JP2015115044A - Feedback control device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To reduce, when proportional operation quantities are to be obtained from deviations, any deviation remaining on account of the presence of a dead zone.SOLUTION: An operation quantity control unit 16 determines for each control period whether or not a deviation is in a dead zone, where a proportional operation quantity calculated by a proportional controller 15 is 0 in spite of the fact that a deviation calculated by a subtractor 14 is not 0. If the deviation is not in the dead zone, the proportional operation quantity is outputted via a change-over switch 30, and the accumulated value is reset to 0. If the deviation is in the dead zone, a unit value is added to or subtracted from the accumulated value according to the polarity of the deviation. When the absolute accumulated value reaches a threshold, the accumulated operation quantity is set to a positive or negative operation value according to the polarity of the accumulated value and outputted via the change-over switch 30 to reset the accumulated value to 0.

Description

  The present invention relates to a digital feedback control device.

  Proportional control, which is one type of feedback control, obtains a deviation by subtracting a detected value from a target value of a controlled amount, and obtains an operation amount to be given to a controlled object by multiplying the deviation by a proportional gain. In order to make the detected value follow the target value so as not to cause a steady deviation while suppressing overshoot, a combination of proportional control (P control) and integral control (I control), so-called proportional integral control (PI control) is used. (Patent Document 1).

  In addition to the problem of the steady deviation, when the control calculation is executed by digital calculation, there is a problem that a value less than the 1LSB equivalent value of the quantized operation amount is rounded down and a quantization error occurs. As a result, there is a dead zone in which the manipulated value becomes zero even though the digitally calculated deviation value is not zero, and the deviation continues to remain.

JP 2006-171480 A

  In order to reduce the residual deviation, it is only necessary to increase the proportional gain so that the operation amount does not easily become zero. However, if the gain is increased, the control system tends to become unstable. Further, in the proportional integral control, the integral controller integrates the deviation and reflects it in the manipulated variable, so that it can escape from the dead zone. However, the integration controller that executes the integration calculation for such a small deviation has a problem that the number of constituent bits increases, the circuit scale increases, or the calculation time increases.

  The present invention has been made in view of the above circumstances, and an object of the present invention is to provide a feedback control device that can reduce a deviation remaining due to the presence of a dead zone of digital calculation for obtaining an operation amount from the deviation.

  The feedback control device according to claim 1 includes a detecting unit that detects a controlled amount to be controlled, and a deviation calculating unit that digitally calculates a deviation by subtracting a detected value of the controlled amount from a target value of the controlled amount. And a proportional control means for digitally calculating the proportional manipulated variable by multiplying the deviation by a proportional gain, and an manipulated variable control means. The roll-up value described below is a deviation for each predetermined control cycle when the proportional manipulated variable is zero, that is, when the deviation is in the dead zone of the digital calculation related to the proportional manipulated variable. It is a value accumulated considering the polarity of.

  The operation amount control means repeatedly executes the following control at a predetermined control cycle. That is, when the deviation is zero and the proportional manipulated variable is non-zero, the product value is reset and the proportional manipulated variable is output to the control target. This control is a normal proportional control. On the other hand, when the deviation is non-zero and the proportional manipulated variable is zero (in the dead zone), the first unit value is added to the product value under the condition that the deviation is positive, and the deviation is The second unit value is subtracted from the accumulated value under a negative condition.

  When the accumulated value after the addition or subtraction reaches a positive first threshold value, the manipulated variable control means sets the accumulated manipulated variable to the first manipulated value and outputs it to the controlled object. When the accumulated value after the addition or subtraction reaches a negative second threshold value, the accumulated operation amount is set to the second operation value and output to the control target. When these roll-up operation amounts are output, the roll-up value is reset. The manipulated variable control means holds the accumulated value and sets the proportional manipulated variable or zero when the accumulated value after addition or subtraction has not reached either the first threshold value or the second threshold value. The set roll operation amount is output to the control target.

  That is, when the deviation is in the dead zone, the operation amount control means adds the first unit value to the product value or subtracts the second unit value according to the sign of the deviation for each control period. When the accumulated value after addition / subtraction reaches the first threshold value or the second threshold value, the accumulated operation amount is set to the first operation value or the second operation value and output to the control target. Reset the rollup value.

