JP2007072898A - Servo controller, control parameter determination method and control parameter determination program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To easily set an accurate control parameter to a servo control system. <P>SOLUTION: Frequency characteristics of a control system are expressed by a complex number and the zero point of the complex number is specified on the basis of the zero crossing frequency fx of open loop frequency characteristics in a control parameter generation part 28. A low cutoff frequency of a controller determined from a disk rotational frequency or the like is determined and a target zero crossing frequency fx is determined. Then, in a control parameter calculation part 41 of the control parameter generation part 28, the zero points cs and ds of the complex number of the controller are determined on the basis of the zero crossing frequency fx of the open loop frequency characteristics, and the provisional value of a controller gain Kp is determined. Then, servo is applied, the open loop frequency characteristics are measured, and the controller gain Kp is determined so that the zero crossing frequency of the open loop frequency characteristics matches with the target zero crossing frequency fx. Then, a phase margin is confirmed from the open loop frequency characteristics, and when a phase margin is 42° to 32°, the control parameter at this time is determined as a final control parameter. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention controls the focusing position of a laser beam by an optical pickup in an optical recording / reproducing apparatus using various optical disks and magneto-optical disks, and controls the position of a magnetic head in a magnetic recording / reproducing apparatus using a floppy disk, hard disk, etc. The present invention relates to a servo control device and a control method therefor.

  In an optical disk recording / reproducing apparatus and a magnetic disk recording / reproducing apparatus, servo control is performed for laser focusing position control of an optical pickup or magnetic head position control (see, for example, Patent Documents 1 to 6). Servo control is realized by subjecting an error signal detected by the optical pickup or magnetic head to a control operation by a servo controller, and using this as a drive signal to drive the position of the optical pickup or magnetic head by an electromagnetic actuator.

  The servo controller is generally composed of a phase advance / delay control and a series connection of a low-frequency booster. In the phase advance / delay control, the high-frequency phase is controlled and stabilized, and the low-frequency booster increases the low-frequency gain to suppress position disturbances such as circumferential blurring and eccentricity.

  Specifically, in the phase advance / delay control, the phase advance frequency is set to 1/3 of the target value fx of the zero cross frequency at which the gain of the open loop frequency characteristic becomes 0 dB, and the phase delay frequency is set to 3 times higher. Servo control is stabilized by advancing the region phase, and the low region booster increases the low region boost frequency to the extent that it does not affect the high region phase and suppresses position disturbance. Next, a control is performed in which the gain is set so that the zero cross frequency matches the target fx by looking at the open loop frequency characteristics, and the final control parameter is determined by fine adjustment based on these control parameters. It has been broken. In the servo control, it is desirable to suppress the low-frequency positional shift as much as possible by enhancing the low-frequency component as much as possible with respect to the error signal. It is also desirable to be able to respond to as high a frequency error signal as possible by phase lead / lag compensation.

  Therefore, a method has been proposed in which the controller zero is a complex number and the low-frequency gain is increased. That is, the two controller zeros determined by the low frequency boost frequency and the phase advance frequency of the phase advance / delay control are automatically set to real numbers in the conventional controller, but can be set as conjugate complex zeros. This is a method of further increasing the low frequency gain and enhancing the position disturbance suppression effect. There has also been proposed a method for calculating the controller zero and the controller pole when the controller is realized by a digital control system by the pole placement method. By approximating the dead time element included in the control system to an integral multiple of the sampling frequency and deriving the pole placement equation, the digital controller can perform the pole placement more accurately. By these methods, it is possible to realize good control with high low-frequency gain and resistance to disturbance, and it is possible to calculate this control parameter by pole placement calculation without trial and error.

JP-A-4-345929 Japanese Patent Laid-Open No. 9-293251 Japanese Patent Laid-Open No. 3-168936 JP-A-2-249139 JP-A-2-263338 JP 2000-123377 A

  However, in the calculation of the control parameters described above, it is necessary to use a control parameter calculation formula based on the above-described pole arrangement calculation, which is more complicated than the conventional control for determining each control parameter. In addition, it is difficult to intuitively understand the meaning of the calculated control parameter, and even if there is an error, it is difficult to find it. Furthermore, it is necessary to use different control parameter calculation formulas depending on the size of the dead time element, which is complicated.

