JP3707346B2 - Brake control device for vehicle - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a vehicle brake control device applied to a regenerative cooperative brake control system or the like that efficiently recovers regenerative energy by controlling pressure reduction of brake hydraulic pressure while controlling regenerative brake torque.
[0002]
[Prior art]
As a conventional vehicle brake control device, the one described in Japanese Patent Laid-Open No. 9-247552 can be cited.
[0003]
This is because the driver's deceleration request amount is calculated from the operation amount of the brake operation member, the brake hydraulic pressure command value is determined so that the actual vehicle deceleration matches this, and the actually measured brake hydraulic pressure is further commanded. The current of the electromagnetic solenoid valve is feedback-controlled so as to match the value. This electromagnetic solenoid valve is connected to an accumulator (a high pressure is secured by a pump) and a reservoir tank (low pressure), and adjusts the hydraulic pressure to a level necessary for suppressing the rotation of the wheel and supplies it to the wheel cylinder. The electromagnetic solenoid valve may be composed of two parts, a pressure-increasing valve and a pressure-reducing valve, or may use a pressure-increasing / reducing valve that integrates these functions into one.
[0004]
However, in such a conventional apparatus, the dead zone of the flow rate characteristic with respect to the current value that the electromagnetic solenoid valve has greatly affects the transient characteristics of the hydraulic control. The electromagnetic solenoid valve normally has a dead band characteristic that the valve does not open and the oil does not actually flow unless a certain amount of current is passed due to the relationship between the spring force, the electromagnetic force, and the fluid force. This dead zone tends to increase as the solenoid valve is designed to be particularly inexpensive and small (see FIG. 8).
[0005]
Even if you configure a hydraulic feedback control system that measures the actual wheel cylinder pressure with a pressure sensor and compares it with the hydraulic pressure command value to control the current value of the solenoid valve, When the control is reversed from pressure reduction to pressure increase and pressure decrease to pressure increase, the feedback valve controller uses the integral characteristics until the current command value of the solenoid valve is integrated to a value that exceeds the dead zone, and the oil pressure does not actually flow. Does not follow the command value at all. That is, in the hydraulic feedback control system, unnecessary dead time occurs when the hydraulic pressure command value is reversed.
[0006]
This is the case when an ON / OFF solenoid valve (current / flow rate characteristic is steep) is used as the solenoid valve, or when a proportional solenoid valve (current / flow rate characteristic is relatively gentle) is used. Is a common problem.
[0007]
In order to solve the above problems, the present applicant has proposed a vehicle brake control device described in Japanese Patent Application Laid-Open No. 2000-071959.
[0008]
In this conventional publication, the current is gradually increased in a vehicle brake control device that feedback-controls the wheel cylinder pressure using an electromagnetic solenoid valve that opens in proportion to the current and changes the amount of brake oil. The dead band width at which the solenoid solenoid valve opens for the first time is learned from the current command value of the solenoid solenoid valve and the wheel cylinder pressure.After learning, the wheel cylinder command pressure and actual wheel cylinder pressure are quickly corrected by offset correction. With features that match
[0009]
[Problems to be solved by the invention]
However, in the above conventional vehicle brake control device, the current supplied to the electromagnetic solenoid valve is gradually increased, and the dead band width is calculated from the current command value when the wheel cylinder pressure changes by a predetermined amount. Since it is configured to learn and offset it after learning, the solenoid characteristics change over time by continuing to energize the electromagnetic solenoid valve, and the dead band width deviates from the first learned value. In this case, since the deviation is compensated by the feedback control (time t3 to t4 in FIG. 9), the wheel cylinder pressure does not become uncontrollable. However, since the dead band width of the pressure increasing / decreasing valve continues to be corrected by feedback control while repeating the pressure increase / decrease in the meantime, in the next learning of the dead band, the one-side valve (pressure increasing valve at time t4 in FIG. 9) There is a problem that the current command value of the valve to be learned (the pressure reducing valve at time t4 in FIG. 9) is erroneously learned as the dead band width (time t4 in FIG. 9) in a state where it is not completely closed.