  In this accumulation process, an operation amount less than a value equivalent to 1LSB obtained by multiplying a deviation in the dead zone by a proportional gain is changed to a specific value, and is determined according to the sign of the deviation. The unit value is associated with the value, and the unit value is added to or subtracted from the product value for each control period. Thereby, even when the deviation is in the dead zone, the deviation can be reflected in the manipulated variable, and the residual deviation can be reduced. According to this means, compared to the case of using proportional integral control, the configuration for holding the product value and the configuration for adding / subtracting the unit value to the product value can be greatly reduced.

  According to the means described in claim 2, the absolute values of the first operation value and the second operation value are the minimum values that can change the state of the controlled object. Thereby, monotonicity of the operation amount at the time of switching between the proportional operation amount and the accumulated operation amount is ensured.

  According to the means described in claim 3, the first unit value and the second unit value are equal, and the absolute values of the first threshold value and the second threshold value are equal. Thereby, the accumulation process of a deviation state becomes a positive / negative symmetry.

  According to the means described in claim 4, the feedback control device is a type 1 or type 2 control system for the target value. In the type 1 or type 2 control system, the steady deviation of the step response becomes zero. When this means is applied to a type 1 or type 2 control system, the detected value can be made to coincide with the target value with high accuracy in the step response.

The block diagram of the electronic controller which shows 1st Embodiment Flow chart of feedback control device Diagram showing changes in digital values of target current and detected current The block block diagram of the feedback control apparatus which shows 2nd Embodiment.

(First embodiment)
A first embodiment will be described with reference to FIGS. 1 to 3. The electronic control device 10 digitally controls the current flowing through the solenoid 11 to be controlled, and includes a microcomputer 12 and a feedback control device 13 (hereinafter referred to as a control device 13). The microcomputer 12 outputs a target current that flows through the solenoid 11 to the control device 13. The control device 13 has a software configuration that operates based on a program, but can also be configured in hardware.

  The control device 13 includes a subtractor 14, a proportional controller 15, an operation amount controller 16, a PWM controller 17, a main circuit 18, a current sensor 19, an analog buffer 20, and an A / D converter 21. The control device 13 is composed of a digital circuit except for some circuits of the main circuit 18, the analog buffer 20, and the A / D converter 21, and the digital circuit operates in synchronization with a common clock signal.

  The main circuit 18 includes N-channel transistors 25 and 26 connected in series between a power supply terminal 22 connected to the positive terminal of the battery and a ground 23 with an output terminal 24 interposed therebetween. Due to the low-side drive method, the solenoid 11 is connected between the power supply terminal 22 and the output terminal 24. The transistors 25 and 26 are complementarily turned on and off by drive signals G1 and G2 output from the PWM control unit 17.

  According to the configuration of the main circuit 18, when the transistor 26 is turned on and the transistor 25 is turned off, the current flowing through the solenoid 11 increases. When the transistor 26 is turned off and the transistor 25 is turned on, the current flowing through the solenoid 11 flows back through the transistor 25 with low loss (synchronous rectification). As a result, a current corresponding to the duty ratio of the drive signal G2 flows through the solenoid 11.

  The current sensor 19 is an element that detects a current that is a controlled amount interposed in the energization path of the solenoid 11 and is configured by using, for example, a Hall element. The analog buffer 20 converts the output signal of the current sensor 19 into a detection signal having a voltage range suitable for A / D conversion. The A / D converter 21 includes a sample and hold circuit, and obtains sampling data as a detection current by sampling the detection signal at a predetermined sampling frequency and performing A / D conversion. Although not shown, the feedback control system is provided with a digital filter for removing noise from the sampling data. The current sensor 19, the analog buffer 20, and the A / D converter 21 constitute a control amount detection means 27.

  The subtractor 14 is a deviation calculating means for digitally calculating a current deviation (hereinafter referred to as deviation) by subtracting the detected current from the target current given from the microcomputer 12. The proportional controller 15 is proportional control means for digitally calculating the proportional operation amount by multiplying the deviation by a proportional gain. The operation amount control unit 16 includes a product value calculation unit 28, a product operation amount setting unit 29, and a changeover switch 30, and outputs an operation amount based on the deviation and the proportional operation amount. The accumulated value calculation unit 28 holds the accumulated value. The PWM control unit 17 generates a drive signal G2 having a duty ratio corresponding to the operation amount and a drive signal G1 that is an inverted signal thereof, and PWM drives the transistors 25 and 26 using these drive signals G1 and G2, respectively.

