JP2006295743A - Digital filter device and digital filter method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To effectively reduce restraining of steep fluctuations of a digital signal due to digital filter. <P>SOLUTION: In a digital filter device which performs a filtering processing for an input unprocessed digital signal and outputs a processed digital signal, when a monitor object signal for monitoring fluctuations of the unprocessed digital signal fluctuates by a predetermined value or more from a target value, the digital filter device will not output the processed digital signal but will output the unprocessed digital signal. After the device has output the unprocessed digital signal, while it makes the cut-off frequency in the filtering processing fluctuate, it outputs the processed digital signal. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a digital filter device and a digital filter method, and more particularly to a digital filter device and a digital filter method for outputting a processed digital signal processed by a digital filter.

  A digital filter is widely used to remove noise of a digital signal when an analog signal is converted into a digital signal by an AD converter. As a filtering method of the digital filter, a first-order lag filter method, a moving average method, and the like are known. However, the digital filter has an effect of removing noise, but has a possibility of suppressing a steep fluctuation of the digital signal. In addition, when a digital filter is used for driving control of a linear solenoid used in, for example, an electromagnetic valve of a vehicle, a steep fluctuation of a digital signal with respect to a target value such as overshoot or undershoot is caused through the digital filter. This may suppress the smooth drive of the linear solenoid, for example, affecting the control of the output of the digital filter.

For this reason, for example, Patent Document 1 proposes a digital filter method in which the number n of sampling data used by moving average is n / 2 when the input signal fluctuates sharply. Further, for example, Patent Document 2 proposes a solenoid valve control device that executes correction processing based on program data for correction with a correction amount that changes in correlation with the difference between the current flowing through the solenoid of the solenoid valve and the target current value. Has been. For example, in Patent Document 3, in a hydraulic pressure control device that controls the brake hydraulic pressure by controlling the supply current to the electromagnetic control valve, a liquid that suppresses a decrease in the control accuracy of the brake hydraulic pressure due to a temperature change. Pressure suppression devices have been proposed. For example, Patent Document 4 proposes a rear solenoid drive control device that energizes the linear solenoid with an appropriate duty during the next duty control even when the resistance value of the linear solenoid changes. For example, Patent Document 5 proposes a vehicle motion control device that suppresses noise caused by the operation of a vacuum booster during automatic pressurization control by appropriately controlling energization of a linear solenoid of a booster drive device. ing.
Japanese Patent Laid-Open No. 3-71713 Japanese Patent Laid-Open No. 11-171495 JP 2001-260868 A JP 2000-62597 A JP 2002-127880 A

  However, even with the digital filter method described above, it is not possible to effectively reduce the suppression of steep fluctuations in the digital signal such as overshoot and undershoot. For example, the digital signal processed by the digital filter can be reduced. This control may be affected when the control is performed. Further, for example, as described above, when a digital filter is used for driving control of a linear solenoid used in, for example, a solenoid valve of a vehicle, smooth driving of the linear solenoid may be hindered.

  The present invention has been made in view of these problems, and an object of the present invention is to effectively reduce suppression of steep fluctuations of a digital signal by a digital filter.

  In order to solve the above problems, a digital filter device according to an aspect of the present invention is a digital filter device that performs filtering processing on an input pre-processing digital signal and outputs the post-processing digital signal. When the monitoring target signal for monitoring fluctuates by a predetermined value or more from the target value, a pre-processing digital signal is output. According to this aspect, for example, it is possible to effectively reduce suppression of steep fluctuations in the digital signal such as overshoot and undershoot by the digital filter. Therefore, for example, it is possible to reduce the influence on the control based on the digital signal output from the digital filter.

  The digital filter device according to the present invention outputs the post-processing digital signal while changing the cutoff frequency in the filtering process after outputting the pre-processing digital signal when the monitoring target signal fluctuates by a predetermined value or more from the target value. May be. According to this aspect, by changing the cut-off frequency, if necessary, it is possible to effectively cut down the steep fluctuation of the digital signal or to cut the digital signal with an emphasis on noise reduction. The digital signal can be processed by the off-frequency.

  The filtering process is a filtering process based on a moving average method that averages the past n pre-processed digital signals from the current time, and the digital filter device according to the present invention has a monitoring target signal that fluctuates by a predetermined value or more from a target value. In some cases, the cutoff frequency may be varied by varying n. According to this aspect, the cut-off frequency can be changed by simple control.

  The digital filter device according to the present invention increases n at every predetermined time interval from the output of the processed digital signal when the monitored signal fluctuates by a predetermined value or more from the target value, and outputs the processed digital signal. After a predetermined time has elapsed, n may be constant. According to this aspect, for example, when only a short time has passed since the processed digital signal fluctuated by a predetermined value or more from the target value, a steep fluctuation of the digital signal is performed by averaging with a small n with a moving average method It is possible to perform digital signal processing with an emphasis on effective reduction of suppression. In addition, when a long time has elapsed since the digital signal after processing fluctuated by a predetermined value or more from the target value, the moving average method is averaged with the increased n to place an emphasis on reducing the noise of the digital signal. Digital signal processing can be performed.

