JP5813217B2 - Control device and control method for electronically controlled valve - Google Patents

Description

  The present invention relates to a control device and a control method for an electronically controlled valve in which a valve used for positioning control in an in-vehicle application such as an electronic throttle valve and an exhaust gas recirculation valve is driven by a brushless DC motor.

  Conventionally, in an electronic throttle device having two throttle opening sensors provided on a valve rotation shaft and two magnetic sensors provided on a motor rotation shaft, any one of multiplexed throttle opening sensors There is known a control device for an electronic throttle device that specifies a failed throttle opening sensor and continues control using the other normal throttle opening sensor when a failure (abnormality) occurs in For example, see Patent Document 1).

  Hereinafter, the invention described in Patent Document 1 will be described. The present invention relates to an electronic throttle valve that drives a DC motor with an H bridge and rotates a valve rotation shaft directly connected to the motor rotation shaft. Further, according to the present invention, based on an accelerator opening degree sensor that detects the depression amount of the accelerator pedal, a throttle opening degree instruction determined from outputs of a vehicle speed sensor, a temperature sensor, etc., the valve rotation shaft follows this instruction. In addition, position control is executed. The actual throttle opening is realized by position control using a throttle opening command and the output of a double throttle opening sensor provided on the valve rotation shaft.

  Here, the failure detection of the throttle opening sensor is performed in the following steps. First, in the first step, it is determined whether or not the absolute value of the difference between the outputs from the two throttle opening sensors (referred to as “SS_A” and “SS_B”) exceeds a predetermined threshold value. When the absolute value of the difference exceeds a predetermined threshold value, the subsequent second step or third step is executed.

  That is, in the second step, the absolute value of the difference between one ES_A of the outputs from the two magnetic sensors (referred to as “ES_A” and “ES_B”) and one of the throttle opening sensor outputs SS_A is predetermined. It is determined whether or not the threshold value is exceeded. At this time, if the absolute value of the difference does not exceed the predetermined threshold value, the throttle opening sensor that outputs SS_B is determined to be faulty, and position control by SS_A is executed.

  In the second step, the third step is executed when the absolute value of the difference between ES_A and SS_A exceeds a predetermined threshold. In the third step, it is determined whether or not the absolute value of the difference between the output ES_A of the magnetic sensor and the other SS_B of the output of the throttle opening sensor exceeds a predetermined threshold value. At this time, if the absolute value of the difference does not exceed the predetermined threshold value, the throttle opening sensor that outputs SS_A is determined to be faulty, and position control by SS_B is executed.

  When it is determined that both of the two throttle opening sensors are out of order, position control is performed using either one of the outputs from the two magnetic sensors (ES_A or ES_B). Thus, the malfunctioning throttle opening sensor is accurately identified, and the control is continued using the other normal throttle opening sensor.

  Conventionally, in the electronic throttle device having two throttle opening sensors provided on the valve rotation shaft and two magnetic sensors provided on the motor rotation shaft, it is possible to determine the failure of the magnetic sensor or the motor rotor. A control device for an electronic throttle device is known (see, for example, Patent Document 2).

  Hereinafter, the invention described in Patent Document 2 will be described. The present invention also relates to an electronic throttle valve, which is the same as that of Patent Document 1, except that the object of failure detection is a magnetic sensor and a magnet rotor (that is, a motor rotor).

  Here, failure detection of the magnetic sensor and the magnet rotor is performed in the following steps. First, in the first step, it is determined whether or not the absolute value of the difference between the output ES from the magnetic sensor and ES_TH converted from the output SS from the throttle opening sensor exceeds a predetermined threshold value. When the absolute value of the difference exceeds a predetermined threshold value, the subsequent second step or third step is executed.

  That is, in the second step, it is determined whether or not the actual throttle opening SS matches the throttle opening command. In the second step, when the actual throttle opening SS does not coincide with the throttle opening command, it is determined in the third step whether or not the rate of change of the actual throttle opening exceeds a predetermined threshold value.

  In the third step, when the predetermined threshold value is exceeded, it is determined that the magnet rotor is normal and the magnetic sensor is abnormal. In the reverse case, the magnet rotor is abnormal and the magnetic sensor is normal. It is judged. As a result, when the magnetic sensor output predicted based on the actual throttle opening SS is not correct, it is specified whether it is a magnetic sensor failure or a magnet rotor failure.

JP 2001-90588 A JP 2001-98987 A

However, the prior art has the following problems.
In the control device for the electronic throttle device described in Patent Document 1, since it is necessary to provide two sensors for the valve rotation shaft and the motor rotation shaft, there is a problem that the cost increases.

  Further, in order to detect a failure of the throttle opening sensor, a threshold value determination between the two throttle opening sensor outputs is executed, and one and the other of the two magnetic sensor outputs and one of the two throttle opening sensor outputs are performed. In addition, it is necessary to take three steps of performing threshold determination with the other in two stages. For this reason, there is a problem in that the determination process becomes complicated and the processing load on the DSP (Digital Signal Processor) or microcomputer increases.

  Furthermore, the control device of the electronic throttle device described in Patent Document 2 is a failure detection of a magnetic sensor and a magnet rotor on the condition that the throttle opening sensor is normal. However, there is a problem that the determination process is further complicated.

  The present invention has been made to solve the above-described problems, and is an electronic device capable of executing failure detection processing and realizing high-speed and high-accuracy position control with a simple and low-cost configuration. It is an object to obtain a control device and a control method for a control valve.

An electronically controlled valve control device according to the present invention controls a drive source that applies a drive torque against a return torque for a valve mechanism in which a return torque acts in a valve opening direction or a valve closing direction. A control device for an electronically controlled valve that controls opening and closing of a mechanism, a brushless DC motor that is a drive source, and a first position that detects a motor shaft rotation angle of the brushless DC motor and outputs a discrete signal in a pulse format A detector, a second position detector that detects a valve shaft rotation angle of the valve mechanism and outputs a continuous signal as an analog voltage, and a first position detection that is driven by a reference command input from outside The relationship between the motor shaft rotation angle and the analog voltage is obtained based on the discrete motor shaft rotation angle output from the motor and the analog voltage output from the second position detector. Calibration processing unit, and a correction processing unit that converts the analog voltage output from the second position detector into the valve shaft rotation angle based on the relationship between the motor shaft rotation angle and the analog voltage obtained by the calibration processing unit. A control unit that generates a drive command for the brushless DC motor based on a valve shaft rotation angle output from the correction processing unit and a valve shaft position command input from the outside, and a first position detector a motor shaft rotation angle to be output, based on the valve shaft rotation angle output from the correction processing unit detects a malfunction of the second position detector, whether to apply the output of the first position detector A failure detection processing unit that performs the above determination , and the calibration processing unit is periodically executed after the valve is assembled or mounted on the vehicle .