[0010]
The present invention has been made paying attention to such a conventional problem, and when the dead zone width of the electromagnetic solenoid valve deviates from the first learning value, the feedback control compensates for the deviation and repeats the pressure increase / decrease However, it aims at providing the brake control apparatus for vehicles which can learn a dead zone width correctly.
[0011]
[Means for Solving the Problems]
  In order to achieve the above object, according to the first aspect of the present invention, there is provided a brake operating member by a driver.Manipulation amountBased on the brake hydraulic pressure command value calculating means for calculating the hydraulic pressure command value acting on the brake to suppress the rotation of the wheel,
Brake hydraulic pressure measuring means for measuring an actual value of hydraulic pressure acting on a brake for suppressing rotation of the wheel;
In order to match the brake hydraulic pressure measurement value with the brake hydraulic pressure command value, both values are input, and a hydraulic feedback control calculation means for calculating a current command value to the electromagnetic solenoid valve;
Electromagnetic solenoid valve current control means for performing current control on the electromagnetic solenoid valve based on the calculated current command value;
A dead band width learning means for learning the dead band width of the electromagnetic solenoid valve based on a current command value when a predetermined amount of change occurs in the brake hydraulic pressure measurement value by gradually increasing the current supplied to the electromagnetic solenoid valve. Prepared,
In the vehicle brake control device that uses the learned dead zone width as an offset correction value of the current command value for current control of the electromagnetic solenoid valve,
The brake hydraulic pressure command valueDifferenceIs provided within the predetermined range for a predetermined time, and when it changes for a predetermined amount for the first time, an initialization means for initializing the internal variable of the hydraulic feedback control calculation means is provided,
The dead band width learning unit is a unit that learns the dead band width of the electromagnetic solenoid valve in synchronization with the initialization by the initialization unit.
[0012]
According to a second aspect of the present invention, in the vehicle brake control device according to the first aspect, the dead zone width learning means is independent of initialization by the initialization means at the time of first learning immediately after the start of brake control. It is a means for learning the dead zone width of the electromagnetic solenoid valve.
[0013]
Operation and effect of the invention
  In the invention according to claim 1, in the brake hydraulic pressure command value calculation means, the brake operation member by the driverManipulation amountThe hydraulic pressure command value acting on the brake that suppresses the rotation of the wheel is calculated based on the hydraulic pressure, and the actual value of the hydraulic pressure acting on the brake that suppresses the rotation of the wheel is measured by the brake hydraulic pressure measuring means. In order to make the brake hydraulic pressure measured value coincide with the brake hydraulic pressure command value, the both values are input to calculate the current command value to the electromagnetic solenoid valve, and the electromagnetic solenoid valve current control means calculates the calculated current. Current control for the electromagnetic solenoid valve is performed based on the command value.
In this feedback control, the dead zone width learning means gradually increases the current supplied to the electromagnetic solenoid valve, and the dead zone of the electromagnetic solenoid valve is based on the current command value when a predetermined amount of change occurs in the brake hydraulic pressure measurement value. The width is learned, and the learned and stored dead band width is used for the current control of the electromagnetic solenoid valve as the offset correction value of the current command value.
On the other hand, in the initialization means, the brake hydraulic pressure command valueDifferenceWhen the pressure changes within a predetermined range for a predetermined time and changes by a predetermined amount for the first time, the internal variable of the hydraulic feedback control calculation means is initialized.
Therefore, the dead band width learning unit learns the dead band width of the electromagnetic solenoid valve in synchronization with the initialization by the initialization unit.
Therefore, when the solenoid characteristics change over time by continuing to energize the electromagnetic solenoid valve, and the dead band width deviates from the initial learning, even if feedback control compensates for the deviation and repeats increasing and decreasing pressure, the feedback Since learning is performed in synchronization with the initialization of the control, there is an effect that the dead band width can be correctly learned.
[0014]
In the invention according to claim 2, the dead zone width learning means learns the dead zone width of the electromagnetic solenoid valve at the first learning immediately after the start of the brake control irrespective of the initialization by the initialization means.