  In the digital filter device according to the present invention, when the monitoring target signal fluctuates by a predetermined value or more from the target value, n pre-processing digital signals after the time when the monitoring target signal fluctuates by a predetermined value or more from the target value. N may be varied so that the pre-processed digital signal becomes According to this aspect, it is possible to calculate the average of only the pre-processing digital signal after the monitoring target signal fluctuates by a predetermined value or more from the target value, and it is possible to suppress the steep fluctuation of the digital signal. Can be reduced.

  Yet another embodiment of the present invention is also a digital filter device. This device performs a filtering process on an input pre-processing digital signal representing an output of a linear solenoid, and outputs the post-processing digital signal to a control unit that controls the output of the linear solenoid. When the monitoring target signal for monitoring the fluctuation fluctuates by a predetermined value or more from the target value, the post-processing digital signal is not output and the pre-processing digital signal is output. A filtering process based on a moving average method that averages the previous n digital signals before processing from the current time after outputting the digital signal before processing, where n is increased every predetermined time interval and n is kept constant after the predetermined time has elapsed. Perform the filtering process. According to this aspect, for example, even when a digital signal representing the output of the linear solenoid undergoes a steep fluctuation due to an overshoot or undershoot in the output of the linear solenoid, the digital signal represents the steep fluctuation of the digital signal. Suppression by the filter can be effectively reduced, and smooth drive control of the linear solenoid can be realized.

  According to the digital filter device of the present invention, it is possible to effectively reduce suppression of steep fluctuations in the digital signal by the digital filter.

  Hereinafter, embodiments of the present invention (hereinafter referred to as “embodiments”) will be described in detail with reference to the drawings.

  FIG. 1 is an overall configuration diagram of a hydraulic system 150 according to the present embodiment. The hydraulic system 150 mainly includes an actuator 80 and a master cylinder 14 other than the actuator 80.

  The brake pedal 12 is provided with a stroke sensor 46 that detects the depression stroke. The master cylinder 14 pumps brake oil, which is hydraulic fluid, in response to the driver's depression operation of the brake pedal 12. A dry stroke simulator 13 is provided between the brake pedal 12 and the master cylinder 14.

  One end of a brake hydraulic control conduit 16 for the right front wheel and a brake hydraulic control conduit 18 for the left front wheel are connected to the master cylinder 14, and these brake hydraulic control conduits exhibit the braking force of the right front wheel and the left front wheel, respectively. It is connected to wheel cylinders 20FR and 20FL for the right front wheel and the left front wheel. A right master shut-off valve 22FR and a left master shut-off valve 22FL (hereinafter collectively referred to as “master shut-off valve 22” as necessary) are respectively provided in the brake hydraulic control conduits 16 and 18 for the right front wheel and the left front wheel. ), And a right master pressure sensor 48FR for measuring the master cylinder hydraulic pressure on the right front wheel side and the left front wheel side, and a left master pressure sensor 48FL (hereinafter collectively referred to as “master” as necessary). A pressure sensor 48 "). When the brake pedal 12 is depressed by the driver, the depression is detected by the stroke sensor 46, but the failure of the stroke sensor 46 is assumed and the measurement of the master cylinder hydraulic pressure by the master pressure sensor 48 is also used. Depression is detected. The master cylinder hydraulic pressure is monitored by two pressure sensors from the viewpoint of fail-safe.

  The master shut-off valve 22 is an electromagnetic valve that is normally open (hereinafter referred to as “normally open type”). When the W / C (wheel cylinder) pressure is controlled by the control of the pressure-increasing linear valve 40 and the pressure-decreasing linear valve 42, which will be described later, the wheel cylinder 20 is closed and the wheel cylinder 20 is disconnected from the master cylinder 14. When these master shut-off valves 22 are opened, the master cylinder 14 and a wheel cylinder 20 described later are brought into communication, and the brake 70 is operated by the hydraulic pressure of the master cylinder 14.

  A reservoir tank 26 is connected to the master cylinder 14, a wet stroke simulator 24 is connected via an on-off valve 23, and one end of a hydraulic supply / discharge conduit 28 is connected to the reservoir tank 26. The hydraulic supply / discharge conduit 28 is provided with an oil pump 34 driven by a pump motor 32. The discharge side of the oil pump 34 is a high-pressure conduit 30, and an accumulator 49 and a relief valve 53 are provided. The accumulator 49 accumulates brake oil that has been brought to a high pressure in the range of 16 to 21.5 MPa (hereinafter referred to as “control range”) by the oil pump 34. The relief valve 53 opens when the accumulator pressure is abnormally high, for example, 30 MPa, and allows high-pressure brake oil to escape to the hydraulic supply / discharge conduit 28.