Further, the electronically controlled valve control method according to the present invention controls a drive source that applies a drive torque against the return torque for a valve mechanism in which a return torque acts in the valve opening direction or the valve closing direction. An electronically controlled valve control method executed by an electronically controlled valve controller that controls opening and closing of a valve mechanism, the electronically controlled valve controller comprising: a brushless DC motor as a drive source; and a brushless DC motor A first position detector that detects the motor shaft rotation angle and outputs a discrete signal in the form of a pulse; and a second position that detects the valve shaft rotation angle of the valve mechanism and outputs a continuous signal as an analog voltage A discrete motor shaft rotation angle output from the first position detector at that time, driven by a reference command input from the outside, and a second position Based on the analog voltage output from the detector, a calibration processing step for determining the relationship between the motor shaft rotation angle and the analog voltage, and based on the relationship between the motor shaft rotation angle and the analog voltage determined in the calibration processing step. The correction processing step for converting the analog voltage output from the second position detector into the valve shaft rotation angle, the valve shaft rotation angle converted in the correction processing step, and the valve shaft position command input from the outside, based on a control step of generating a drive command to the brushless DC motor, a motor shaft rotation angle outputted from the first position detector, based on the converted valve shaft rotational angle correction processing steps, the detecting a malfunction of the two-position detector, comprising: a failure detection processing step of determining whether or not to apply the output of the first position detector, a key Calibration processing steps are those regularly executed after the valve assembly or vehicle mounted.

According to the control device and control method for an electronically controlled valve according to the present invention, the calibration processing unit (step) is driven by a reference command input from the outside, and is output from the first position detector at that time. Based on the discrete motor shaft rotation angle and the analog voltage output from the second position detector, the relationship between the motor shaft rotation angle and the analog voltage is obtained, and the correction processing unit (step) is a calibration processing unit. The analog voltage output from the second position detector is converted into the valve shaft rotation angle based on the relationship between the motor shaft rotation angle and the analog voltage obtained in step 5, and the control unit (step) is output from the correction processing unit. A drive command to the brushless DC motor is generated based on the valve shaft rotation angle and the valve shaft position command input from the outside, and the failure detection processing unit (step) A motor shaft rotation angle output from the first position detector, based on the valve shaft rotation angle output from the correction processing unit, for detecting at least one of the failure of the first position detector and a second position detector.
Therefore, the failure detection process can be executed with a simple and low-cost configuration, and high-speed and highly accurate position control can be realized.

It is a block block diagram which shows the control apparatus of the electronically controlled valve | bulb which concerns on Embodiment 1 of this invention with a valve mechanism. It is a block block diagram which shows the calibration process part of FIG. 1 in detail. It is a block block diagram which shows the correction process part of FIG. 1 in detail. It is a block block diagram which shows the control part of FIG. 1 in detail. It is a block block diagram which shows the 1st control part of FIG. 4 in detail. It is a block block diagram which shows the 2nd control part of FIG. 4 in detail. It is a block block diagram which shows the failure detection process part of FIG. 1 in detail. It is a flowchart which shows the condition determination process in the failure detection process part of FIG. It is a block block diagram which shows the control apparatus of the electronically controlled valve | bulb which concerns on Embodiment 2 of this invention with a valve mechanism. It is a block block diagram which shows the 2nd control part in the control part of FIG. 9 in detail. It is a block block diagram which shows the failure detection process part of FIG. 9 in detail. It is a flowchart which shows the 1st condition determination process in the failure detection process part of FIG. It is a flowchart which shows the 2nd condition determination process in the failure detection process part of FIG. It is explanatory drawing which shows the input / output relationship of the phase corrector in the control apparatus of the electronically controlled valve | bulb which concerns on Embodiment 3 of this invention. It is explanatory drawing which shows the process which calculates | requires the specific phase correction amount in the phase corrector of the control apparatus of the electronically controlled valve which concerns on Embodiment 3 of this invention. It is a block block diagram which shows the 2nd control part of the control apparatus of the electronically controlled valve | bulb which concerns on Embodiment 4 of this invention in detail. It is explanatory drawing which shows the process of a dead zone correction | amendment device in the 2nd control part of the control apparatus of the electronically controlled valve | bulb which concerns on Embodiment 4 of this invention. It is explanatory drawing which shows the example of a measurement of a phase voltage command and winding current in describing the process of a dead zone correction | amendment device of the control apparatus of the electronically controlled valve which concerns on Embodiment 4 of this invention.

  Hereinafter, preferred embodiments of a control device and a control method for an electronically controlled valve according to the present invention will be described with reference to the drawings. In the drawings, the same or corresponding parts are denoted by the same reference numerals. To do.

Embodiment 1 FIG.
A control device for an electronically controlled valve according to Embodiment 1 of the present invention will be described with reference to FIGS. FIG. 1 is a block configuration diagram showing a control device for an electronically controlled valve according to Embodiment 1 of the present invention, together with a valve mechanism 1. Hereinafter, the valve mechanism 1 and the electronically controlled valve control device are collectively referred to as a control system.

  In FIG. 1, this control system includes a valve mechanism 1, a brushless DC motor (BLDCM) 2, an inverter 3 that is a drive circuit for the brushless DC motor 2, a PWM processing unit 4, and a motor rotating shaft (specifically, a brushless DC motor 2). , A first position detector 5 for detecting the angle (motor shaft rotation angle) of the motor rotor and a second position detector 6 for detecting the angle of the valve rotation shaft (valve shaft rotation angle) of the valve mechanism 1. ing.

  The control system also includes a calibration processing unit 10 that performs calibration based on the detection values of the first position detector 5 and the second position detector 6, and the detection values of the second position detector 6. A correction processing unit 20 for applying correction, a control unit 30 for setting the valve shaft rotation angle to a predetermined position, and a failure detection processing unit 40 for determining a failure of the first position detector 5 and the second position detector 6 Is provided.

  The actuator that drives the valve mechanism 1 is a brushless DC motor 2. Further, the brushless DC motor 2 is provided with a first position detector 5 that outputs, for example, the magnetic pole position of a magnet magnetized on the rotor in a pulse output format such as a Hall IC. The angular resolution of the first position detector 5 is, for example, 60 deg, 30 deg or 15 deg in electrical angle, and the motor shaft rotation angle is discretely detected.

  The second position detector 6 that detects the valve shaft rotation angle of the valve mechanism 1 includes, for example, a magnet on the valve shaft, and a non-contact type magnetism that outputs a magnetic field change accompanying the rotation of the valve shaft in an analog output format. Let it be a sensor.

