the noise NS are updated and stored in sequence as the servo calculation results of the MPU 40 at each sampling period. a6+al as a constant
,b, ,K are stored.

(b) Description of operation of one embodiment FIG. 4 is a servo constant setting process flow diagram.
A) is a processing flow diagram of the MPU 40, and FIG. 4 (B) is a process flow diagram of the MPU 40.
The processing flowchart of U44, FIG. 5, is an explanatory diagram of servo constant determination.

First, the process shown in FIG. 4 is started at power-on or adjustment.

■ MPU40 reads servo constants a0, a, a, from ROM41.
, bl, , are set in the RAM 42.

(2) Next, set the noise NS in the RAM 42 from the random number (noise) generation program in the ROM 41.

Note that the noise includes frequency components within the analysis range.

(2) Then, the MPU 40 performs the servo control calculation processing shown in FIG.
The variables are calculated using the servo constants a0, al, bl, and K, and the variables are written to the RAM 42 and read out for calculation.Finally, the control NUS is determined and the data bus 4-3
d to a digital-to-analog converter (DAC) 5a, and current drives the actuator 1a via an amplifier IC.

In this case, the target position CS is set to zero or a value close to zero.

■ When the MPU 40 finishes this servo control calculation, the state of the RAM 42 at this time @BS (i), the error amount ES
C+) to common R via buffer y4Ba and lunch 49a
Write to AM47.

- The MPU 40 determines whether this has been repeated a predetermined number of times, and if it has not been repeated the predetermined number of times, returns to step
Write S (i) and ES (i).

(2) On the other hand, when the writing is repeated a predetermined number of times, the MPU 40 notifies the measurement MPU 44 via the control line CS that the writing of a predetermined number of variables to the common RAM 47 is completed. At this time, common R
BS(i) and 23(i) in time series are written in AM47.

As a result, the characteristics measurement process of the MPU 44 shown in FIG. 4(B), which will be described later, starts. MPU44 via 11 line CS
If you do not check whether the measurement process is completed from step ■
Returns to , repeats the servo calculation, and servo controls 'M target 1, but at this time, the BS (i) to the common RAM 47 is
, BS (i) is not written.

Conversely, upon receiving completion of the measurement process, MP is transferred from the common RAM 47.
The gain and phase at each frequency of the servo system transfer function calculated by U44 are successively output.

■ Mptr4o is as shown in Fig. 5 < Design value (frequency 1k) f from ROM41, target phase value θ at (servo band)
0 and the gain, and the frequency r of the common RAM 47
Compare with phase and gain of 0.

If these match the target values, it is determined that the servo characteristics are good, and the servo constant setting process ends.

(2) Conversely, if the MPU 40 does not match the target value, it changes the servo constant.

Here, the gain is changed so that the gain of the transfer function becomes OdB (target value) at frequency r0. For example, if the difference is 6 dB, the gain K is set to 0.5 times, and the RAM 42
Set to . Also, regarding the phase constant, the servo band f
Measured phase at 0 and target value θ. Compare the values, reset the time constant T+ and TtO value of the phase lead/lag filter, and a
Change o, al, and bI, set them in RAM 42, and return to step and b.

In this way, the servo control calculation using the changed servo constant, the output (writing) of BS (i) and ES (i), and the measurement of the transfer function in the MPU 44 are repeated until the servo characteristic value reaches the target value. Repeat this.

Next, the operation of the measurement processor 4B will be explained.

■ In the above step ■, when the MPU 44 receives the write completion notification, it stores the time series BS (i), ES N) (i=1.2) written by the MPU 3O from the common RAM 47.
, -n) one after another.

■ Next, the MPU 44 outputs each time series data BS (i),
Each ES (i) is subjected to Fourier transform (FFT) and converted into a frequency series.

(2) Furthermore, the MPU 44 calculates the gain and phase for each frequency of the open loop transfer function BS/ES based on the relationship between the two.

■ Then, write these to the common RAM 47 and control f
! The MPU 44 notifies the control MPU 40 of measurement completion via the cs, and executes this every time there is a write completion notification from the MPU 40, and uses it to determine the servo characteristics of the control MPU 40.

In this way, the control MPU 40 performs servo control calculations based on the servo constants set for each predetermined cycle, servo-controls the control object 1 with noise, and uses the common RAM 47
When a predetermined number of variables are collected, the measurement MPU
44 reads this from the common RAM 47, performs Fourier transform to calculate a transfer function, and writes this to the common RAM 47.

Then, the control MPU 40 reads out the calculated transfer function from the common RAM 47, compares it with the target value, determines whether the servo characteristics are good or bad, and if the servo characteristics are bad, changes the servo constant, and similarly controls the servo for measurement. Repeat control calculation.

On the other hand, if the servo characteristics are good, this servo constant setting process is finished, and the third
The servo calculation shown in the figure is performed to servo control the controlled object 1.

In this way, the servo constants can be optimally set in the digital servo control section, and the process shown in FIG. 4 is executed at the time of device startup or adjustment, and thereafter normal servo control B is performed.

(C) Description of other embodiments In the above embodiment, the MPtJ40 repeats the servo control calculations from step ■ onwards without making any judgment even after writing in step ■, but after writing in step ■, it does not return to step ■. First, it may be possible to wait for the MPU 44 to complete the calculation.

In addition, a measurement processor is provided in addition to the control processor,
Although the measurement processor performs the servo characteristic measurement process, the measurement processor may not be provided and the control processor may further execute the measurement process.

Furthermore, although we have explained the example of finding the open-loop transfer function and setting the phase constant and gain, it is also possible to find the open-loop transfer function and error characteristics and set other constants, which also prevents the measurement signal from being affected by noise. However, other functions such as a sine wave may be used.

Although the present invention has been described above using examples, the present invention can be modified in various ways according to the gist of the present invention, and these are not excluded from the present invention.

〔Effect of the invention〕

According to the present invention, the servo control unit is configured with a digital servo control unit and a servo characteristic determination function is added to the collar described above, so that the optimum servo constant can be automatically set in the digital servo control unit. With this effect, it is possible to optimally set the servo constants without using an external measuring device.

In other words, since the servo constants are automatically adjusted, there is no need for manual adjustment at the time of product shipment, and even after shipment, adjustments are made automatically according to changes in characteristics, ensuring servo control with the optimal servo constants. This greatly contributes to improving the reliability of the servo control system and reducing the effort required for 31f] adjustment.

[Brief explanation of drawings]

FIG. 1 is a diagram explaining the principle of the present invention, FIG. 2 is a configuration diagram of an embodiment of the present invention. FIG. 3 is an explanatory diagram of the servo control calculation process of the configuration shown in FIG. Figure 4 is a flow diagram of the servo constant setting process configured in Figure 2; Figure 5 is an explanatory diagram of servo constant determination in the 4th time, FIG. 6 is an explanatory diagram of the prior art. In the figure, 1-controlled object, 4...Digital servo control section, 4a--servo control function, 4b-Servo characteristic determination function.

Claims (1)

[Claims] A controlled object (1) and a state quantity from the controlled object (1) are fed back, and a servo control calculation is performed using a servo constant set based on the command quantity and the state quantity. In the servo control system, the digital servo control unit (4) generates a predetermined measurement signal and performs the servo control calculation. and a servo characteristic determination function for determining the characteristics of the servo control system by observing desired variables in the servo control calculation, and setting optimal servo constants that will improve the characteristics of the servo control system. A servo constant setting method for a servo control system characterized by the following.