That is, if the initialization by the initialization means is always included in the condition for learning the dead band, the dead band of the electromagnetic solenoid valve cannot be learned at the first opportunity immediately after the start of the brake control.
On the other hand, in the first learning immediately after the start of the brake control, there is an effect that the dead zone width of the electromagnetic solenoid valve can be learned at the first opportunity immediately after the start of the brake control by irrelevant to the initialization.
[0015]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, embodiments of the present invention will be described with reference to the drawings.
[0016]
FIG. 1 is a conceptual diagram showing a vehicle brake control apparatus according to the present invention, in which a is a brake hydraulic pressure command value calculating means, b is a brake hydraulic pressure measuring means, c is a hydraulic feedback control calculating means, and d is an electromagnetic solenoid valve current control means. , E is a dead band width learning means, and f is an initialization means. In the dead band width learning means e, except for the first learning immediately after the start of brake control, the electromagnetic solenoid valve is synchronized with the initialization by the initialization means. By learning the dead band width, when the dead band width deviates from the initial learning, the dead band width can be correctly learned even when the feedback control compensates for the deviation and repeats increasing and decreasing pressure.
[0017]
FIG. 2 is an overall system diagram showing the vehicle brake control device according to the embodiment of the present invention. This is because the vehicle brake control device of the present invention is applied to a “regenerative cooperative brake control system” that efficiently recovers regenerative energy by controlling the brake hydraulic pressure to be reduced during regenerative brake torque control by an AC synchronous motor. Is.
[0018]
To explain the configuration, reference numeral 1 denotes a brake pedal operated by a driver, which is connected to 2 hydraulic boosters and 3 master cylinders (hereinafter, the master cylinder is referred to as M / CYL). The hydraulic booster 2 uses the high pressure (sequence-controlled by the pressure switch 21) accumulated in the accumulator 19 by the pump 18 to boost the brake pressure and supply it to the M / CYL 3. This high pressure is also used as a source pressure for hydraulic feedback control. 4 and 5 are oil passage switching valves, and both valves are controlled synchronously. The figure shows a state when power is not supplied, and M / CYL3 oil is supplied to the wheel cylinder 7 as it is (normal mode).
[0019]
When energized, the M / CYL 3 is connected to the 6 stroke simulator (oil load equivalent to the wheel cylinder 7) to ensure the same brake pedal feeling as in the normal mode. At the same time, the wheel cylinder 7 (hereinafter, the wheel cylinder is referred to as W / CYL) is connected to the solenoid valves 10 and 11 for hydraulic control, and is disconnected from the M / CYL 3.
[0020]
A pressure increasing valve 10 is located between the high pressure line and the W / CYL 7 and adjusts the amount of oil flowing into the W / CYL 7. A pressure reducing valve 11 is located between the W / CYL 7 and the reservoir (low pressure), and adjusts the amount of oil flowing out of the W / CYL 7. The hydraulic pressure of W / CYL 7 is controlled using a set of solenoid valves 10 and 11.
[0021]
Reference numeral 8 denotes an M / CYL pressure sensor that measures the M / CYL pressure (the amount of braking required by the driver). Reference numeral 9 denotes a W / CYL pressure sensor that measures the W / CYL pressure (for feedback control). Reference numeral 20 denotes an original pressure sensor that measures an accumulator pressure (original pressure).
[0022]
A brake control controller 12 includes a CPU, a ROM, a RAM, a digital port, an A / D port, a one-chip microcomputer incorporating various timer functions (or a plurality of chips realizing the same function), a high-speed communication circuit, The brake control controller 12 is constituted by each actuator drive circuit and each function of each means shown in the conceptual diagram (FIG. 1) of the present invention is achieved. FIG. 2 shows a configuration related to W / CYL 7 for one wheel, but the other three wheels have the same configuration and are not shown.