  The high pressure conduit 30 is provided with an accumulator pressure sensor 51 that measures the accumulator pressure. The electronic control unit 200 described later receives the accumulator pressure that is the output of the accumulator pressure sensor 51, and controls the pump motor 32 so that the accumulator pressure falls within the control range.

  Each of the high-pressure conduits 30 is normally in a closed state (referred to as a “normally closed type”). When necessary, the front right wheel pressure-increasing linear valve 40FR and the left front wheel pressure-increasing linear are used for pressure increase of the wheel cylinder. A right front wheel wheel cylinder 20FR is connected via a valve 40FL, a right rear wheel pressure-increasing linear valve 40RR, and a left rear wheel pressure-increasing linear valve 40RL (hereinafter collectively referred to as “pressure-increasing linear valve 40” as necessary). The left front wheel wheel cylinder 20FL, the right rear wheel wheel cylinder 20RR, and the left rear wheel wheel cylinder 20RL (hereinafter collectively referred to as “wheel cylinder 20” as necessary).

  Each of the wheels of the vehicle (not shown) includes a right front wheel brake 70FR, a left front wheel brake 70FL, a right rear wheel brake 70RR, and a left rear wheel brake 70RL (hereinafter collectively referred to as necessary). (Referred to as “brake 70”), and the brake force is exerted by pressing the brake shoe against the drum by driving the wheel cylinder 20, respectively.

  The right front wheel wheel cylinder 20FR and the left front wheel wheel cylinder 20FL are an electromagnetic flow control valve used for pressure reduction when necessary, that is, a normally closed right front wheel pressure reducing linear valve 42FR that is a linear valve, and a left front wheel pressure reducing pressure. The hydraulic supply / discharge conduit 28 is connected via a linear valve 42FL. Further, the right rear wheel wheel cylinder 20RR and the left rear wheel wheel cylinder 20RL are respectively connected to the hydraulic supply / discharge conduit 28 via the normally open right rear wheel pressure reducing linear valve 42RR and the left rear wheel pressure reducing linear valve 42RL. It is connected. Hereinafter, the pressure reducing linear valves 42FL, 42FR, 42RL, and 42RR are collectively referred to as “the pressure reducing linear valve 42” as necessary.

  In the vicinity of each wheel cylinder 20, a right front wheel W / C pressure sensor 44FR, a left front wheel W / C pressure sensor 44FL, and a right rear wheel W / C pressure sensor 44RR that measure the hydraulic pressure in the wheel cylinder, respectively. A left rear wheel W / C pressure sensor 44RL (hereinafter collectively referred to as "W / C pressure sensor 44" as necessary) is provided.

  FIG. 2 is a functional block diagram of an electronic control unit 200 (hereinafter referred to as “ECU 200”) according to the present embodiment. The ECU 200 has a main control unit 210 including an arithmetic unit using a microcomputer, a ROM 204, a RAM 206, an I / O port, and the like. The I / O port of the main control unit 210 includes the above-described stroke sensor 46, accumulator pressure, and the like. A sensor 51, a master pressure sensor 48, and a W / C pressure sensor 44 are connected. The ECU 200 also controls the linear valve 40 for pressure increase, the linear valve control unit 100 for controlling the linear valve 42 for pressure reduction, the pump motor control unit 130 for controlling the pump motor 32, and the master cutoff valve control unit 140 for controlling the master cutoff valve 22. The linear valve control unit 100 is connected to the pressure-increasing linear valve 40 and the pressure-reducing linear valve 42, the pump motor control unit 130 is connected to the pump motor 32, and the master cutoff valve control unit 140 is connected to the master cutoff valve 22. Each is connected.

  The arithmetic unit 202 includes a stroke signal indicating the depression stroke of the brake pedal 12 from the stroke sensor 46, an accumulator pressure signal indicating the accumulator pressure from the accumulator pressure sensor 51, a master pressure signal indicating the master pressure from the master pressure sensor 48, and W / C. A W / C pressure signal indicating the W / C pressure is input from the pressure sensor 44.

  The ROM 204 stores a plurality of programs such as a pressure-increasing linear valve control program and a plurality of tables such as an accumulator pressure setting range determination table. The RAM 206 is used as a work area for data storage and program execution. For example, the calculation unit 202 calculates a target W / C pressure for each wheel based on a stroke signal, a master pressure signal, a W / C pressure signal, and the like using a program stored in the ROM 204.

  The arithmetic unit 202 calculates a PSB (Pressure Servo Brake) command current, which is a control current supplied to the pressure increasing linear valve 40 and the pressure reducing linear valve 42, from the calculated target W / C pressure. The calculated PSB command current is output to the linear valve control unit 100. Upon receiving the PSB command current, the linear valve control unit 100 converts the input PSB command current into a drive signal for driving the linear solenoids of the pressure-increasing linear valve 40 and the pressure-decreasing linear valve 42, and the pressure-increasing linear valve 40. And output to the pressure-reducing linear valve 42. The control hydraulic pressure obtained by the operation of the driven pressure increasing linear valve 40 and pressure reducing linear valve 42 is fed back to the arithmetic unit 202. As described above, the arithmetic unit 202 performs control so that the W / C pressure of each wheel becomes the target W / C pressure.