  Note that a spring (not shown) is connected to the valve mechanism 1 as an urging means. A preload is applied to the spring, and a return torque by the spring acts, for example, in a valve closing direction of a valve shaft (not shown) of the valve mechanism 1.

  On the other hand, a spring is connected to the brushless DC motor 2 via a power transmission mechanism (not shown) such as a gear reducer interlocked with the motor rotation shaft, and the valve mechanism 1 is not controlled (for example, energization cut). In a state where the valve shaft is pressed against the machine end by a return torque due to the preload of the spring, a so-called mechanical fail-safe function is provided.

  Next, a schematic flow of control of the electronically controlled valve control device will be described. This control device is mainly intended to set the valve shaft rotation angle to a predetermined angle, and feedback control is set so that the actual position (valve shaft rotation angle) follows the position command.

  Specifically, the control unit 30 calculates a drive command based on the position command and the actual position, and inputs the drive command to the PWM processing unit 4 so that the PWM processing unit 4 outputs the PWM command. . Further, the brushless DC motor 2 is driven to follow the target value by switching the high side arm and the low side arm of the inverter 3 at an appropriate timing based on the PWM command.

  Here, since the second position detector 6 provided in the valve mechanism 1 is an analog output type, it is necessary to scale the actual position, which is the feedback amount in the feedback control, to the dimension of the angle. Therefore, the following initialization operation is executed.

  First, before assembling feedback control, for example, a certain amount of voltage is given as a drive command as a reference command, and the brushless DC motor 2 is driven in one direction in an open loop. At this time, the output of the first position detector 5 that detects the motor shaft rotation angle of the brushless DC motor 2 and the valve shaft rotation angle as an analog voltage that is the output of the second position detector 6 provided in the valve mechanism 1. Based on the above, the calibration processing unit 10 obtains the relationship between the motor shaft rotation angle and the analog voltage.

  Note that in this initialization operation, the reference command and the PWM processing unit 4 are connected, and the calibration processing unit 10 and the correction processing unit 20 are not connected. When this initialization operation is completed, the calibration processing unit 10 and the correction processing unit 20 temporarily transfer the relationship between the motor shaft rotation angle obtained by the calibration processing unit 10 and the analog voltage to the correction processing unit 20. Connected and disconnected again when porting is complete. Thereafter, the PWM processing unit 4 and the control unit 30 are connected, and feedback control is possible.

  Further, the correction processing unit 20 has, as information, the relationship between the motor shaft rotation angle obtained by the calibration processing unit 10 and the analog voltage, and the reduction ratio of the power transmission mechanism, from the second position detector 6. The feedback control system is configured to correct the valve shaft rotation angle to the valve shaft rotation angle and output this as the actual position, thereby causing the valve shaft rotation angle to follow the target value.

  Further, the failure detection processing unit 40 determines the failure of the second position detector 6 based on the valve shaft rotation angle that is the output of the correction processing unit 20 and the motor shaft rotation angle that is the output of the first position detector 5. The determination is executed, and the determination result is output to the PWM processing unit 4. Note that the failure detection processing unit 40 always performs failure determination of the second position detector 6 during operation of the control device.

  The above is a schematic flow of the control of the electronically controlled valve control device. Next, each part of the calibration processing unit 10, the correction processing unit 20, the control unit 30, and the failure detection processing unit 40 that executes the arithmetic processing will be described in detail.

  First, the calculation processing of the calibration processing unit 10 will be described with reference to FIG. FIG. 2 is a block diagram showing in detail the calibration processing unit 10 of FIG. The calibration processing unit 10 includes an electrical angle time-series table 11, a voltage time-series table 12, and a table 13 indicating a relationship between the voltage and the electrical angle. The output of the first position detector 5 and the second position detection are performed. The output of the device 6 is taken as input.

Since the calculation processing of the calibration processing unit 10 is executed in a state where the brushless DC motor 2 is forcibly driven in a state where the reference command and the PWM processing unit 4 are connected, the calculation of the first position detector 5 is performed. By counting up the pulse-like input at the edge of the pulse, the electrical angle time series table 11 is generated. Here, the output of the first position detector 5 is a discrete signal as described above, and one step in a step shape corresponds to the electrical angle resolution θ _e0 .

  At the same time, the voltage time series table 12 is generated based on the analog input voltage of the second position detector 6. At this time, the electrical angle time-series table 11 and the voltage time-series table 12 are acquired in synchronism, so that when the two are combined, the voltage for each electrical angle step is acquired discretely. As a result, a table 13 showing the relationship between the voltage and the electrical angle is generated. In addition, the relationship between this voltage and an electrical angle is represented as a function of following Formula (1).

In the formula (1), f (V) indicates that it is a function of the voltage V. The simplest example of the expression (1) is an approximation in which the electrical angle θ _e is a linear expression of the voltage V, but the electrical characteristics of the second position detector 6 are not completely linear. In consideration, it may generally be approximated as a polynomial of voltage V. As a result, the calibration processing unit 10 outputs Expression (1) that is a relationship between the voltage and the electrical angle.

  That is, the calibration processing unit 10 outputs information on the voltage and the electrical angle corresponding to the expression (1), but the output format is not limited to this, and for example, the relationship between the voltage and the electrical angle is expressed. The table 13 shown may be output as it is, and the table may be extrapolated or interpolated in the correction processing unit 20 described later.

  In addition, the arithmetic processing in the calibration processing unit 10 may be executed not only once after the assembly of the electronically controlled valve, but also periodically after the valve is mounted on the vehicle.

  Next, the calculation processing of the correction processing unit 20 will be described with reference to FIG. FIG. 3 is a block diagram showing in detail the correction processing unit 20 of FIG. In FIG. 3, the correction processing unit 20 includes a correction processing unit 21 of a transmission mechanism system and a detector system, and is based on the relationship between the voltage and the electrical angle described above and the analog voltage from the second position detector 6. Then, the voltage is corrected using the following equation (2).

In the formula (2), θ _v represents the valve shaft rotation angle, which is the output of the correction processing section 20, p _n denotes the motor pole pairs, eta represents the reduction ratio. Further, in the equation (2), since _pn and η are constants, the valve shaft rotation angle corrected from the voltage that is the output of the second position detector 6 can be obtained.

  In this way, the calibration processing unit 10 obtains the relationship between the voltage and the electrical angle in the initialization operation and transplants the relationship to the correction processing unit 20, so that the output voltage of the second position detector 6 can be scaled to the angle. It becomes possible. Further, since the calibration processing unit 10 expresses f (V) in the equation (1) by polynomial approximation, the nonlinear characteristic peculiar to the analog output format of the second position detector 6 is also corrected, and the valve shaft It is possible to increase the resolution of the rotation angle.