  The present invention has been devised in view of the above-mentioned problems, and is intuitive as in the case of conventional phase advance / delay control while having a complex zero point in the controller and a high disturbance suppression characteristic compared to the conventional controller. An object of the present invention is to provide a servo control device and a control parameter determination method for generating a control parameter for servo control that is clearly understandable and clear in meaning, and a program that causes the servo control device to execute control parameter determination processing.

  The present invention provides a recording / reproducing element for writing data to a disk-shaped recording medium or reading data recorded on the disk-shaped recording medium, and the recording / reproducing element for the recording surface of the disk-shaped recording medium. Detects an error signal that is proportional to the difference between the position where the recording / reproducing element should perform recording / reproduction and the actual position, and detects an error signal corresponding to the difference signal. An error signal detection means for performing control calculation on the error signal detected by the error signal detection means using a control parameter, and calculating a control output for making the error signal as small as possible, and a drive means And a control parameter for calculating control parameters for realizing the servo performance required for the servo control system constituted by the error signal detection means and the control calculation means. Comprising a meter calculating means, the control parameter calculation means, by and calculating a complex zeros from the zero cross frequency of the open loop frequency characteristic is an index representing the servo performance, to solve the problems described above.

  Specifically, the provisional value of the control parameter is determined from the zero cross frequency fx in the open loop frequency characteristic, the gain is set so that the zero cross frequency matches the target fx according to the open loop frequency characteristic, and the phase margin is further increased. A fine control parameter is finely adjusted accordingly to finally determine a good control parameter.

  In particular, in the control parameter calculation means, when the zero cross frequency is set to fx, the filter in the control calculation means has a conjugate zero represented by the expressions (1) and (2), and the expressions (1), ( 2) The zero point frequency ratio Ar in the equation is 0.28 to 0.3, and the zero point angle θ is 143 to 145 degrees.

  Further, in the control parameter calculation means, when the zero cross frequency is fx, the filter in the control calculation means has a pole set based on the equation (3), and the pole frequency ratio Br in the equation (3) is 3 ˜7.

  The control parameter determination method according to the present invention calculates a control parameter for an error signal that is proportional to a difference between a position to be localized and an actual position and that corresponds to the difference signal in the control of the drive system, In a control parameter determination method for drive control that performs control calculation using a control parameter to reduce an error signal as much as possible, an open loop zero cross frequency in a servo control system, a pole that is a ratio of a controller pole and a zero cross frequency An acquisition step of acquiring a frequency ratio and a controller gain, a first calculation step of calculating a controller zero using at least the acquired zero cross frequency, and a control based on the acquired pole frequency ratio and zero cross frequency A second calculation step for calculating a device pole, an open loop frequency calculated from the calculated controller zero and the controller pole Obtaining a characteristic by having an acquisition step of reacquiring controller gain according to the measurement results of the open loop frequency characteristic, to determine the optimal control parameters.

  In the first calculation step, when the zero cross frequency is set to fx, there are conjugate zeros represented by the following formulas (1) and (2), and zeros in the formulas (1) and (2): The frequency ratio Ar is 0.28 to 0.3, and the zero point angle θ is 143 to 145 degrees.

  Further, in the second calculation step, when the zero cross frequency is set to fx, the pole is set based on the following formula (3), and the pole frequency ratio Br in the formula (3) is set to 3-7.

  In the present invention, the provisional value of the control parameter is determined from the zero cross frequency fx in the open loop frequency characteristic, and the gain is set so that the zero cross frequency matches the target fx according to the open loop frequency characteristic. As a result, the control parameters can be obtained intuitively and easily, and even if there is an error, it can be easily detected. Next, the finest control parameters are finally determined by fine adjustment so as to realize a sufficient phase margin. This makes it possible to determine parameters with the same algorithm even for systems with different dead times.

  As an embodiment of the present invention, an application example to the optical disc recording / reproducing apparatus 3 shown in FIG. 1 will be described.

  The optical disc recording / reproducing apparatus 3 outputs a control signal by executing predetermined calculation processing with a recording / reproducing element 31 for writing data to the optical disc 4 or reading data recorded on the optical disc 4. The control calculation part 15 and the control parameter generation part 28 which produces | generates the control parameter required for the control function in the said control calculation part 15 are provided.