  FIG. 3 is a diagram showing the configuration of the pressure-increasing linear valve 40 and the pressure-decreasing linear valve 42 according to this embodiment. The pressure increasing linear valve 40 includes a seating valve 58 and a linear solenoid 50. The seating valve 58 includes a valve seat 52, a valve element 54 that can be seated on and separated from the valve seat 52, and a spring 56 that biases the valve element 54 in a direction in which the valve element 54 is seated on the valve seat 52. ing. The linear solenoid 50 has a coil 60 that generates a solenoid force by supplying a current. The pressure-increasing linear valve 40 is a normally closed valve that keeps the valve element 54 seated on the valve seat 52 by the elastic force of the spring 56 while no current is supplied to the coil 60.

  In the pressure-increasing linear valve 40, when a current is supplied to the coil 60, a solenoid force acts in a direction to separate the valve element 54 from the valve seat 52, and the valve element 54 is separated from the valve seat 52. The valve 58 is opened. When the distance between the valve element 54 and the valve seat 52 is small and the opening degree is small, the flow rate of the hydraulic fluid flowing through the seating valve 58 becomes small, and the pressure increase gradient of the hydraulic pressure in the wheel cylinder 20 becomes small. Further, when the distance between the valve element 54 and the valve seat 52 is large and the opening degree is large, the flow rate of the hydraulic fluid flowing through the seating valve 58 becomes large, and the pressure increase gradient of the hydraulic pressure in the wheel cylinder 20 becomes large. The pressure-increasing linear valve 40 is disposed between the oil pump 34 and the wheel cylinder 20, and controls the flow rate of the working fluid flowing from the oil pump 34 to the wheel cylinder 20 by adjusting the opening of the seating valve 58. The pressure increase of the W / C pressure is controlled.

  The pressure reducing linear valve 42 has the same structure as the pressure increasing linear valve 40, and when the opening of the seating valve 58 is small, the pressure reducing gradient of the hydraulic pressure in the wheel cylinder 20 is reduced, and the opening of the seating valve 58 is reduced. Is large, the pressure reduction gradient of the hydraulic pressure of the wheel cylinder 20 becomes large. The pressure reducing linear valve 42 is disposed between the reservoir tank 26 and the wheel cylinder 20, and controls the flow rate of the working fluid flowing from the wheel cylinder 20 to the reservoir tank 26 by adjusting the opening of the seating valve 58. Controls the reduction of the W / C pressure.

  The linear valve control unit 100 of the ECU 200 is connected to the coil 60 of the pressure increasing linear valve 40 and the pressure reducing linear valve 42. The linear valve control unit 100 includes a pulse output unit 108 which will be described later, and the pulse output unit 108 controls a duty (time ratio) for supplying a current by PWM control to control a supply current to the coil 60. To do. In the pressure increasing linear valve 40 and the pressure reducing linear valve 42, the opening degree of the linear solenoid 50 is adjusted in accordance with the current supplied to the coil 60, and the W / C pressure is controlled.

  FIG. 4 is a block diagram showing the configuration of the linear valve control unit 100 according to this embodiment. The linear valve control unit 100 includes a current-duty conversion map 102, a temperature correction unit 104, a clamp 106, a pulse output unit 108, an A / D conversion unit 110, a digital filter 112, a switch 114, a PI control unit 116, and an interrupt processing unit 118. The current monitoring unit 120 is included. The linear valve control unit 100 is connected to the main control unit 210 of the ECU 200 and receives a PSB command current input from the main control unit 210.

  The PSB command current input from the main control unit 210 is input to the current-duty conversion map 102 of the linear valve control unit 100. The current-duty conversion map 102 is a map showing the duty of the current supplied to the linear solenoid 50 with respect to the input PSB command current, and is a map in which the horizontal axis is current and the vertical axis is duty. The current-duty conversion map 102 converts the PSB command current input according to this map into a duty that is a ratio of supplying a current to the linear solenoid 50. On the other hand, since this current-duty conversion map 102 is mapped in a room temperature state, it does not correspond to all temperature regions. For this reason, for example, when the linear valve or the control liquid becomes high temperature, it is optimal to convert the current supplied to the linear solenoid 50 to the duty by the current-duty conversion map 102 mapped in the normal temperature state. May not be converted to a correct duty. For this reason, the temperature correction unit 104 reduces the influence due to the temperature change by rewriting a part of the current-duty conversion map 102 in order to realize control corresponding to all temperature regions. As described above, the feed-forward control for converting the PSB current input from the main control unit 210 into the duty of an appropriate current supplied to the linear solenoid 50 is performed by the current-duty conversion map 102 and the temperature correction unit 104.