  Furthermore, it is possible to correct variations in output caused by tolerances during assembly, such as adjusting the gap between the magnet and the detection surface, and individual sensor differences. Compared with the case where the single characteristic is applied as it is, the output variation of the sensor can also be suppressed.

  Next, the calculation process of the control unit 30 will be described with reference to FIGS. FIG. 4 is a block diagram showing in detail the control unit 30 of FIG. In FIG. 4, the control unit 30 includes a first control unit 31 and a second control unit 32, and a position command, an actual position (valve shaft rotation angle), and a bus voltage of the inverter 3 are input and a drive command is output. To do.

  In the feedback control state described below, the reference command and the PWM processing unit 4 are disconnected, and the control unit 30 and the PWM processing unit 4 are connected. Further, the calibration processing unit 10 and the correction processing unit 20 are disconnected.

  FIG. 5 is a block diagram showing in detail the first control unit 31 of FIG. In FIG. 5, the first control unit 31 includes a dead zone 311, a PID controller 312, and a saturator 313. The first control unit 31 receives a position command and an actual position (valve shaft rotation angle) and outputs a current command, specifically a q-axis current command. This is a control system in which a current command is calculated and output so that the deviation is zero.

  The dead zone 311 acts on this actual deviation, and a specific calculation is represented by the following expression (3). In Equation (3), e represents the actual deviation between the position command and the actual position, ε represents the dead zone width, k represents the slope between the dead zones, and e ′ represents the corrected deviation.

  This dead zone 311 is generally used in PID control. However, in an electronically controlled valve in which a load resistance such as a working fluid acts, particularly in a region where the actual deviation is small, an output of the first control unit 31 is used. By reducing the sensitivity of a certain current command, it is possible to suppress fine vibration when the valve is held at an arbitrary opening degree.

  The PID controller 312 is a controller composed of a proportional unit, an integrator, and a differentiator (preferably, a pseudo-differentiator). FIG. 5 shows a configuration of differential preceding type PI-D control as an example. However, any other configuration is possible as long as it falls within the category of classical control theory such as I-PD control. The saturator 313 limits the current command with the upper and lower limit values.

  FIG. 6 is a block diagram showing in detail the second control unit 32 of FIG. In FIG. 6, the second control unit 32 includes a current command distributor 321, an induced voltage estimator 322, two current controllers 323A and 323B, two current estimators 324A and 324B, a saturator 325, a voltage corrector 326, A valve shaft / electrical angle converter 327, a filter 328, and a phase corrector 329 are provided.

  Hereinafter, the operation of each unit of the second control unit 32 will be described. The second control unit 32 receives a current command, an actual position (valve shaft rotation angle) and a bus voltage of the inverter 3 and outputs a phase voltage command which is a drive command, and performs so-called current control. Realized without a current sensor.

The current command distributor 321 performs processing for distributing the current command (q-axis current command) i _{q_com} to the U-phase current command i _{u_com} and the V-phase current command _{iv_com} . Here, the electrical angle required for distribution is obtained by the following equation (4) in which the valve shaft rotation angle, which is the actual position, is converted into the motor shaft rotation angle (electrical angle) by the valve shaft / electrical angle converter 327. .

  Since the electric angle which changes continuously is given by the equation (4), sinusoidal driving, which is one of the driving methods of the brushless DC motor 2, can be performed. Needless to say, in addition to sine wave drive, 120-degree conduction rectangular wave drive and 180-degree conduction rectangular wave drive are also possible.

  The current command distributor 321 distributes the direct current to the alternating current using the following equation (5).

Here, the right side C (θ _e , Δθ _e ) of Expression (5) is a function of the electrical angle θ _e and a phase correction amount Δθ _e described later, and a 3 × 2 direction cosine matrix having a trigonometric function as an element. It is. Further, the W-phase current command i _{w_com on} the left side of the equation (5) needs to be calculated because in the current control of the brushless DC motor 2, if two of the three phases can be controlled, the remaining one phase can be automatically controlled. There is no. Also, the d-axis current command i _{d_com on} the right side of Equation (5) is also set to zero for simplicity.

Therefore, the input / output relationship in the current command distributor 321 is that the input is the current command, the electrical angle output from the valve shaft / electrical angle converter 327, and the electrical angle obtained by pseudo-differentiating this electrical angle with the filter 328. The angular velocity and the phase correction amount Δθ _e output from the phase corrector 329, and the outputs are the U-phase current command i _{u_com} and the V-phase current command _{iv_com} . The filter 328 and the phase corrector 329 will be described later.

The induced voltage estimator 322 estimates an induced voltage generated in each phase winding of the brushless DC motor 2 during driving. The U-phase estimated induced voltage _{eu_est} and the V-phase estimated induced voltage _{ev_est} , which are the outputs of the induced voltage estimator 322, are expressed by the following equation (6).

Here, ω _{e on} the right side of Equation (6) is an electrical angular velocity, and e _norm is, for example, U generated in the brushless DC motor 2 when the brushless DC motor 2 is driven at a certain rotational speed from the outside. This is a voltage per unit angular velocity obtained by dividing one amplitude of the phase induced voltage and the V phase induced voltage by the rotation speed, and D (θ _e , Δθ _e ) is a function of the electrical angle θ _e and the phase correction amount Δθ _e. Is a 2 × 1 vector having trigonometric function elements expressed as

Subsequently, the filter 328 and the phase corrector 329 will be described. The filter 328 generates an electrical angular velocity ω _e by time-differentiating the electrical angle obtained as the output of the valve shaft / electrical angle converter 327. For example, if the filter 328 is a pseudo-differentiator, it is expressed by the following equation (7).

  In the formula (7), T and α are constants. By adding a first-order lag element that is a characteristic of the right-hand side denominator as in Expression (7), it can be made less susceptible to observation noise and the like. As a matter of course, the filter 328 may be a simple element such as a moving average filter, a forward difference, and a backward difference in addition to the pseudo-differentiator represented by the equation (7). Any element may be used as long as it is a calculation for obtaining.

  Next, the phase corrector 329 will be described. Generally, when there is a current loop in the motor control as in the second control unit 32, the actual position reading delay (phase delay) is caused by the influence of the control period of the current loop as the motor angular velocity increases. Increase.

  In comparison with FIG. 6, the phase of the phase current command that is the output of the current command distributor 321 and the phase of the estimated induced voltage that is the output of the induced voltage estimator 322 are the responses of the actual phase current and the actual induced voltage of the actual motor. As a result, the response of the brushless DC motor 2 decreases. Therefore, the phase corrector 329 is introduced from the viewpoint of correcting the phase delay of the control calculation caused by the control cycle in order to eliminate this decrease in response.