  The recording / reproducing element 31 generates an error signal by focusing the light beam on each recording layer of the optical disk 4 to be rotated and detecting the return light reflected by the recording surface of the optical disk 4 to control this. It transmits to the calculating part 15. The recording / reproducing apparatus 31 is composed of a semiconductor laser or the like using semiconductor recombination emission, and receives a return light from the optical source 4 and a laser light source 32 that emits laser light of a predetermined wavelength and photoelectrically converts it. A light receiving unit 35 that guides the laser light emitted from the laser light source 32 to the optical disc 4 side, and transmits the laser light reflected by the optical disc 4 as it is to guide the laser light to the light receiving unit 35. And a lens 34 that condenses the laser light from the light separation unit 33 to focus on each recording layer of the optical disc 4 and makes the return light from the optical disc 4 parallel light. Is provided with an electromagnetic actuator 36 that moves the recording / reproducing element 31 close to and away from the recording surface of the optical disc 4 according to the control signal.

  The control calculation unit 15 in the automatic control device 2 detects an error signal generated in the light receiving unit 35, thereby generating an error signal detection circuit 51 that generates a focus error signal, and a focus error generated in the error signal detection circuit 51. By subjecting the AD converter 52 for A / D conversion of the signal and the focus error signal digitized by the AD converter 52 to predetermined control using the control parameter generated by the control parameter generator 28 A control arithmetic circuit 53 for generating a control signal, a DA converter 54 for D / A converting the control signal generated by the control arithmetic circuit 53, and an electromagnetic wave based on the control signal digitized by the DA converter 54 And an actuator driving circuit 55 for driving the actuator 36.

  Further, the control parameter generation unit 28 in the automatic control device 2 includes a control parameter calculation unit 41 that actually calculates the control parameters, and an input unit 44 that sets numerical values for the control parameter calculation unit 41.

  The control parameter calculation unit 41 calculates the control parameter used for each control function of the control calculation circuit 53 based on the zero cross frequency fx designated by the input unit 44 and outputs the calculation result to the control calculation circuit 53. To do.

  That is, the control parameter generation unit 28 generates control parameters necessary for the control calculation. Specifically, a desired zero-cross frequency fx in the control system open-loop transfer function is input to the input unit 44, and the control parameter calculation unit 41 calculates the input zero-cross frequency fx from the input zero-cross frequency fx according to a pole arrangement formula described later. A control parameter equivalent to the control parameter is simply calculated using an approximate expression. By transferring this control parameter to the control arithmetic circuit 53, the servo control of the recording / reproducing element 31 becomes possible.

  Hereinafter, the pole placement formula calculated by the control parameter generation unit 28 will be described. FIG. 2 shows a control block diagram when the control system constituting the automatic control device 2 having the above-described configuration is realized by digital control.

  In FIG. 2, “r” (= ref) is a target position that represents the position where the recording / reproducing element 31 should be. “D” (= dis) is a disturbance such as a disc runout applied to the target position. “E” (= err) represents a position error. This position error err corresponds to a difference ref + dis−y between the target position ref + dis of the recording / reproducing element changed according to the disturbance dis and the actual position y of the controlled object. In the control system shown in FIG. 2, the disturbance dis applied to the recording / reproducing element 31 and the actual position y cannot be directly detected, and the error signal detection circuit 51 detects the position error err = ref + dis−y corresponding to the focus error signal. Servo control is performed.

  The position error err is sampled and a control calculation is performed by the control calculation unit K (z). The control parameters of the control system are: low-frequency emphasis low-frequency cutoff discretization angular frequency a, low-frequency emphasis high-frequency cutoff discretization angular frequency c, phase advance discretization angular frequency d, phase lag discretization angular frequency b , And controller gain Kp. After this control calculation, the calculation result is output after one sampling time. Note that the calculation time delay in the control calculation unit K (z) corresponds to a dead time element in FIG. The output calculation result is output as a control signal u through the 0th-order holder. The zero-order holder corresponds to the DA converter 54 in FIG. The actuator is driven by the control signal u to determine the position y of the recording / reproducing element. A transfer function from the control signal u to the position y of the recording / reproducing element is represented by a control object P (s) in FIG. Further, the control object P (s) is represented by a quadratic integration element for simplicity.

  Here, an example of a conventional method for calculating the pole placement equation for the control system of FIG. 2 will be described. The control object P (s) and the 0th-order holder in FIG. 2 are z-transformed to P (z), and the whole is represented in FIG. 3 as a discrete system. The closed loop transfer function of this control system is given by the following equation.