[0023]
Reference numeral 15 denotes an AC synchronous motor connected to the drive wheels 16 of the vehicle via a speed reduction mechanism, and collects vehicle kinetic energy to the battery 17 by drive torque control and regenerative brake control. Reference numeral 14 denotes a DC / AC conversion current control circuit, which is located between the DC battery and the AC synchronous motor 15, and based on the three-phase PWM signal from the motor controller 13, generates an AC current and a DC current. Perform conversion. The motor controller 13 controls the regenerative brake torque based on the regenerative brake torque command value received from the 12 brake control controllers through communication. Further, drive torque control by the AC synchronous motor 15 is performed during driving. Furthermore, the maximum allowable regenerative torque value determined by the state of charge, temperature, etc. of the battery 17 is calculated and transmitted to the brake controller 12 via communication.
[0024]
3, 4, and 5 are flowcharts showing control operations performed by the microcomputer of the brake controller 12, and the routine shown in FIG. 3 is executed at regular intervals (for example, 5 msec). Note that (k) represents the current computation, and (k−n) represents the previous computation n times.
[0025]
In S10, the signals of the M / CYL pressure sensor 8, the W / CYL pressure sensor 9, and the original pressure sensor 20 are measured using an A / D converter built in the microcomputer, converted into a predetermined physical unit, and M / CYL The pressure Pmc (k), the W / CYL pressure Pwc (k), and the original pressure Pline (k) are calculated.
[0026]
In S20, the maximum allowable regenerative motor torque Tmmax (k) is read from the high-speed communication reception buffer with the motor control controller 13.
[0027]
In S30, the driver braking torque request value Tbdem (k) is calculated from the following equation using the M / CYL pressure Pmc (k) and the vehicle specification constant K1 stored in advance in the ROM.
Tbdem (k) = Pmc (k) × constant K1
Constant K1 = W / CYL area × brake pad area × brake rotor effective radius × brake friction coefficient
[0028]
In S40, it is determined whether to perform regenerative cooperative brake control. If the following equation is satisfied, the process proceeds to S50, and if not, the process proceeds to S310.
M / CYL pressure Pmc (k) ≧ constant K2, and maximum allowable regenerative motor torque Tmmax (k) ≧ constant K3
[0029]
A case where regenerative cooperative brake control is not performed first will be described.
In S310, the output process is performed when the regeneration cooperative brake control is stopped (that is, when the hydraulic feedback control is stopped). D / A output is used to output the port so that the oil passage switching valves 4 and 5 are in the normal mode, and to output the current command values (= 0) of the pressure increasing valve 10 and the pressure reducing valve 11 to each current control circuit. And a transmission process for transmitting the regenerative motor torque command value (= 0) to the motor controller 13 using high-speed communication. One control routine is completed.
[0030]
The following is a description of when regenerative cooperative brake control is performed.
In S50, the driver braking torque request value Tbdem (k) is distributed and calculated between the hydraulic brake torque command value Tbcom (k) and the regenerative brake torque command value Tmcom (k).
When Tbdem (k) ≧ Tmmax (k)
Tbcom (k) = Tbdem (k) −Tmmax (k), Tmcom (k) = Tmmax (k)
When Tbdem (k) <Tmmax (k)
Tbcom (k) = 0, Tmcom (k) = Tbdem (k)
Further, the brake hydraulic pressure command value Pcom (k) is calculated using the hydraulic brake torque command value Tbcom (k) and the vehicle specification constant K1.
Pcom (k) = Tbcom (k) ÷ constant K1
Constant K1 = W / CYL area × brake pad area × brake rotor effective radius × brake friction coefficient
[0031]
In S60, the brake pressure command value difference ΔPcom (k) is calculated, and the process proceeds to S70.
ΔPcom (k) = Pcom (k) −Pcom (k−1)
[0032]
In S70, it is determined whether or not there is a predetermined change in the brake hydraulic pressure command value. If ΔPcom (k) ≦ predetermined value K10, the process proceeds to S80. Otherwise, the process proceeds to S90.
[0033]
In S80, assuming that there is no predetermined change in the brake hydraulic pressure command value, in order to measure the elapsed time, the time measurement counter CNT1 is incremented (CNT1 (k) = CNT1 (k-1) +1), and the process proceeds to S120.
[0034]
In S90, the elapsed time is examined assuming that there is a predetermined change in the brake hydraulic pressure command value. If CNT1 (k)> predetermined time K11, the process proceeds to S100. Otherwise, the process proceeds to S110.