  Next, processing executed by the control device 13 will be described with reference to FIG. When the power is supplied, the control device 13 resets the accumulated value held in the accumulated value calculation unit 28 to 0 (step S1), and initializes the values and variables of various registers such as the target current input register. To do. Thereafter, the control device 13 repeatedly executes the processing from steps S2 to S16 at a predetermined control cycle.

  The control device 13 samples the detection signal by the A / D converter 21 and performs A / D conversion to obtain a detection current (steps S2 and S3), and subtracts the detection current from the target current by the subtractor 14 to obtain a deviation ( Step S4). Thereafter, the proportional operation amount is obtained by multiplying the deviation by the proportional gain by the proportional controller 15 (step S5). The deviation, proportional gain, and proportional manipulated variable are all digital values, and amounts less than 1LSB equivalent are discarded. For this reason, when the proportional gain is small, there is a dead zone in which the proportional manipulated variable becomes zero even though the deviation is not zero.

  The operation amount control unit 16 determines whether the deviation is in the dead zone of the proportional operation calculation for each control period, and outputs the proportional operation amount via the changeover switch 30 when the deviation is not in the dead zone. On the other hand, when the deviation is in the dead zone, the unit value 1 is added to or subtracted from the accumulated value according to the polarity of the deviation. When the absolute value of the product value reaches the threshold value 4, the product operation amount is set to the operation value 1 or −1 according to the polarity of the product value and output via the changeover switch 30. At this time, the accumulated value is reset to zero. Among these processes, the roll-up operation amount setting unit 29 executes a process for setting an operation value to the roll-up operation amount, and the roll-up value calculation unit 28 executes other processes. The unit value, threshold value, and operation value are examples, and can be changed to other values. The operation will be specifically described below.

  The operation amount control unit 16 determines whether or not the proportional operation amount calculated by the proportional controller 15 is 0 (step S6). If the proportional manipulated variable is not 0, there is no deviation in the dead zone, so the proportional manipulated variable is selected and output by the changeover switch 30, and the product value is reset to 0 (step S7). If the proportional operation amount is 0, it is further determined whether or not the deviation is 0 (step S8). When the deviation is 0, the proportional manipulated variable is naturally 0, so that the processing similar to step S7 is executed instead of the dead zone. The process of step S7 is normal proportional control.

  The operation amount control unit 16 determines that the deviation is in the dead zone when the deviation is not 0 and the proportional operation amount is 0 (step S8: NO). The manipulated variable control unit 16 determines the sign of the deviation (step S9). If the deviation is positive, the unit value 1 is added to the accumulated value (step S10). If the deviation is negative, the unit is calculated from the accumulated value. The value 1 is subtracted (step S11).

  Subsequently, the operation amount control unit 16 determines whether or not the absolute value of the product value is smaller than the threshold value 4 (step S12). When the value is smaller than the threshold value 4, the roll-up operation amount is set to 0, and the roll-up operation amount is selected and output by the changeover switch 30. At this time, the accumulated value is held (step S13). On the other hand, when the absolute value of the product value is equal to or greater than the threshold value 4, the sign of the product value is determined (step S14).

  If the product value is positive, the product operation amount is set to the operation value 1, the product operation amount is selected and output by the changeover switch 30, and the product value is reset to 0 (step S15). If the product value is negative, the product operation amount is set to the operation value -1, the product operation amount is selected and output by the changeover switch 30, and the product value is reset to 0 (step S16). The operation values 1 and −1 are minimum values that can change the current flowing through the solenoid 11 via the PWM control unit 17. The operation amount control unit 16 returns to the process of step S2 again after the processes of steps S7, S13, S15, and S16.

  According to the processing of the operation amount control unit 16 described above, even when the proportional operation amount becomes 0 because the deviation is in the dead zone, the unit value is added to or subtracted from the accumulated value for each control period according to the sign of the deviation. In addition, the deviation state can be accumulated to produce a lifting operation amount. Thus, even when the deviation is in the dead zone, the deviation can be reflected in the operation amount, and the residual deviation can be reduced.

  When the control device 13 is a type 1 or type 2 feedback control system with respect to the target current, that is, when the one-round transfer function has one or more poles at the origin, a steady deviation remains in the step response to the target current. Absent. For this reason, as the detected current approaches the target current, the deviation becomes smaller and it becomes easier to enter the dead zone. According to the control device 13, even in such a control system, the detected current can be made to coincide with the target current with high accuracy.