The phase corrector 329 outputs the phase correction amount Δθ _e calculated using the following equation (8) to the current command distributor 321 and the induced voltage estimator 322. In Expression (8), T _θ represents the time of one electrical angle cycle at a constant rotational speed of the brushless DC motor 2, and Δt _c represents the control cycle of the second control unit 32.

In other words, the phase correction that is the output of the phase corrector 329 is performed with respect to the electrical angle that is required when estimating the phase current command that is the output of the current command distributor 321 and the estimated induced voltage that is the output of the induced voltage estimator 322. By adding the amount Δθ _e , it is possible to suppress the calculation delay of the current command of each phase and the calculation delay of the induced voltage due to the control cycle of the second control unit 32 that controls the current. The phase correction amount Δθ _e can be set as a fixed value determined from, for example, the maximum angular velocity that can be generated by the brushless DC motor 2.

Subsequently, the current controllers 323A and 323B respectively output a U-phase current command i _{u_com} and a V-phase current command i _{v_com} that are outputs of the current command distributor 321 and a U-phase estimated current i _{u_est} and a V-phase estimated current i _{v_est} described later. preparative _i u_com -i u_est a current deviation respectively difference calculation, based on the _i v_com -i v_est, and generates a U-phase voltage command and the V-phase voltage command. As the simplest example, the current controllers 323A and 323B are configured with only a proportional device so as to output a voltage command proportional to the current deviation, which is a known technique. In addition, the current controllers 323A and 323B may be configured by a proportional device and an integrator.

  Next, the voltage corrector 326 will be described. In electronically controlled valves used for in-vehicle applications, a battery is often used as the power source. Generally, the battery undergoes a voltage change such as a decrease in battery voltage due to aging, or a change in battery voltage when the key is turned on. Specifically, for example, when the brushless DC motor 2 is driven at the rated voltage of the inverter 3 and the voltage level decreases, even if the drive command output from the control system is the same, the brushless DC motor 2 The rotational speed decreases.

  From the viewpoint of the control system, this is equivalent to a reduction in the gain of the control system. Therefore, it is necessary to correct the drive command output from the control system in order to realize stable driving without impairing responsiveness even when the power source changes in this way. Therefore, the voltage corrector 326 corrects the phase voltage command, which is the output of the current controllers 323A and 323B, using the following equation (9).

Here, the left sides V _{u_comp} , V _{v_comp,} and V _{w_comp} of Equation (9) are phase voltage commands for each phase as the output of the voltage corrector 326. Further, regarding the right side of the equation (9), V _base represents a reference voltage (fixed value), V _bus represents the bus voltage of the inverter 3, and V _u ^* and V _v ^* are outputs of the current controllers 323A and 323B. It is a phase voltage command of a certain U phase and V phase, and V _w ^* is a phase voltage command of W phase calculated from the above V _u ^* and V _v ^* . Thus, when V _bus is smaller than V _base , the phase voltage command is increased, and conversely, when V _bus is larger than V _base , the operation of reducing the phase voltage command is automatically performed. Therefore, it is possible to prevent deterioration of responsiveness due to voltage fluctuation.

  Next, the saturator 325 limits the single amplitude of the U-phase voltage command, the V-phase voltage command, and the W-phase voltage command according to the voltage level of the inverter 3.

Subsequently, the current estimators _{324 </ b} ^{> A} and _{324 </ b} ^> _B output the U-phase voltage command V _{u_comp} ^* and the V-phase voltage command V _{v_comp} ^* obtained as outputs of the saturator 325 and the U-phase obtained as outputs of the induced voltage estimator 322 described above. Based on the estimated induced voltage _{eu_est} and the V-phase estimated induced voltage _{ev_est} , the phase current flowing in the U phase and the V phase is estimated.

  In general, in general current control in which the phase current of the three-phase motor is individually feedback-controlled, the gain of the second control unit 32 (that is, the gain of the current controllers 323A and 323B) is set so that the current response does not oscillate. The high gain is set to 1 and the current control bandwidth is widened to improve the response.

  On the other hand, when pseudo current feedback control is executed without using a current sensor, the estimation accuracy of the induced voltage and the accuracy of the current model of the current estimators 324A and 324B are high, and the second control unit 32 If the gain (that is, the gain of the current controllers 323A and 323B) is set to a high gain to such an extent that the current response does not oscillate, the actual phase current and the estimated phase current match well, and the current response is improved. Further, in such a state, the phase voltage command can be regarded as matching the actual phase voltage of the three-phase motor.

Therefore, in order to calculate the U-phase estimated current i _{u_est} and the V-phase estimated current i _{v_est} , as a current model of the current estimators 324A and 324B, for example, the winding resistance per phase including the harness is R, A first-order lag element having a winding inductance L is set. Here, the specific calculation of the current estimators 324A and 324B is expressed by the following expressions (10) and (11), where the control period of the second control unit 32 is Δt _c . In Expressions (10) and (11), n is the number of samplings and is a positive integer.

The above is the current estimator 324A, U-phase estimated current i _{U_est} and V-phase estimated current i _{V_est} calculation method in the case where modeled as first-order lag element 324B.

  As described above, when the valve mechanism 1 is controlled to open and close, the estimated induced voltages of the U and V phases that are the output of the induced voltage estimator 322 and the phase voltage commands of the U and V phases that are the output of the saturator 325 are used. Based on the above, estimated currents (Equation (10) and Equation (10) for each phase calculated by current estimators 324A and 324B, in which the relationship between the voltage and current per phase of the three-phase motor is modeled by a first-order lag element) 11) is fed back to the phase current command, a pseudo current feedback control system that does not require a current sensor, that is, a current sensorless control system can be configured.

  Also, in applications where the valve mechanism 1 is controlled to be opened and closed by the brushless DC motor 2, particularly in the valve mechanism 1 provided with a spring for fail-safe purposes, the return torque due to the spring is superimposed on the motor-generated torque, and the speed is likely to increase. When driving in the direction, the increase in the induced voltage due to the increase in speed reduces the current flowing in the brushless DC motor 2, making it difficult to flow the current for operating the brake near the stop position, and the valve collides with the machine end. The situation of bouncing can occur. However, since the current control due to the influence of the winding inductance can be compensated for by the action of the second control unit 32 constituting the above-described pseudo current feedback control system, a high-speed response is realized without colliding with the machine end. be able to.

  Next, the calculation process of the failure detection processing unit 40 will be described with reference to FIG. FIG. 7 is a block diagram showing in detail the failure detection processing unit 40 of FIG. In FIG. 7, the failure detection processing unit 40 includes an electrical angle time-series table 41, an electrical angle / valve shaft converter 42, and a condition determination unit 43. The pulse-like output of the 1-position detector 5 is input, and a drive availability determination flag is output to the PWM processing unit 4.