  The clamp 106 performs control so that a voltage higher than a certain target voltage is not applied even when a voltage fluctuation occurs. Normally, the control is performed between 0 volts and 12 volts, but it is possible to control up to 16 volts in consideration of voltage variation. A voltage of 16 volts or more is configured such that error processing by the ECU 200 is performed.

  The pulse output unit 108 is connected to the linear solenoid 50. The pulse output unit 108 has a switching element, and supplies current to the linear solenoid 50 by opening / closing control of the switching element. The pulse output unit 108 changes the current supplied to the linear solenoid 50 at the duty input by the feedforward control performed by the current-duty conversion map 102 and the temperature correction unit 104, thereby controlling the drive of the linear solenoid 50. Done.

  The A / D conversion unit 110 A / D converts the voltage signal related to the linear solenoid and outputs a digital signal. The digital filter 112 performs a filtering process on the digital signal output from the A / D conversion unit 110. Hereinafter, a digital signal input to the digital filter 112 will be described as a pre-processing digital signal, and a digital signal that has been subjected to filtering processing by the digital filter 112 will be described as a post-processing digital signal.

  In the present embodiment, the digital filter 112 employs a filtering method based on the moving average method. The filtering by the moving average method is a method of averaging several data and sequentially shifting the data at the time of averaging. Usually, among the pre-processed digital signals sampled at predetermined time intervals, the pre-process digital signal at the current time is the first, and the average of the pre-process digital signals from the current time to the past n-th is output. At this time, the digital filter 112 outputs a signal that does not vary within the predetermined time interval at every predetermined time interval. By performing the filtering process on the unprocessed digital signal in this way, a noise component included in the unprocessed digital signal can be removed. In the present embodiment, the resolution of the digital filter 112 is 480 microseconds.

  The digital filter 112 is connected to the switch 114 via a connector. A current monitoring unit 120 is provided between the connector between the digital filter 112 and the switch 114 and between the current-duty conversion map 102 and the temperature correction unit 104. The current monitoring unit 120 first detects a voltage from the post-processing digital signal output from the digital filter 112, and calculates the current supplied to the linear solenoid 50 from this voltage value. The detection result of the current monitoring unit 120 is input to the digital filter 112 and the switch 114.

  The switch 114 compares the monitored current value, which is the current value supplied to the linear solenoid 50 input from the current monitoring unit 120, with the target current value, and the monitored current value has fluctuated by a threshold current value or more than the target current value. It has a switching judgment part which judges whether it is. The switching determination unit compares and determines the monitored current value and the target current value at predetermined time intervals. In this embodiment, the predetermined time interval is 3 milliseconds, and the threshold current value is 50 milliamperes. It is said that.

  The switch 114 is connected to the PI control unit 116 and the interrupt processing unit 118, and the PI control unit 116 and the interrupt processing unit 118 are connected to a connector provided between the temperature correction unit 104 and the clamp 106. The switch 114 inputs the post-processing digital signal output from the digital filter 112 to the PI control unit 116 when it is determined that the monitoring current value does not fluctuate by more than the threshold current value or more than the target current value. Thus, the digital filter 112 and the PI control unit 116 are connected. In addition, the switch 114 switches from the digital filter 112 when the monitored current value fluctuates more than the threshold current value than the target current value due to, for example, a steep fluctuation of the monitored current value when overshoot or undershoot occurs. The digital filter 112 and the processing unit 118 are connected so that the output processed digital signal is input to the interrupt processing unit 118.

  When the monitored current value does not fluctuate by more than the threshold current value than the target current value, the PI control unit 116 performs PI control, that is, proportional control (ie, proportional control) so that the processed digital signal input from the digital filter 112 approaches the target value. P control) and integral control (I control) are performed. The PI control unit 116 outputs the PI control result as a duty correction value. The PI control unit 116 performs this PI control every 3 milliseconds. The pulse output unit 108 supplies current to the linear solenoid 50 with the corrected duty. As described above, the linear valve control unit controls the driving of the linear solenoid by feedback control that detects the voltage of the linear solenoid 50 and controls the PI control unit to approach the target value. Therefore, when the monitored current value does not fluctuate by more than the threshold current value from the target current value, the linear valve control unit 100 performs the feedforward control by the current-duty conversion map 102 and the temperature correction unit 104, and the PI control unit 116. The drive of the linear solenoid 50 is controlled by both types of feedback control such as the above.