  FIG. 3 is a simulation waveform showing the digital value of the detected current when the target current is increased at a constant rate. For comparison, a digital value of the detected current when using a control device having a conventional configuration that does not include the operation amount control unit 16 is also shown. When the control device 13 of this embodiment is used, the current flowing through the solenoid 11 can be controlled within an error range of ± 1 LSB with respect to the target current. On the other hand, when the control device having the conventional configuration is used, the residual deviation becomes large due to the influence of the dead zone related to the proportional control calculation.

  As described above, if proportional integral control is used instead of proportional control, the integral controller integrates the deviation and reflects it in the manipulated variable, so that it is possible to escape from the dead zone. However, the circuit scale of an integration controller that can execute an integration operation even for a minute deviation of about 1 LSB increases. The manipulated variable control unit 16 of this embodiment can be configured with a smaller circuit scale than such a proportional-integral controller.

  In the present embodiment, the first unit value to be added when the deviation is positive and the second unit value to be subtracted when the deviation is negative are set equal to each other, and the positive first threshold value to be compared with the product value is negative. The absolute value of the second threshold was set equal. Furthermore, a first operation value that is set as the accumulated operation amount when the accumulated value reaches the first threshold value, and a second operation value that is set as the accumulated operation amount when the accumulated value reaches the second threshold value. The absolute value of the operation value was set equal. As a result, the stacking process of the deviation state becomes symmetric with respect to the positive and negative, and the change in the amount of the accumulated operation due to the deviation in the dead zone occurs with the positive and negative symmetry.

(Second Embodiment)
A second embodiment will be described with reference to FIG. The feedback control device 31 (hereinafter referred to as the control device 31) performs proportional control for each control cycle so that the controlled amount of the control target 32 coincides with the target value, and is configured by, for example, an ASIC. The control device 31 includes a target value input unit 33 for inputting a target value, a controlled amount detection unit 34 for detecting a controlled amount, a deviation calculating unit 35 for calculating a deviation obtained by subtracting the detected value from the target value, and a gain proportional to the deviation The proportional control unit 36 that calculates the proportional operation amount by multiplying by, the operation amount control unit 37 that inputs the deviation and the proportional operation amount and outputs the final operation amount, the initialization unit 38, and the like.

  Since the control device 31 also performs calculations with digital values, a dead zone in which the proportional manipulated variable is 0 occurs even though the deviation is not 0, as in the first embodiment. The operation amount control unit 37 determines whether or not the deviation is in the dead zone of the proportional operation calculation for each control cycle, and outputs the proportional operation amount as the operation amount when the deviation is not in the dead zone. When the deviation is in the dead zone, the first unit value or the second unit value is added to or subtracted from the product value according to the polarity of the deviation. When the absolute value of the product value reaches the first threshold value or the second threshold value, the product operation amount is set to the first operation value or the second operation value according to the polarity of the product value. The upper operation amount is output as the operation amount. A specific operation will be described below.

  When power is supplied to the control device 31, the initialization unit 38 resets the register of the target value input unit 33 to 0, initializes an operation amount switching unit 42 described later to a proportional operation amount switching state, and will be described later. The accumulated value of the accumulated value holding unit 45 is reset to zero. The deviation presence / absence determination unit 39 determines whether or not there is a deviation, that is, whether or not the deviation is zero. The proportional operation amount presence / absence determination unit 40 determines whether or not the proportional operation amount is present, that is, whether or not the proportional operation amount is zero.

  The operation amount switching determination unit 41 outputs a switching signal to the proportional operation amount when the proportional operation amount is not 0 and when the deviation is 0, that is, when the deviation is not in the dead zone. When the deviation presence / absence determination unit 39 determines that the deviation is 0, it outputs an initialization signal to a product value holding unit 45 described later. On the other hand, the operation amount switching determination unit 41 outputs a switching signal to the accumulated operation amount when the proportional operation amount is 0 and the deviation is not 0, that is, when the deviation is in the dead zone. The operation amount switching unit 42 selects a proportional operation amount or a product operation amount according to the switching signal, and outputs the operation amount to the control target 32.

  The deviation sign determination unit 43 determines whether the deviation is positive or negative. When the deviation is in the dead zone, the accumulated value addition / subtraction unit 44 adds the first unit value to the accumulated value of the accumulated value holding unit 45 when the deviation is positive, and the product of the accumulated value holding unit 45 when the deviation is negative. The second unit value is subtracted from the upper price. The first unit value and the second unit value are both positive values. The accumulated value of the accumulated value holding unit 45 is reset to 0 when an initialization signal is input.