  In the above equation, there are five poles because the denominator is a quintic equation. There are five control parameters as described above, but the low-frequency emphasis low-frequency cut-off discretization angular frequency a is governed by a factor other than the control performance. There are essentially four parameters. Since five poles are arranged with four control parameters, one pole cannot be arranged. When the four poles that can be arranged are p1, p2, p3, and p4, and the position of the pole that cannot be arranged is q, the closed loop transfer function is expressed by the following equation.

  When the coefficient comparison formula is created so that the denominators of the two formulas are equal, the following formula is obtained.

  Thus, when the equations are solved so that b, c, d, Kp, and q are calculated, the following pole placement equations (4) to (10) are obtained. Thus, the control parameter calculation unit 41 is realized by an arithmetic unit that outputs b, c, d, Kp, and q for the inputs p1, p2, p3, and p4 in accordance with the equations (4) to (10). . The stability determination unit 42 is realized by a device that determines whether the expression (11) is true or false. Here, p1, p2, p3, and p4 are positions of four poles to be arranged, and q represents a position of “poles that cannot be arranged” that is automatically determined. Further, the low-frequency cut-off discretization angular frequency a is designated in advance from factors such as the disk rotation frequency.

In the above example, as shown in FIG. 2, the pole placement type is used when there is a dead time element with a delay of 1 sampling time in the control system. Even when there is a dead time, the pole placement equation can be obtained based on the same concept. For example, the calculation of the pole placement formula when there is a delay time element of 3 sampling times will be described. A block diagram of the control system at this time is shown in FIG. Compared with the control system shown in FIG. 3, the dead time element can be dealt with by changing from 1 / z to 1 / z ³ . The closed loop transfer function of this control system is

  In the above equation, since the denominator is a seventh-order equation, there are seven poles. The control parameters are substantially four as in the case where the calculation time delay is one sampling time. Since seven poles are arranged with four control parameters, three poles cannot be arranged. When the four poles that can be placed are p1, p2, p3, and p4, and the positions of the poles that cannot be placed are q1, q2, and q3, the closed loop transfer function is expressed by the following equation.

  When the coefficient comparison formula is created so that the denominator formulas of the two formulas are equal, the following formula is obtained.

  From this, when the equations are solved so that b, c, d, b, c, d, Kp, Q1, Q2, and Q3 are calculated, the polar arrangement equations (12) to (21) are obtained. Thus, the control parameter calculation unit 28 is realized by an arithmetic unit that calculates b, c, d, b, c, d, Kp, Q1, Q2, and Q3 by inputting p1, p2, p3, and p4. The In addition, the stability determination unit 42 is realized as means for determining stability whether all the roots of the following equation (21) are within the unit circle.

  Therefore, by using such a method, it is possible to calculate a control parameter that realizes a control system that is resistant to disturbance by arranging the four poles p1, p2, p3, and p4 at the same position by using the pole arrangement formula. .

  However, it is difficult to intuitively understand the meaning of the calculated parameter, and the number of pole arrangement formulas is complicated as shown in formulas (4) to (10). Further, for example, a control parameter of a control system having a dead time element having a delay of one sampling time can be obtained from the above-described equations (4) to (10). Is complicated because it is necessary to use the equations (14) to (20) as the pole arrangement formula, and to select the pole arrangement formula depending on the size of the dead time element.

  Therefore, in one embodiment of the present invention, the control parameters calculated by the control parameter generation unit 28 described above are intuitively easy to understand, have a clear meaning, and are determined by the same algorithm even for systems with different dead times. We propose a possible method. That is, the control parameter generation unit 28 calculates a complex zero point based on the zero-cross frequency of the open-loop frequency characteristic. Specifically, the control parameter calculated from the pole placement formula is expressed by another simple approximate formula only when the four poles p1, p2, p3, and p4 are placed at the same position.

  That is, when the pole arrangement is limited to a position where a control system that is resistant to disturbance can be realized, the calculated control parameters are represented by the approximate expressions shown in the following expressions (1), (2), and (3).

  FIG. 5 shows the frequency characteristics of this control system and the conventional frequency characteristics. The solid line shown in FIG. 5 shows the frequency characteristic in this embodiment, and the broken line shows the conventional frequency characteristic. The vertical axis indicates the level (dB, deg), and the horizontal axis indicates the frequency (rad / s). In FIG. 5A, the constant a determines the lower limit frequency of the enhancement characteristic of the low frequency component, and the constant c determines the upper limit frequency of the enhancement characteristic of the low frequency component. The constant d determines the phase advance frequency at high frequency, the constant b determines the phase lag frequency at high frequency, and the constant k determines the gain of the entire control system. Usually, a, b, c, and d take real values, and a <c <d <b.