The pulse-like input of the first position detector 5 is counted up at the edge of the pulse, and the electrical angle time series table 41 is generated. At this time, since the output of the first position detector 5 is a discrete signal, one step in a step shape corresponds to the electrical angle resolution θ _e0 . The electrical angle / valve shaft converter 42 converts the discrete signal output from the electrical angle time series table 41 into the valve shaft rotation angle using the following equation (12).

In equation (12), θ _{ed_v} indicates the output of the electrical angle / valve shaft converter 42, θ _ed indicates the input of the electrical angle / valve shaft converter 42, pn indicates the _number of motor pole pairs, and η indicates deceleration. The ratio is shown. The number of motor pole pairs _pn and the reduction ratio η are constants.

Subsequently, the condition determination unit 43 executes failure determination of the second position detector 6 based on the deviation between θ _{ed_v} calculated by the electrical angle / valve shaft converter 42 and the valve shaft rotation angle θ _v. . A specific calculation process of the condition determination unit 43 will be described with reference to the flowchart of FIG.

First, the condition determining unit 43 determines whether the absolute value of the deviation between the calculated theta _{Ed_v} Valve shaft rotation angle theta _v and the electrical angle / bulb axis converter 42 is less than or equal to angle threshold theta _th (Step S1).

In step S1, when it is determined that the absolute value of the deviation is equal to or smaller than the angle threshold value _θth (that is, Yes), it is determined that the second position detector 6 is normal and the drive is enabled or disabled. A determination flag is output (step S2), and the process of FIG.

On the other hand, if it is determined in step S1 that the absolute value of the deviation is not equal to or smaller than the angle threshold _θth (that is, No), it is determined that the second position detector 6 is abnormal and cannot be driven. A drive propriety determination flag is output (step S3), and the process of FIG. 8 ends.

  This drive propriety determination flag is output to the PWM processing unit 4. If the received flag is a flag that cannot be driven, the drive command is forcibly set to zero in the PWM processing unit 4 and the brushless DC motor 2 is forcibly stopped. Let That is, the first position detector 5 and the first position detector 5 are compared by comparing the valve shaft rotation angle, which is the output of the correction processing unit 20, with the value obtained from the output of the first position detector 5 using Expression (12). Even if the two-position detector 6 has a single configuration, a failure of the second position detector 6 can be detected with simple logic, and the motor can be forcibly stopped.

  As described above, the feedback control is executed with the amount of the output of the second position detector 6 provided in the valve mechanism 1 with high resolution, and the current control system without the current sensor is configured in the inner loop. Highly accurate position control can be realized. Further, by providing a single sensor for each of the valve mechanism 1 and the brushless DC motor 2 to reduce the number of sensors, it is possible to easily and cost-effectively detect the position detector.

As described above, according to the first embodiment, the calibration processing unit (step) is discrete in the form of pulses output from the first position detector when driven by a reference command input from the outside. The relationship between the motor shaft rotation angle and the analog voltage is obtained based on the correct motor shaft rotation angle and the analog voltage output from the second position detector, and the correction processing unit (step) is obtained by the calibration processing unit. Based on the relationship between the motor shaft rotation angle and the analog voltage, the analog voltage output from the second position detector is converted into the valve shaft rotation angle, and the control unit (step) outputs the valve output from the correction processing unit. A drive command to the brushless DC motor is generated based on the shaft rotation angle and the valve shaft position command input from the outside, and the failure detection processing unit (step) is output from the first position detector. A motor shaft rotation angle that, based on the valve shaft rotation angle output from the correction processing unit, for detecting at least one of the failure of the first position detector and a second position detector.
For this reason, failure detection processing is executed with a simple and low-cost configuration, and high-speed and high-accuracy position control is realized using the valve shaft rotation angle in which the output variation of the second position detector is suppressed as a feedback amount. Can do.

Embodiment 2. FIG.
An electronically controlled valve control apparatus according to Embodiment 2 of the present invention will be described with reference to FIGS. FIG. 9 is a block configuration diagram showing an electronically controlled valve control apparatus according to Embodiment 2 of the present invention together with a valve mechanism.

  When the second embodiment of the present invention is different from the first embodiment described above with reference to FIG. 9, the actual position input to the control unit 30 </ b> A is the output of the correction processing unit 20 and the first position detector 5. The point selected by the output and the position command input to the control unit 30A (not shown) are changed in accordance with the selection of the actual position.

  FIG. 10 is a block diagram showing in detail the second control unit 32A in the control unit 30A of FIG. In FIG. 10, the point that the output of the valve shaft / electrical angle converter 327 and the output of the first position detector 5 are selected in accordance with the selection of the actual position is different from the above-described first embodiment.

  Further, in FIG. 9, the configuration of the failure detection processing unit 40A is different from that of the above-described first embodiment by considering the selection of the actual position. Hereinafter, the configuration and calculation processing of the failure detection processing unit 40A will be described in detail.

  The main object of the second embodiment of the present invention is to continue the opening / closing control of the valve mechanism 1 using the output of the first position detector 5 even when the second position detector 6 fails. It is to do.

  FIG. 11 is a block diagram showing in detail the failure detection processing unit 40A of FIG. In FIG. 11, the failure detection processing unit 40A includes a two-stage first condition determining unit 44 and second condition determining unit 45 instead of the condition determining unit 43 shown in FIG.

A specific calculation process of the first condition determination unit 44 will be described with reference to the flowchart of FIG.
First, as in the first embodiment, the first condition determination unit 44 determines that the absolute value of the deviation between the valve shaft rotation angle θ _v and the θ _{ed_v} calculated by the electrical angle / valve shaft converter 42 is the angle threshold θ It is determined whether it is equal to or less than _th (step S1).

In step S1, when it is determined that the absolute value of the deviation is equal to or smaller than the angle threshold _θth (that is, Yes), it is determined that the second position detector 6 is normal, and the second position detector 6 A flag indicating normality is output (step S11), and the process of FIG. 12 is terminated.

On the other hand, if it is determined in step S1 that the absolute value of the deviation is not equal to or smaller than the angle threshold _θth (that is, No), it is determined that the second position detector 6 is abnormal, and the second position detector 6 Is output as abnormal (step S12), and the process of FIG. 12 is terminated.

Next, a specific calculation process of the second condition determination unit 45 will be described with reference to the flowchart of FIG.
First, the second condition determination unit 45 determines whether or not the second position detector 6 is normal based on the flag output from the first condition determination unit 44 (step S21).