  When the monitored current value fluctuates by more than the threshold current value than the target current value, it is difficult for the feedback control by the PI control unit 116 to bring the current supplied to the linear solenoid 50 close to the target value. For this reason, in this case, the interrupt processing unit 118 performs a process of suppressing the fluctuation of the monitoring current value that fluctuates sharply. In the present embodiment, the interrupt processing by the interrupt processing unit is performed in 480 microseconds which is shorter than 3 milliseconds which is the control time interval of the PI control unit. The interrupt processing by the interrupt processing unit does not perform feedback control based on the detected voltage value of the linear solenoid 50 as in the PI control unit 116. Therefore, when the monitored current value fluctuates more than the threshold current value than the target current value, the linear valve control unit 100 performs feedforward control by the current-duty conversion map 102 and the temperature correction unit 104 and interrupt processing by the interrupt processing unit 118. To control the driving of the linear solenoid.

Here, the digital filter 112 has an effect of removing noise, but may suppress a steep fluctuation of the digital signal. Therefore, when the voltage of the linear solenoid 50 fluctuates sharply due to overshoot or undershoot, the digital filter 112 may perform a filtering process to suppress fluctuations in the processed digital signal. As a result, a possibility that in FIG. 6, so that the difference in the digital signal I _b after treatment relative to pretreatment digital signal I _r occurs, the deviation between the pre-processing the digital signal and the processed digital signal occurs There is. Such a divergence between the pre-processing digital signal and the post-processing digital signal hinders smooth driving of the linear solenoid 50, leads to fluid pressure pulsation, and may be a cause of noise.

  Therefore, when the digital filter 112 receives the detection result input from the current monitoring unit 120 and the monitored current value fluctuates more than the target current value by the threshold current value, the digital filter 112 determines that the monitored current value is the target current value. The digital signal before the filtering process by the digital filter 112 is not performed for a predetermined time from when the current fluctuates more than the threshold current value, and the pre-processed digital signal is output. As a result, the digital signal can be output without suppressing a steep fluctuation of the pre-processing digital signal. Thereby, for example, the control accuracy in the control unit or processing unit downstream from the digital filter 112 such as the interrupt processing unit 118 can be increased, and the linear solenoid 50 can be controlled to be closer to the target driving. .

  After a predetermined time has elapsed since the monitoring current value fluctuated by more than the threshold current value from the target current value, the digital filter 112 is a digital signal before processing at the current time among digital signals before processing sampled at predetermined time intervals. Filtering by the moving average method is performed again, where the first signal is output and the average of the pre-processing digital signals from the current time to the past nth signal is output. In this case, the digital filter 112 cuts off the digital filter 112 by changing this n at predetermined time intervals after a predetermined time has elapsed since the monitored current value fluctuated by a threshold current value or more than the target current value. Vary the frequency.

  Specifically, the digital filter 112 increases n every predetermined time interval from the time when the processed digital signal is output, and makes n constant after the predetermined time has elapsed since the output of the processed digital signal. As a result, when only a short time has passed since the processed digital signal fluctuated by more than the threshold value from the target value, it is effective to suppress steep fluctuations in the digital signal by performing averaging by the moving average method with a small n. It is possible to process digital signals with an emphasis on reduction. In addition, when a long time has elapsed from the time when the processed digital signal fluctuates by more than the threshold value from the target value, the digital signal is focused on reducing the noise of the digital signal by averaging with the increased n by the moving average method It is possible to perform the process.

  In addition, the digital filter 112 changes n so that the n pre-processing digital signals become the pre-processing digital signals after the monitoring target signal changes by a predetermined value or more from the target value. As a result, the digital filter 112 can calculate the average of only the pre-processing digital signal after the monitoring target signal fluctuates by a predetermined value or more from the target value, and steep fluctuation of the digital signal is suppressed. Can be effectively reduced.

  FIG. 5 is a flowchart showing processing of the digital filter 112 according to the present embodiment. The process in this flowchart is started and repeated every predetermined time. In the present embodiment, the predetermined time for repeatedly starting the processing in this flowchart is 3 milliseconds.

Digital filter 112, monitoring current input from the current monitoring unit 120 determines whether the variation from target current threshold current I _th or more (S11). In the present embodiment, I _th is set to 50 milliamps. When it is determined that the monitored current has fluctuated more than the threshold current from the target current (Y in S11), the digital filter 112 clears the counter j and sets j to zero (S12). When it is determined that the monitoring current does not fluctuate more than the threshold current from the target current (N in S11), the digital filter 112 increments the counter j (S13).

Next, the digital filter 112 determines whether j is 2 or less (S14). When it is determined that j is 2 or less (Y in S14), the digital filter 112 outputs the actual current I _rj calculated from the pre-processing digital signal when the counter is j as the output current I _a . (S15). That is, if the counter is zero, i.e., when the monitoring current is varied from the target current threshold current I _th or more, outputs the actual current I _r0 at the present time as the output current I _a. Further, when the counter j is 1, that is, when 3 milliseconds have elapsed since the monitoring current last changed more than the threshold current I _th from the target current, the actual current I _r1 at the current time when the counter j is set to 1 is obtained. and outputs as the output current _{I a.} Likewise if the counter is 2, i.e., when the monitoring current is passed 6 milliseconds from the time of change at the end than the target current threshold current I _th or more, the actual current I _r2 of current time counter j is 2 and outputs as the output current _{I a.} When only a short time has passed since the current fluctuated sharply, the influence of the ripple is small and the need for filtering processing is small. Also, initially the monitoring current varies from the target current threshold current I _th or more, mainly for controlling the current supplied to the linear solenoid 50 by the feed forward control, the need to monitor accurately the current by the current monitoring unit 120 There are few. For this reason, when only a short time has elapsed since the current changed sharply, the digital filter 112 outputs the actual current I _rn at the current time as the output current I _a .