  On the other hand, in the embodiment according to the present invention, the frequency characteristic is expressed by a complex number, and the control parameter generation unit 28 designates a complex zero point based on the zero-cross frequency fx of the open-loop frequency characteristic. That is, the control parameters cs and ds are determined from the zero-cross frequency fx in the open loop frequency characteristic using the equations (1) and (2), and the provisional value of bs is determined with Br = 3 in the equation (3). . Using this provisional value, the open loop frequency characteristic is measured, and the gain is set so that the zero cross frequency matches the target fx. Furthermore, finest adjustment of the controller pole bs is finally performed to determine the best control parameter. Thereby, it is easy to understand intuitively, and the control parameter can be determined with the same algorithm even for systems with different dead times.

  First, the basis of the equations (1), (2), and (3) for determining the control parameter from the zero cross frequency fx will be described. For each case of including a dead time element with a delay of one sampling time, including a dead time element of a delay of two sampling times, and including a dead time element of a delay of three sampling times, four poles are stacked on various frequencies. The control parameters for the arrangement are calculated and shown in Table 1.

  At this time, the low-frequency cut-off frequency as corresponding to the low-frequency cut-off discretization angular frequency a is usually about 40 Hz or less and does not greatly affect the high-frequency characteristics of the control system. For example, when a sampling parameter is set to 100 kHz and a = 1.0 (as: equivalent to 0 Hz) and a control parameter for arranging the poles at 500 Hz is calculated, and only a is changed to a = 0.9975 (as: equivalent to 40 Hz). FIG. 6 shows the disturbance suppression characteristics. Although the low frequency of 100 Hz or less is different, almost the same characteristics can be obtained at 100 Hz or more. Therefore, for the sake of simplicity, in the present invention, the control parameter was examined with a = 1.0.

  Even if the low frequency cut-off frequency as is set to about 40 Hz according to the disk rotation frequency, the control characteristics of 100 Hz or higher are hardly affected as can be seen from FIG. Further, the gain Kp can be set so that the zero cross frequency matches the target fx from the open loop frequency characteristic as described above, and therefore only b, c, and d are considered here. The results in the case of one sampling time delay are shown in Table 1, the results in the case of two sampling time delays are shown in Table 2, and the results in the case of three sampling time delays are shown in Table 3.

  From Table 1, Table 2, and Table 3, if the pole position (Hz) is normalized by the sampling frequency and the normalized pole position is the same, the calculated control parameters b, c, and d are the same even if the sampling frequency is different. It turns out that it becomes.

  Therefore, it is understood that it is sufficient to grasp the behavior at each normalized frequency, not the absolute value of the sampling frequency or the pole position frequency. The following description will be made based on an example of a sampling frequency of 500 kHz, but it should be noted that if the value of the normalized frequency is the same, the same applies to other sampling frequencies, and the description can be made with generality.

  Next, the relationship between the frequency of pole arrangement and the zero cross frequency fx is confirmed. For example, FIG. 7 shows open loop frequency characteristics when the sampling frequency is 500 kHz and the pole position is 1 kHz. FIG. 7A shows gain (dB), FIG. 7B shows phase, the vertical axis shows level (dB, deg), and the horizontal axis shows frequency (rad / s). From FIG. 7, it can be seen that in this case, the zero cross frequency fx at which the gain is 0 dB is 1.42 kHz. Similarly, Table 4 shows the results of reading the relationship between the pole position and the zero cross frequency fx for each case of Table 1, Table 2, and Table 3 from the graph. The phase margin read from the graph is also included. In the case of a delay of 3 sampling times, the pole position 12.5 kHz is blank. This is because the pole that cannot be placed becomes unstable and the pole cannot be placed.

The control parameters are intuitively easy to understand when considered in the continuous system. Therefore, b, c, and d are converted into the continuous system, and further converted into a frequency expression and expressed as a ratio with the zero cross frequency fx. Specifically, b, c, and d in Tables 1, 2, and 3 are converted to a continuous system based on Z = ^{es sT} and expressed as bs, cs, ds, and conjugate zeros cs, ds are expressed as ( As shown in equations (22) and (23), the zero point frequency fr and the zero point angle θ of the controller are divided and expressed.