[0035]
In S100, it is determined that the brake hydraulic pressure command value has started to move for the first time after the brake hydraulic pressure command value has not been changed for a predetermined time, and initialization of the internal variable of the feedback compensator and learning of the dead zone of the proportional valve are performed. I need. CNT1 (k) = 0, FRESET = 1, FSTUDY = 1. Proceed to S120.
[0036]
In S110, it is determined that the state in which the brake hydraulic pressure command value has not been changed does not continue for a predetermined time and the brake hydraulic pressure command value has started to move, and it is necessary to initialize an internal variable of the feedback compensator and learn the dead zone of the proportional valve. do not do. CNT1 (k) = 0. Proceed to S120.
[0037]
In S120, an initialization flag is determined. If FRESET = 1, the process proceeds to S130. If FRESET = 0, the process proceeds to S140.
[0038]
In S130, an internal variable of the feedback compensator and an initialization flag are initialized. Xr (k) = 0, Xf (k) = 0, Xμ (k) = 0, FRESET = 0. Proceed to S140.
[0039]
In S140 to S170, as shown in FIG. 6, a hydraulic control compensator using a known “two-degree-of-freedom control structure” and “μ synthesis” is calculated. Note that the initial values of the internal variables Xr (0), Xf (0), and Xμ (0) are all zero.
[0040]
In S140, the reference model W / C pressure Pref (k) is calculated using the target W / C pressure Pcom (k) as an input. Equation (1) is a discrete-time state equation, and (k + 1) represents a value after one sample period. Xr (k) is a variable vector, and Ar, Br, Cr, and Dr are constant matrices previously stored in the ROM. Proceed to S150.
Xr (k + 1) = Ar.Xr (k) + Br.Pcom (k) (1)
Pref (k) = Cr · Xr (k) + Dr · Pcom (k)
[0041]
In S150, the feedforward compensator (= reference model / plant model) is calculated with the target W / C pressure Pcom (k) as an input. Proceed to S160.
Xf (k + 1) = Af.Xf (k) + Bf.Pcom (k) (2)
Icom_ff (k) = Cf · Xf (k) + Df · Pcom (k)
[0042]
In S160, the deviation compensator ΔPwc (k) between the reference model W / C pressure Pref (k) and the W / C pressure Pwc (k) is input, and a feedback compensator (designed in advance using μ synthesis) is calculated. The process proceeds to S170.
ΔPwc (k) = Pref (k) −Pwc (k) (3)
Xμ (k + 1) = Aμ · Xμ (k) + Bμ · Pwc (k)
Icom_fb (k) = Cμ · Xμ (k) + Dμ · Pwc (k)
[0043]
In S170, after adding the output Icom_ff (k) of the feedforward compensator and the output Icom_fb (k) of the feedback compensator, the electromagnetic valve current command value Icom (k) is limited by the upper and lower limit values (± Imax). calculate. The process proceeds to S180.
Icom (k) = Icom_ff (k) + Icom_fb (k)
[0044]
In S180, in order to eliminate the influence of the difference in the electromagnetic valve flow rate characteristics due to the differential pressure, a correction coefficient corresponding to the differential pressure is calculated from the table data for each of the pressure increasing / decreasing electromagnetic valves 10 and 11. The constant Pcom_atm corresponds to atmospheric pressure. Proceed to S190.
K_in (k) = TABLE_in (Pline (k) −Pwc (k))
K_out (k) = TABLE_out (Pwc (k) −Pcom_atm)
[0045]
In S190, a learning flag is determined. If FSTUDY = 1, the process proceeds to S200. If FSTUDY = 0, the process proceeds to S210. In the learning opportunity after the start of control, the initial value of the learning flag is set to FSTUDY (0) = 1 so that the learning can be performed without synchronizing with the initialization of the feedback control.
[0046]
  First, the non-learning process will be described.