If it is determined that j is greater than 2 (N in S14), the digital filter 112 determines whether j is 3 (S16). If j is determined to be 3, that is, when the monitoring current is passed 9 milliseconds from the time of change at the end than the target current threshold current I _th or more (S16 of Y), the digital filter 112, the counter 3 and is the actual current I _r3 and the current time was a past one preprocessing the digital signal from the current time, the average output current of the counter and the actual current I _r2 of 3 ms before the current time, which is that the 2 Output as _Ia (S17). At this time, the actual current _Ir3 is also in a transient state, and it is preferable not to perform the filtering process as much as possible in order to improve the responsiveness. However, since the actual current tends to be stable, the reduction of ripple is taken into consideration, and the digital filter 112 outputs a post-processing digital signal obtained by performing a filtering process using a moving average method with n = 2.

If it is determined that j is not 3 (N in S16), the digital filter 112 determines whether j is 4 (S18). If j is determined to be 4, that is, when the monitoring current has passed 12 ms from the time of change at the end than the target current threshold current I _th or more (S18 of Y), the digital filter 112, the counter is a 4 The current current I _{r4 at} the current time, the two pre-processing digital signals from the current time, the actual current I _r3 3 milliseconds before the current time at which the counter was set to 3, and the counter 2 been outputs an average of 6 milliseconds before the actual current _{I r2} as the output current _{I a} from the current time (S19). Since the actual current tends to be more stable than when j is 3, the digital filter 112 outputs a processed digital signal that has been subjected to filtering processing by increasing n.

If j is 4 judged not, that is, when the monitoring current has passed more than 15 ms when the variation in the end than the target current threshold current I _th or more (S18 of N), the digital filter 112, the counter and the j The current current I _{rj at the} current time and the four pre-processing digital signals in the past from the current time, the actual current I _{rj-1 at} 3 milliseconds before the current time at which the counter is j−1, and the counter There outputs an average of j-4 and the actual current of 12 milli-seconds before the current time that is _{I rj-4} as an output current _{I a} (S20). In this case, since it is considered that the actual current is stable, the digital filter 112 outputs the processed digital signal obtained by performing the filtering process by the moving average method in which n is 5 similarly to n in the normal filtering process. Output.

Figure 6 is a diagram showing an output current I _a by a digital filter 112 of the present embodiment. I _r shows the actual current supplied to the linear solenoid 50 which is calculated from the pre-processed digital signal. I _a indicates an output current from the digital filter 112 according to the present embodiment. Also, I _b shows the output current by a digital filter for performing a filtering process by conventional moving average method.

As shown in this figure, if the actual current I _r varies sharply, the output current I _b with a digital filter for performing a filtering process by conventional moving average method, will deviate significantly from the actual current I _r. In contrast, the output current I _a by a digital filter 112 according to this embodiment, since the counter j is output as the actual current I _r is 2 or less, and outputs without actual current I _r to suppress an abrupt change be able to. Further, when the counter j is 3 or more, filtering processing by the moving average method is performed. In this case, the digital filter 112 performs the filtering process while increasing n averaged by the moving average method every 3 milliseconds when the counter j is 3 to 5. Thus, while suppressing the deviation of the actual current I _r and the output current I _a, the noise contained in the actual current I _r can be suppressed. When the counter j is 5 or more, the digital filter 112 shifts to a normal filtering process with n being constant.

  FIG. 7 is a diagram illustrating the basis of n of the moving average method in the digital filter 112 according to the present embodiment. In this embodiment, the pulse output unit 108 controls the linear solenoid 50 by PWM control of 5 kilohertz. For this reason, ripples at intervals of 200 microseconds are generated in the current supplied to the linear solenoid 50 as shown in FIG. In order to measure the current of the linear solenoid 50 with high accuracy, it is preferable to calculate the median value of the ripple. For this reason, a digital filter is required.

On the other hand, the resolution of the digital filter 112 is 480 microseconds. Therefore, if you start sampling at T _1, after the two cycles of the ripple is sampled at T _2. By repeating the sampling in this way, a sampling current from T ₁ to T ₅ is obtained every 40 microseconds for one ripple as shown in the lower part of FIG.