  The ratio of the zero point frequency fr and the zero cross frequency fx is set to the zero point frequency ratio Ar (= fr / fx), and the pole bs is converted to the frequency and the ratio to the zero cross frequency fx is taken to obtain the pole frequency ratio Br (= bs / 2π · fx). Are displayed in Table 5, Table 6, and Table 7. Both Ar and Br are controller parameters expressed as a ratio to the zero cross frequency fx, and the phase lead frequency and phase delay frequency in the conventional controller are set to 1/3 times and 3 times the zero cross frequency fx. It corresponds to being. A normalized zero cross frequency fxn obtained by normalizing the zero cross frequency fx with the sampling frequency is also shown.

  Based on Table 5, Table 6, and Table 7, a method for calculating cs and ds based on the equations (1) and (2) will be described first. When cs and ds in each case of Table 5, Table 6, and Table 7 are plotted on the complex plane, it can be confirmed that they are almost in a straight line as shown in FIG. Further, in comparison with the table, it can be seen that the greater the normalized zero cross frequency fxn, the greater the distance between the origin and the zero point, that is, the zero point frequency fr.

  Check in more detail. First, the angle between cs and ds plotted in FIG. 8 and the real axis, that is, the zero point angle θ in the equations (22) and (23) is graphed in FIG. Even if the standardized zero cross frequency fxn changes, it becomes a substantially constant value in the range of 143 ° to 145 °. Therefore, in the control parameter adjustment method according to the present invention proposed here, the parameters are determined so that the controller zero always exists on a straight line of 143 ° to 145 ° from the real axis regardless of the normalized zero cross frequency fxn. To do. Here, for example, it is 144 °. Next, a graph of the zero point frequency ratio Ar is shown in FIG. This is also substantially constant in the range of 0.28 to 0.3 even if the normalized zero cross frequency fxn changes. Therefore, in the control parameter determination method proposed this time, the zero point frequency fr is 0.28 to 0.3 times the zero cross frequency fx regardless of the standardized zero cross frequency fxn. Here, for example, it is 0.288 times.

  From the above, the controller zeros cs and ds can be approximately calculated by the equations (1) and (2) from the zero cross frequency fx with the zero point frequency ratio Ar = 0.288 and the zero point angle θ = 144 °.

  Next, a method for simply determining the provisional value of bs based on equation (3) will be described. The pole frequency ratio Br is plotted in FIG. When the standardized zero cross frequency fxn is small, bs is approximately three times fx and is about Br = 3. However, when the standardized zero cross frequency fxn is large or the dead time element is large, Br becomes large and Br = about 7. It turns out that it becomes large. This is considered to be because bs is calculated to be large in the pole placement formula in order to reduce the influence of the phase delay. Conversely, in the control parameter determination method according to the present invention, it is considered that Br should be set large enough to sufficiently reduce the influence of the phase delay. Therefore, Br is set to a provisional value of 3, the open-loop frequency characteristics are measured, and adjusted to a level where a sufficient phase margin can be secured. However, if it is too large, the high frequency noise becomes large. As shown in FIG. 11, Br = 3 to 7 is considered to be a realistic range.

  In order to confirm how much phase margin is sufficient, the relationship between the normalized zero cross frequency fxn and the phase margin in each case is plotted in FIG. It can be seen that it is in the range of 42 ° to 32 °. From this, it can be seen that the closed-loop poles can be arranged on the real axis even if the phase margin is 32 °. From the above, as a method of determining bs, the provisional value is Br = 3, servo is applied, the closed loop frequency characteristic is measured, and Br is increased so as to be 42 ° to 32 ° if the phase margin is small. If the phase margin cannot be larger than 32 ° even if Br <7 is increased, the control parameter cannot be calculated at the zero cross frequency fx, and fx is decreased.

From the above, the algorithm of the control parameter determination method according to the present invention is as follows.
1) The zero cross frequency fx to be realized is determined, and the cs and ds are determined from the equations (1) and (2) with the zero point frequency ratio Ar = 0.288 and the zero point angle θ = 144 °.
2) The controller pole is determined from bs = Br · 2πfx with the pole frequency ratio Br = 3.
3) Obtain a digital controller, apply servo, and adjust the controller gain Kp so that the zero cross frequency becomes fx.
4) Measure the phase margin. If it is not sufficient, gradually increase Br until the phase margin reaches 32 ° or more. If the phase margin of 32 ° cannot be secured when Br is 7 or less, the zero cross frequency fx is decreased.