In S210, using the offset values OFFSET_in (k) and OFFSET_out (k) corresponding to the dead zones (see FIG. 8) of the pressure increasing and pressure reducing valves 10, 11, the current command value I_in (k) for the pressure increasing and pressure reducing valves, I_out (k) is offset corrected. In order to cope with the dead zone width before learning, the representative values of the dead zones of the pressure increasing and depressurizing valves 10 and 11 are set as initial values in OFFSET_in (0) and OFFSET_out (0). Proceed to S240.
I_in (k) = OFFSET_in (k) + Icom (k) ・ K_in (k)
However, 0 ≦ I_in ≦ constant K4
I_out (k) = OFFSET_out (k) + Icom (k) ・ K_out (k)
However, 0 ≦ I_out ≦ constant K5
The constant K4 is a current value for obtaining the maximum flow rate of the pressure increasing valve 10, and the constant K5 isDecompressionThe current value when obtaining the maximum flow rate of the valve 11.
[0047]
In S240, output processing is performed during regenerative cooperative brake control. To perform port output so that the oil passage switching valves 4 and 5 are set to the control mode, and to output the current command values I_in (k) and I_out (k) of the pressure increasing valve 10 and the pressure reducing valve 11 to each current control circuit. D / A output is performed, and a transmission process for transmitting the regenerative motor torque command value to the motor controller 13 using high-speed communication is performed. One control routine is completed.
[0048]
Next, the learning time will be described.
In S200, it is determined whether the pressure is increased or decreased. When Icom (k) ≧ 0, the process proceeds to S220 as pressure increase. If Icom (k) <0, the process proceeds to S230 as decompression.
[0049]
In S220, since the pressure increasing valve 10 is energized and the pressure reducing valve 11 is closed, the electromagnetic valve current command value after the differential pressure correction is used as the current value of the pressure increasing valve 10. Proceed to S250.
I_in (k) = Icom (k) ・ K_in (k) I_out (k) = 0
[0050]
In S230, since the pressure reducing valve 11 is energized and the pressure increasing valve 10 is closed, the electromagnetic valve current command value after the differential pressure correction is set as the current value of the pressure reducing valve 11. Proceed to S250.
I_out (k) = | Icom (k) · K_in (k) | I_in (k) = 0
[0051]
In S250, output processing is performed during regenerative cooperative brake control. To perform port output so that the oil passage switching valves 4 and 5 are set to the control mode, and to output the current command values I_in (k) and I_out (k) of the pressure increasing valve 10 and the pressure reducing valve 11 to each current control circuit. D / A output is performed, and a transmission process for transmitting the regenerative motor torque command value to the motor controller 13 using high-speed communication is performed. Proceed to S260.
[0052]
In S260, the difference ΔPwc (k) of the W / C pressure is calculated. Proceed to S270.
ΔPwc (k) = Pwc (k) −Pwc (k−1)
[0053]
In S270, in order to learn the dead zone width of the proportional valve, it is determined whether or not there is a predetermined change in the W / C pressure. If ΔPwc (k)> predetermined value K12, the process proceeds to S280. Otherwise, one control routine is terminated.
[0054]
In S280, it is determined whether the pressure is increased or decreased. When Icom (k) ≧ 0, the process proceeds to S290 as pressure increase. When Icom (k) <0, the process proceeds to S300 as decompression.
[0055]
In S290, the current command value of the pressure increasing valve 10 is learned as the dead band width of the pressure increasing valve 10. From the next control routine, since there is an offset correction of the dead band width in S210, the offset correction overlaps with Icom (k) including the dead band width calculated up to this time. In order to prevent this, the initialization flag FRESET = 1 is set so that the feedback control can be initialized in the next control routine. Therefore, learning is not performed in synchronization with this initialization. One control routine is completed.
OFFSET_in (k) = I_in (k), FSTUDY = 0, FRESET = 1
[0056]
In S300, the current command value of the pressure reducing valve 11 is learned as the dead zone width of the pressure reducing valve 11. From the next control routine, since there is an offset correction of the dead band width in S210, the offset correction overlaps with Icom (k) including the dead band width calculated up to this time. In order to prevent this, the initialization flag FRESET = 1 is set so that the feedback control can be initialized in the next control routine. Therefore, learning is not performed in synchronization with this initialization. One control routine is completed.