  In the present embodiment, when the counter j is 3, n is set to 2, and when the counter j is 4, n is set to 3. This is because it is considered that a value close to the median value of the ripple can be calculated by averaging these n sampling currents. However, when the counter j is 5 or more, n is set to 5, and n is set to 4, and the sampling current is not averaged. This is because when n is set to 4 and four sampling currents are averaged, sampling is performed at three points in 0 to 100 microseconds and only one point is sampled in 100 to 200 microseconds as shown in this figure. Will be. For this reason, when it is considered that the linear solenoid 50 is driven in the entire range from 0% to 100%, the detection result by the current monitoring unit 120 may vary. Therefore, in this embodiment, n is set to 4 and sampling current averaging is not performed.

  The present invention is not limited to the above-described embodiments, and an appropriate combination of the elements of each embodiment is also effective as an embodiment of the present invention. Various modifications such as design changes can be added to each embodiment based on the knowledge of those skilled in the art, and embodiments to which such modifications are added can also be included in the scope of the present invention. Here are some examples.

  The digital filter 112 does not have to change n every predetermined time interval. For example, when the counter j is 3 and 4, n is 2, and when the counter j is 5 to 7, n is 3. N may be varied at different time intervals. Thereby, various filtering processes are possible.

  The digital filter 112 outputs a pre-processing digital signal when the monitored current fluctuates by more than a threshold current from the target current, and then outputs a post-processing digital signal that has been filtered by the moving average method without changing n. May be. Thereby, control can be simplified.

1 is an overall configuration diagram of a hydraulic system according to the present embodiment. It is a functional block diagram of ECU which concerns on this embodiment. It is a figure which shows the structure of the linear valve for pressure increase which concerns on this embodiment, and the linear valve for pressure reduction. It is a block diagram which shows the structure of the linear valve control part which concerns on this embodiment. It is a flowchart which shows the process of the digital filter which concerns on this embodiment. It is a figure which shows the output current by the digital filter which concerns on this embodiment. It is a figure which shows the basis of n of the moving average method in the digital filter which concerns on this embodiment.

Explanation of symbols

  40 linear valve for pressure increase, 42 linear valve for pressure reduction, 50 linear solenoid, 100 linear valve control unit, 102 current-duty conversion map, 104 temperature correction unit, 108 pulse output unit, 110 A / D conversion unit, 112 digital filter, 120 current monitoring unit, 200 ECU, 210 main control unit.

Claims (8)

In a digital filter device that performs a filtering process on an input pre-processing digital signal and outputs a post-processing digital signal,
When the monitoring target signal for monitoring the fluctuation of the pre-processing digital signal fluctuates by a predetermined value or more from a target value, the pre-processing digital signal is not output but the pre-processing digital signal is output. Digital filter device.   When the monitoring target signal fluctuates by a predetermined value or more from a target value, after outputting the pre-processing digital signal, the post-processing digital signal is output while changing a cutoff frequency in the filtering process. Item 2. The digital filter device according to Item 1. The filtering process is a filtering process by a moving average method that averages the past n digital signals before processing from the current time,
When the monitoring target signal fluctuates by a predetermined value or more from a target value, when the filtering process is performed after outputting the pre-processing digital signal, the cutoff frequency is fluctuated by fluctuating the n. The digital filter device according to claim 2, wherein   When the monitoring target signal fluctuates by a predetermined value or more from a target value, the n is increased every predetermined time interval from the output of the processed digital signal, and after a predetermined time has elapsed from the output of the processed digital signal 4. The digital filter device according to claim 3, wherein n is constant.   When the monitoring target signal fluctuates by a predetermined value or more from the target value, the n pre-processing digital signals become pre-processing digital signals after the monitoring target signal fluctuates by a predetermined value or more from the target value. The digital filter device according to claim 3, wherein the n is varied as described above. In the digital filter device that performs filtering on the pre-processed digital signal that represents the output of the input linear solenoid, and outputs the processed digital signal to the control unit that controls the output of the linear solenoid,
When the monitoring target signal for monitoring the fluctuation of the output of the linear solenoid fluctuates by a predetermined value or more from the target value, the digital signal before processing is not output, and the digital signal before processing is output,
After the output of the pre-processed digital signal, a filtering process using a moving average method that averages the previous n digital signals before the process from the current time, wherein the n is increased every predetermined time interval, and after a predetermined time has elapsed A digital filter device that performs a filtering process to make n constant. In a digital filter method for performing a filtering process on an input pre-processing digital signal and outputting a post-processing digital signal,
Determining whether the monitored signal has fluctuated from a target value by a predetermined value or more;
And a step of outputting the pre-processing digital signal when it is determined that the monitoring target signal has fluctuated by a predetermined value or more from a target value. In a digital filter method for performing a filtering process on an input pre-processing digital signal and outputting a post-processing digital signal,
Determining whether the monitored signal has fluctuated from a target value by a predetermined value or more;
Outputting the pre-processing digital signal when it is determined that the monitoring target signal has changed from a target value by a predetermined value or more;
And outputting the post-processing digital signal while varying the cutoff frequency after outputting the pre-processing digital signal.

Images