  FIG. 13 shows a flowchart when this algorithm is realized by the servo control system shown in FIG. First, in step S1, the low-frequency cutoff frequency as of the controller is determined from the disk rotation frequency or the like. Next, in step S <b> 2, as is acquired from the input unit 44 of the control parameter generation unit 28. Next, in step S3, a target zero cross frequency fx is determined. Next, in step S4, fx is acquired from the input unit 44. Next, in step S5, the control parameter calculation unit 41 of the control parameter generation unit 28 determines the controller zeros cs and ds based on the equations (1) and (2). Next, in step S6, a provisional value is determined. Here, Br = 3 as a provisional value. In step S7, the determined provisional value Br (= 3) is acquired. Subsequently, in step S8, the control parameter calculation unit 41 calculates bs based on equation (3). In step S9, the calculated and input as, bs, cs, and ds are discretized in accordance with the sampling frequency, and control parameters a, b, c, and d of the digital controller are calculated. In step S <b> 10, a provisional value of the controller gain Kp is acquired from the input unit 44. In step S 11, the control parameters a, b, c, d, and Kp are transferred from the control parameter generation unit 28 to the control arithmetic circuit 53. Next, in step S12, the servo is applied to measure the open loop frequency characteristics. In step S13, the controller gain Kp is determined so that the measured zero-cross frequency of the open-loop frequency characteristic matches the target zero-cross frequency fx. In step S14, the phase margin is confirmed from the open loop frequency characteristics. If the margin is sufficient (from 42 ° to 32 °) (step S14, yes), the control parameter at this time is determined as the final control parameter. . If the phase margin is not sufficient (less than 32 °) (step S14, no), the value of Br is set large in step S15. In step S16, bs is calculated from equation (3) to measure the open-loop frequency characteristics, and the process is repeated until the phase margin is sufficient. However, if the phase margin of 32 ° cannot be ensured even if Br exceeds 7 in step S16, the target zero-cross frequency fx is inappropriate, and therefore, in step S17, fx is reduced and repeated again. .

  As described above, in the present invention, the provisional value of the control parameter is determined from the zero cross frequency fx in the open loop frequency characteristic, and the gain is set so that the zero cross frequency matches the target fx according to the open loop frequency characteristic. . As a result, an intuitively easy-to-understand control parameter can be obtained, and even when the control parameter is inappropriate, it can be easily detected. Further, the finest control parameters are finally determined by performing fine adjustment so as to realize a sufficient phase margin. This makes it possible to determine parameters with the same algorithm even for systems with different dead times.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a configuration diagram illustrating the entirety including a servo control system of an optical disc recording / reproducing apparatus shown as an embodiment of the present invention. It is a block diagram when the control system which comprises the automatic control apparatus shown as an Example of this invention is implement | achieved by digital control. FIG. 3 is a configuration diagram illustrating a control system in which the control target P (s) and the zeroth-order holder in FIG. FIG. 4 is a block diagram showing a control system in the case where there is a dead time with a delay of 3 sampling times in FIG. It is a characteristic view explaining the frequency characteristic of the control system shown as one Example of this invention. It is a characteristic view which shows the disturbance suppression characteristic when the low frequency cut-off discretization angular frequency a is changed. It is a characteristic view which shows an open loop frequency characteristic when a pole position is 1 kHz. It is a figure which shows the position on the complex plane of the controller zero point cs in each case of 1, 2, 3 sampling time delay. It is a figure which shows the zero point angle of the controller zero in each case of 1, 2, 3 sampling time delay. It is a figure which shows the relationship of ratio Ar of the zero cross frequency with respect to the zero point frequency of the controller zero point in each case of 1, 2, 3 sampling time delay. It is a figure which shows the relationship of ratio Br of the zero cross frequency with respect to the controller pole bs in each case of 1, 2, 3 sampling time delay. It is a figure which shows the relationship of the phase margin with respect to the normalization zero cross frequency in each case of 1, 2, 3 sampling time delay. It is a flowchart explaining the control parameter determination process in the control system shown as one Example of this invention.