  If it is determined in step S21 that the second position detector 6 is normal (that is, Yes), a drive enable / disable determination flag indicating that the second position detector 6 can be driven is output to the PWM processing unit 4 (step S22). The process of FIG. 13 is terminated. Thereby, the opening / closing control of the valve mechanism 1 is executed using the output of the second position detector 6 as a feedback amount.

  On the other hand, if it is determined in step S21 that the second position detector 6 is abnormal (that is, No), the second condition determination unit 45 applies the output of the first position detector 5 as a feedback amount. It is determined whether or not to perform (step S23).

  Note that this determination may be made simply by determining whether or not to apply the output of the first position detector 5. Further, the present invention is not limited to this. For example, when the feedback loop is once cut and the brushless DC motor 2 is driven at a constant rotational speed by a reference command, the count number of the first position detector 5 calculated from the constant rotational speed Whether the absolute value of the difference between the theoretical time width and the time width when the pulses actually output from the first position detector 5 are counted is equal to or smaller than the predetermined time width is determined by the first position detector 5. You may decide whether to apply the output.

  If it is determined in step S23 that the output of the first position detector 5 is to be applied (that is, Yes), the second condition determination unit 45 sets a drive availability determination flag indicating that the drive is possible to the PWM processing unit 4. (Step S24), and the process of FIG. 13 is terminated.

  As a result, the control is shifted to the opening / closing control of the valve mechanism 1 using the output of the first position detector 5 as the feedback amount. The actual position referring to the device 5 and the position command converted to the motor shaft rotation angle are respectively switched. Further, the actual position that is an input to the second control unit 32 </ b> A in the control unit 30 </ b> A is also switched to the actual position with reference to the first position detector 5.

  On the other hand, if it is determined in step S23 that the output of the first position detector 5 is not applied (that is, No), the second condition determination unit 45 sets a drive availability determination flag indicating that the drive is impossible as a PWM. It outputs to the process part 4 (step S25), and complete | finishes the process of FIG. Thereby, the brushless DC motor 2 is stopped.

  In this way, even if the second position detector 6 is abnormal, it is possible to continue drive control using the output of the first position detector 5 as a position feedback amount, and the position detectors can be connected to each other. Together with simple logic to monitor, the fault tolerance of the electronically controlled valve can be improved.

In the second embodiment, when the output of the first position detector 5 is applied as the feedback amount, a discrete electrical angle signal is used as the motor shaft rotation angle of the brushless DC motor 2. Therefore, the matrix C (θ _e , Δθ _e ) and the vector D (θ _e , Δθ _e ) shown in the equations (5) and (6) are discrete values for each electrical angle resolution. .

In this case, the calculation of the trigonometric function used in the first embodiment is not necessary, and the elements of the matrix C (θ _e , Δθ _e ) and the vector D (θ _e , Δθ _e ) are assumed to be zero-order held for each electrical angle resolution. Can be computed without using trigonometric functions. Therefore, it is possible to reduce a calculation load on a processing device such as a DSP (Digital Signal Processor) or a microcomputer that performs an operation for driving an electronically controlled valve.

Embodiment 3 FIG.
An electronically controlled valve control apparatus according to Embodiment 3 of the present invention will be described with reference to FIGS. FIG. 14 is an explanatory diagram showing the input / output relationship of the phase corrector 329A in the electronically controlled valve control apparatus according to Embodiment 3 of the present invention. The phase corrector 329A is applied to the second control units 32 and 32A of the first and second embodiments described above.

  As described in the first embodiment, the phase corrector 329 is introduced for the purpose of correcting the phase delay due to the control period of the second control units 32 and 32A. 2 explained that the correction amount for the phase delay is a fixed value. However, immediately after starting from the state where the brushless DC motor 2 is stopped, there is no phase delay in capturing the actual position. Therefore, if the correction amount is a fixed value, the brushless DC motor 2 cannot be started or the brushless DC The motor 2 may be driven in an unintended direction and may oscillate in some cases, which may result in an unsuitable state for an electronically controlled valve.

  Accordingly, the phase corrector 329A shown in FIG. 14 has a configuration in which the phase correction amount can be changed based on the angular velocity (electrical angle conversion) that is the output of the filter 328. FIG. 15 shows a process for obtaining a specific phase correction amount in the phase corrector 329A.

  FIG. 15A shows the profile of the motor shaft rotation angle (electrical angle) when the valve is driven and controlled at a fully closed state of 0 deg and is set to a predetermined angle, and FIG. 15B shows the angle profile. The angular velocity profile of the motor in is shown.

  Here, in the phase corrector 329A, as shown in FIG. 15C, the phase correction amount in the stopped state is set to 0, and the angular velocity threshold is set as a boundary while referring to the angular velocity profile in FIG. Outputs a constant phase correction amount. As another form, as shown in FIG. 15D, an output with variable phase correction amount is obtained by using curve approximation or linear approximation as a function of angular velocity.

  By doing so, it is possible to eliminate a state unsuitable for an electronically controlled valve such that the brushless DC motor 2 cannot be started or the brushless DC motor 2 is driven in an unintended direction and oscillates in some cases. .

Embodiment 4 FIG.
A control device for an electronically controlled valve according to Embodiment 4 of the present invention will be described with reference to FIGS. FIG. 16 is a block diagram showing in detail the second control unit 32B in the control device for an electronically controlled valve according to Embodiment 4 of the present invention. Note that the dead zone corrector 330 can be similarly applied to the second control units 32 and 32A of the first to third embodiments described above.

  In a system including a brushless DC motor and an inverter, there may be a dead zone where a predetermined current cannot flow in a region where the phase voltage command is small due to the characteristics of the motor predriver and the influence of a parasitic diode associated with the inverter.

  Specifically, a voltage corresponding to this duty ratio is applied to the winding of the brushless DC motor 2 in a state where the phase voltage command which is the output of the current controllers 323A and 323B is small, that is, the duty ratio of the phase voltage command is small. In other words, a state in which a predetermined current obtained by dividing the voltage corresponding to the duty ratio by the winding resistance cannot flow.

Therefore, based on the output of the saturator 325, as shown in FIG. 17, a table in which the horizontal axis is the phase voltage command which is the output of the saturator 325 and the vertical axis is the post-correction phase voltage, It has. Here, the dead band δ of the voltage is obtained by actually measuring the behavior of the system including the inverter 3 and the brushless DC motor 2. The behavior of the system is a behavior based on the inverter 3 and the brushless DC motor 2. For example, as a drive command that is an input of the PWM processing unit 4, from the phase voltage command 0 to the bus voltage V _bus equivalent. What is necessary is just to measure the winding current when the DC value of the phase voltage command is given.