OFFSET_out (k) = I_out (k), FSTUDY = 0, FRESET = 1
[0057]
Next, operational effects will be described.
First, the current control for the pressure increasing / decreasing valves 10 and 11 is performed in the flow of S10 → S20 → S30 → S40 in FIG. 3 and the flow from S40 to S50 to S300 when the regeneration cooperative brake control is performed.
[0058]
Basically, the W / CYL pressure Pwc (k) is calculated in S10, and the brake hydraulic pressure command value Pcom (k) is calculated using the hydraulic brake torque command value Tbcom (k) and the vehicle specification constant K1 in S50. ) Is calculated, and after adding the output Icom_ff (k) of the feedforward compensator and the output Icom_fb (k) of the feedback compensator in S140 to S170, it is limited by the upper and lower limit values (± Imax), and the electromagnetic valve current The command value Icom (k) is calculated, and current control is performed on the pressure increasing / decreasing valves 10 and 11 that are electromagnetic solenoid valves based on the calculated electromagnetic valve current command value Icom (k).
[0059]
In the current control by the feedforward compensation + feedback compensation, the dead band width of the pressure increasing / decreasing valves 10 and 11 is learned in synchronization with the initialization of the internal variable of the feedback control. In other words, when the brake oil pressure command value Pcom (k) changes within a predetermined time for a predetermined time and then changes by a predetermined amount for the first time, the pressure increasing / reducing valves 10, 11 are synchronized with the initialization of the internal variable of the feedback control. 4, when the process proceeds from S60 → S70 → S90 → S100 in FIG. 4, in S100, the reset flag FRESET is set to FRESET = 1, the learning flag FSTUDY is set to FSTUDY = 1, and FRESET is set. When S = 1, the process proceeds from S120 to S130, and internal variables for feedback control are initialized.
[0060]
In S190 of FIG. 5, if FSTUDY = 1, the process proceeds to S200 and subsequent steps, and the pressure increasing valve 10 or pressure reducing valve 11 to be learned is gradually energized, so that the pressure reducing valve 11 or pressure increasing valve 10 on one side is completely closed. In the closed state (S200, S220, S230), when the brake hydraulic pressure changes by a predetermined amount (S270), the dead band width of the pressure increasing valve 10 or the pressure reducing valve 11 is learned (S280, S290, S300).
[0061]
Therefore, when the energization of the pressure increasing valve 10 and the pressure reducing valve 11 is continued, the solenoid characteristics change with time, and when the dead band width deviates from the first learning, the feedback control repeats the pressure increasing / decreasing to compensate for the deviation. Even in such a case, since the learning is performed in synchronization with the initialization of the feedback control, there is an effect that the dead band width can be correctly learned.
[0062]
That is, in the dead band learning of the pressure reducing valve 11 shown in FIG. 7, the dead band widths of the pressure increasing and decreasing valves 10 and 11 are corrected by feedback control from time t3 to time t4 while repeating pressure increasing and decreasing. Since the dead zone width is learned in synchronization with the time t4 when control initialization is performed, the current command value Iout (k) of the pressure reducing valve 11 to be learned is obtained with the pressure increasing valve 10 on one side completely closed. Since learning is performed, the dead zone width can be correctly learned at the time t5 in FIG.
[0063]
Further, in S190 of FIG. 5, it is determined whether or not FSTUDY = 1, but since the initial value of the learning flag FSTUDY is set to FSTUDY (0) = 1, at the first learning immediately after the start of the brake control, Regardless of the initialization of the feedback control, the dead zone width of the pressure increasing / decreasing valves 10 and 11 is learned.
[0064]
That is, if the initialization condition of the feedback control is always included in the dead band width learning condition, the dead band width of the pressure increasing / decreasing valves 10 and 11 cannot be learned at the first opportunity immediately after the start of the brake control.
[0065]
On the other hand, at the time of the first learning immediately after the start of the brake control, the initial value of the learning flag FSTUDY is set to FSTUDY (0) = 1 and is irrelevant to the initialization. There is an effect that the dead zone width of the pressure increasing / decreasing valves 10 and 11 can be learned at an opportunity.