Explanation of symbols

2 recording / reproducing unit, 3 optical disc recording / reproducing device, 4 optical disc, 31 recording / reproducing element, 15 control operation unit, 28 control parameter generating unit, 32 laser light source, 35 light receiving unit, 33 light separating unit, 34 lens, 36 electromagnetic actuator, 52 AD converter, 53 control arithmetic circuit, 54 DA converter, 55 actuator drive circuit

Claims (10)

A recording / reproducing element for writing data to the disk-shaped recording medium or reading data recorded on the disk-shaped recording medium;
Driving means for driving the recording / reproducing element in a horizontal direction or a vertical direction with respect to a recording surface of the disc-shaped recording medium;
An error signal detecting means for detecting an error signal that is proportional to a difference between a position where the recording / reproducing element should perform recording and reproduction and an actual position and detects an error signal corresponding to the difference signal;
Control calculation means for performing a control calculation on the error signal detected by the error signal detection means using a control parameter and calculating a control output for making the error signal as small as possible;
Control parameter calculation means for calculating a control parameter for realizing the servo performance required for the servo control system constituted by the drive means, the error signal detection means, and the control calculation means;
A servo control device characterized in that the control parameter calculation means calculates a complex zero from a zero cross frequency of an open loop frequency characteristic that is an index representing servo performance. In the control parameter calculation means, when the zero cross frequency is fx, the filter in the control calculation means has a conjugate zero point expressed by the expressions (1) and (2), and the expressions (1), ( 2. The servo control device according to claim 1, wherein the zero point frequency ratio Ar in the equation (2) is 0.28 to 0.3, and the zero point angle [theta] is 143 to 145 degrees.

In the control parameter calculation means, when the zero cross frequency is fx, the filter in the control calculation means has a pole set based on the expression (3), and the pole frequency ratio Br in the expression (3) is 3 The servo control device according to claim 1, wherein the servo control device is.

In drive system control, a control parameter is calculated for an error signal that is proportional to the difference between the position to be localized and the actual position and that corresponds to the difference signal, and a control calculation is performed using the control parameter to obtain an error signal. In the control parameter determination method for drive control so as to reduce as much as possible,
An acquisition step of acquiring an open loop zero cross frequency in a servo control system, a pole frequency ratio that is a ratio of a controller pole and a zero cross frequency, and a controller gain;
A first calculation step of calculating a controller zero using at least the acquired zero-cross frequency;
A second calculation step of calculating a controller pole based on the acquired pole frequency ratio and zero cross frequency;
Obtaining an open loop frequency characteristic calculated from the calculated controller zero and controller pole;
An acquisition step of re-acquiring a controller gain according to the measurement result of the open loop frequency characteristic. In the first calculation step, when the zero cross frequency is fx, there are conjugate zeros represented by the equations (1) and (2), and the zeros in the equations (1) and (2) 5. The control parameter determination method according to claim 4, wherein the frequency ratio Ar is 0.28 to 0.3 and the zero point angle θ is 143 to 145 degrees.

In the second calculation step, when the zero cross frequency is fx, the pole is set based on the equation (3), and the pole frequency ratio Br in the equation (3) is set to 3-7. The control parameter determination method according to claim 4, wherein:

  5. The control parameter determination method according to claim 4, further comprising a step of adjusting a controller gain so that the obtained zero-cross frequency of the open-loop frequency characteristic matches a target zero-cross frequency fx.   Check the phase margin from the open-loop frequency characteristics, and if the phase margin is between 42 ° and 32 °, determine the zero cross frequency, the pole frequency ratio, and the controller gain at this time as control parameters. The control parameter determination method according to claim 4.   If the phase margin of 42 ° to 32 ° cannot be ensured even if Br exceeds 7 in the above formula (3), the gain of the zero cross frequency fx in the open loop frequency characteristic is reduced and the setting of the control parameter is repeated. The control parameter determination method according to claim 8. In order to write data to the disk-shaped recording medium or read data recorded on the disk-shaped recording medium, the recording / reproducing element is driven horizontally or vertically with respect to the recording surface of the disk-shaped recording medium. In drive control, computer control is used to calculate a control parameter for an error signal that is proportional to the difference between the position at which the recording / reproducing element should perform recording / reproduction and the actual position and that corresponds to the difference signal. For servo control devices that perform control calculations and perform control output to minimize the error signal,
An acquisition step of acquiring an open loop zero cross frequency in a servo control system, a pole frequency ratio that is a ratio of a controller pole and a zero cross frequency, and a controller gain;
A first calculation step of calculating a controller zero using at least the acquired zero-cross frequency;
A second calculation step of calculating a controller pole based on the acquired pole frequency ratio and zero cross frequency;
Obtaining a measurement result of open loop frequency characteristics from the calculated controller zero and controller pole;
A control parameter determination program that executes a control parameter determination process including: an acquisition step of reacquiring a controller gain according to the measurement result of the open loop frequency characteristic.

Images