  The roll-up value determination unit 46 determines whether the roll-up value has reached a positive first threshold value or a negative second threshold value. The product value sign determination unit 47 determines whether the product value is positive or negative. The roll-up operation amount setting unit 48 sets the roll-up operation amount to 0 when the roll-up value has not reached any threshold value. The roll-up operation amount setting unit 48 sets the roll-up operation amount to a positive first operation value when the roll-up value reaches the first threshold value, and the roll-up operation amount when the roll-up value reaches the second threshold value. Is set to a negative second operating value. The set up operation amount is given to the operation amount switching unit 42. When the accumulated operation amount is set to the first operation value or the second operation value, the accumulated operation amount setting unit 48 outputs an initialization signal to the accumulated value holding unit 45.

  According to the present embodiment, similarly to the first embodiment, even when the deviation is in the dead zone in the proportional control, an operation amount can be created based on the deviation, and the residual deviation can be reduced. In addition, the circuit scale can be reduced as compared with the case where proportional integral control is used. The first unit value and the second unit value are set to different values, the absolute values of the first threshold value and the second threshold value are set to different values, or the absolute values of the first operation value and the second operation value. If the values are set to different values, the accumulation process of the deviation state can be configured asymmetrically positive and negative, so that it can be applied to various nonlinear control objects.

(Other embodiments)
As mentioned above, although preferred embodiment of this invention was described, this invention is not limited to embodiment mentioned above, A various deformation | transformation and expansion | extension can be performed within the range which does not deviate from the summary of invention.

  Also in the first embodiment, the first unit value and the second unit value are set to different values, and the absolute values of the first threshold value and the second threshold value are set to different values, or the first operation is performed. The absolute value of the value and the second operation value may be set to different values.

  In the first and second embodiments, the absolute value of the operation value (first operation value, second operation value) is preferably a minimum value (for example, 1) that can change the state of the control target. Thereby, the monotonous increase or monotonic decrease of the operation amount at the time of switching between the proportional operation amount and the accumulated operation amount is ensured.

  In the first and second embodiments, when the accumulated value after addition or subtraction does not reach the threshold value, the accumulated value is held and the proportional operation amount is selected and output to the control target. Good.

  In the drawing, 11 is a solenoid (control target), 14 is a subtractor (deviation calculation means), 15 is a proportional controller (proportional control means), 16 and 37 are operation amount control units (operation amount control means), and 27 is detection. Means, 32 is a control target, 34 is a controlled amount detection unit (detection unit), 35 is a deviation calculation unit (deviation calculation unit), and 36 is a proportional control unit (proportional control unit).

Claims (4)

Detecting means (27, 34) for detecting a controlled amount of the controlled object (11, 32);
Deviation calculating means (14, 35) for digitally calculating a deviation by subtracting the detected value of the controlled quantity from the target value of the controlled quantity;
Proportional control means (15, 36) for digitally calculating a proportional manipulated variable by multiplying the deviation by a proportional gain;
When the deviation is zero and the proportional manipulated variable is non-zero, the product value is reset and the proportional manipulated variable is output to the control object, and the deviation is non-zero and the proportional manipulated variable Is zero, the first unit value is added to the product value under the condition that the deviation is positive, and the second unit value is subtracted from the product value under the condition that the deviation is negative. When the accumulated value after the addition or subtraction reaches a positive first threshold value, the accumulation operation amount is set to the first operation value and output to the control object, and after the addition or subtraction When the accumulated value of the output reaches the negative second threshold value, the accumulated operation amount is set to the second operation value and output to the control object, and the accumulated operation amount is output as described above. The accumulated value after resetting the accumulated value and adding or subtracting is the first threshold value. And when the value does not reach either of the second threshold values, a control for holding the product value and outputting the proportional operation amount or the product operation amount set to zero to the control target is performed. A feedback control device comprising operation amount control means (16, 37) repeatedly executed at a control cycle.   The feedback control device according to claim 1, wherein absolute values of the first operation value and the second operation value are minimum values that can change a state of the control target.   3. The feedback control device according to claim 1, wherein the first unit value and the second unit value are equal, and the absolute values of the first threshold value and the second threshold value are equal.   The feedback control device according to any one of claims 1 to 3, wherein the control system is a type 1 or type 2 control system for a target value.

Images