[0066]
Although the embodiment of the present invention has been described above, the specific configuration is not limited to this embodiment, and is included in the present invention unless the gist described in the claims is changed.
[0067]
For example, in the embodiment, the application example to the regenerative cooperative brake control device has been described. However, the wheel cylinder pressure is fed back using an electromagnetic solenoid valve that opens in proportion to the current and changes the amount of brake oil. The brake control system can be applied to other systems.
[Brief description of the drawings]
FIG. 1 is a conceptual diagram showing a vehicle brake control device of the present invention.
FIG. 2 is an overall system diagram illustrating the vehicle brake control device according to the first embodiment.
FIG. 3 is an overall flowchart showing a control operation performed by a microcomputer of a brake control controller of the apparatus according to the first embodiment.
FIG. 4 is a flowchart of S60 to S130 showing a control operation performed by a microcomputer of a brake control controller of the apparatus according to the first embodiment.
FIG. 5 is a flowchart of S190 to S300 showing a control operation performed by the microcomputer of the brake control controller of the apparatus according to the first embodiment.
FIG. 6 is a control block diagram showing a configuration of a two-degree-of-freedom control system.
7 is a time chart showing dead band width learning of the pressure reducing valve in the brake control in Embodiment 1. FIG.
FIG. 8 is a current-flow rate characteristic diagram showing dead band widths of a pressure increasing valve and a pressure reducing valve.
FIG. 9 is a time chart showing dead band learning of a pressure reducing valve in conventional brake control.
[Explanation of symbols]
a Brake hydraulic pressure command value calculation means
b Brake hydraulic pressure measuring means
c Hydraulic feedback control calculation means
d Electromagnetic solenoid valve current control means
e Dead band width learning means
f Initialization means
1 Brake pedal
2 Hydraulic booster
3 Master cylinder
4,5 Oil passage switching valve
6 Stroke simulator
7 Wheel cylinder
8 Master cylinder pressure sensor
9 Wheel cylinder pressure sensor
10 Booster valve (Electromagnetic solenoid valve)
11 Pressure reducing valve (electromagnetic solenoid valve)
12 Brake control controller
13 Motor controller
14 Current control circuit for DC / AC conversion
15 AC synchronous motor
16 Drive wheels
17 battery
18 Pump
19 Accumulator
20 source pressure sensor
21 Pressure switch

Claims (2)

Brake hydraulic pressure command value calculating means for calculating a hydraulic pressure command value acting on the brake that suppresses the rotation of the wheel based on the operation amount of the brake operating member by the driver;
Brake hydraulic pressure measuring means for measuring an actual value of hydraulic pressure acting on a brake for suppressing rotation of the wheel;
In order to match the brake hydraulic pressure measurement value with the brake hydraulic pressure command value, both values are input, and a hydraulic feedback control calculation means for calculating a current command value to the electromagnetic solenoid valve;
Electromagnetic solenoid valve current control means for performing current control on the electromagnetic solenoid valve based on the calculated current command value;
A dead band width learning means for learning the dead band width of the electromagnetic solenoid valve based on a current command value when a predetermined amount of change occurs in the brake hydraulic pressure measurement value by gradually increasing the current passing through the electromagnetic solenoid valve; Prepared,
In the vehicle brake control device that uses the learned dead zone width as an offset correction value of the current command value for current control of the electromagnetic solenoid valve,
When the difference between the brake hydraulic pressure command values stays within a predetermined range for a predetermined time and then changes by a predetermined amount for the first time, an initialization unit is provided for initializing an internal variable of the hydraulic feedback control calculation unit,
The vehicular brake control device according to claim 1, wherein the dead band width learning means is means for learning a dead band width of the electromagnetic solenoid valve in synchronization with the initialization by the initialization means. The vehicle brake control device according to claim 1,
The dead zone learning means is a means for learning a dead zone width of an electromagnetic solenoid valve at the first learning immediately after the start of brake control, irrespective of initialization by the initialization means. Control device.

Images