As shown in FIG. 18, if the measurement data is arranged with the phase voltage command on the horizontal axis and the winding current on the vertical axis, the voltage dead zone δ can be obtained. Here, the dead band δ of the voltage is a positive real number. FIG. 18 shows only the case where the direction of current flow is positive, but the direction of current flow is negative, that is, as a drive command, the voltage command from the voltage command 0 to the bus voltage −V _bus equivalent is shown. From the result of measuring the winding current when a DC value is applied, the dead band of the negative voltage can be obtained.

  However, when the current flow direction is a negative direction, the measurement data is normally point-symmetric with respect to the origin of FIG. 18, and thus a negative sign is added to the dead zone δ of the voltage when the current flow direction is the positive direction. -Δ can be determined. Accordingly, the voltage dead zone can be determined as a fixed value based on the measurement result. Naturally, if there is no dead band of voltage as the measurement result, it may be zero. In this way, a specific relational expression of the table provided in the dead zone corrector 330 is the following expression (13).

In Expression (13), V ^* _{j_comp_c} is a corrected post-phase voltage that is an output of the dead band corrector 330, and g (V ^* _{j_comp} ) and g ′ (V ^* _{j_comp} ) are _used to increase the input V ^* _{j_comp} . On the other hand, the outputs g (V ^* _{j_comp} ) and g ′ (V ^* _{j_comp} ) are monotonically increasing functions. Further, a is a positive real number and is determined as a value commensurate with the measurement result.

Here, g (V ^* _{j_comp} ) is a function that is defined in a region where V ^* _{j_comp} is smaller than the voltage dead zone −δ and satisfies g (−δ) = − δ. Further, g ′ (V ^* _{j_comp} ) is a function that is defined in a region where V ^* _{j_comp} is larger than the voltage dead zone δ and satisfies g ′ (δ) = δ. The subscript j means input / output corresponding to the u phase or the v phase.

That is, the dead zone corrector 330, a dead zone -δ smaller area of the output _V ^* j_comp voltage of saturator 325, a monotonically increasing function to the output _V ^* j_comp the saturator 325 and variable aδ as intercept g (V ^The corrected post-phase voltage V ^* _{j_comp_c} is output in the first line of the equation (13) consisting of ^* _{j_comp} ).

Further, the dead band corrector 330 outputs the corrected phase voltage V ^* _{j_comp_c} as zero in the second row of the equation (13) when the output V ^* _{j_comp of the saturator} 325 is in the _range of the voltage dead band −δ or more and + δ or less. , in the region output _V ^* j_comp is greater than the dead zone δ of the voltage of the saturator 325 consists of monotonically increasing function g in which the output _V ^* j_comp the saturator 325 and variable -aδ as intercept _'(V ^* j_comp) The corrected post-phase voltage V ^* _{j_comp_c} is output in the third row of the equation (13). As a simplest example of the monotonically increasing functions g (V ^* _{j_comp} ) and g ′ (V ^* _{j_comp} ), a relational expression can be given by the following expression (14) where V ^* _{j_comp} .

Next, the effect of the dead zone corrector 330 will be described. First, the case where there is no dead zone corrector 330 will be described. Since the current command and the estimated current output from the current estimators 324A and 324B match within the current control band determined by the current controllers 323A and 323B, the current when the current command and the estimated current completely match The current deviation that is input to the controllers 323A and 323B becomes zero, and the phase voltage command V _j ^* (j = u to v) that is the output of the current controllers 323A and 323B becomes zero. Further, the drive command after the phase voltage command V _j ^* (j = u to v) passes through the voltage corrector 326 and the saturator 325 is also zero.

  Here, it should be noted that when a pseudo current feedback control system is built without a current sensor, the winding current of the brushless DC motor 2 driven by the inverter 3 is unknown. That is, the pseudo current feedback control system without a current sensor is substantially a current feedforward control. Therefore, in the control loop composed of the current controllers 323A and 323B, the voltage corrector 326, the saturator 325, and the current estimators 324A and 324B, how the system composed of the inverter 3 and the brushless DC motor 2 is faithfully reproduced. It is important to be able to generate a voltage command after reproduction.

  Thus, in the fourth embodiment of the present invention, a dead zone corrector 330 is added as a model for further elaborating the behavior of the system including the inverter 3 and the brushless DC motor 2. This is to solve the problem that when the current command is small, the winding current (hereinafter also referred to as the real phase current) cannot flow as the current command because the current deviation becomes zero.

As shown in FIG. 18, the dead zone corrector 330 measures the relationship between the phase voltage command and the winding current in advance, and configures a table based on the measurement result as shown in FIG. output _V ^* j_comp of saturator 325 (j = u~v) generates and outputs a more corrected phase voltage _{^{V * j_comp_c (j = u~v)}} . As a result, even when the output of the saturator 325 is within the range of the voltage dead zone, the current deviation between the current command and the estimated currents output from the current estimators 324A and 324B is not zero, so the current controller 323A 323B always generates a predetermined phase voltage command.

  By doing in this way, since the winding current according to the current command can be passed through the winding of the brushless DC motor 2 even in a region where the current command is small, the current estimation accuracy can be further improved. In particular, the accuracy of the estimation of the winding current with respect to the DC current command is improved, and the load of the valve mechanism 1 is relatively light. For example, the current control in the low current region near the fully closed state can be precisely performed. Without making it, it is possible to obtain a more stable and smooth response.

  In the first embodiment, feedback control by the second position detector 6 in the analog output format, and in the second embodiment, feedback control by the second position detector 6 in the analog output format, and the second position detector 6 are provided. In the case where it is determined that there is a failure, the control mode of feedback control by the first position detector 5 in the pulse output format has been described. Here, as these combined forms, feedback control by the second position detector 6 and motor commutation timing control by the first position detector 5 may be used.

  In this control mode, for example, when 120-degree energization rectangular wave driving is applied as the energization method of the brushless DC motor 2, the commutation pattern is determined by using an analog signal as in the second position detector 6. However, in the sense that the commutation pattern is determined by the pulse output format in which the discrete electrical angle can be directly detected as in the first position detector 5, the commutation pattern can be determined only by pulse edge detection. The logic of the flow pattern can be simplified.

  In all the embodiments of the present invention, the forced driving of the motor by the reference command is naturally possible in both directions, that is, in the valve opening direction and in the closing direction. In a configuration in which a power transmission mechanism such as a gear reduction mechanism is interposed between the motor shaft and the valve shaft, the output of the second position detector 6 is caused by back and forth between the open direction and the close direction due to backlash. The characteristics may differ by the amount of play.

  Therefore, the calibration processing unit 10 uses the first position detector 5 and the second position detector 6 when the motor is forcibly driven over the entire range of the valve shaft operating angle in both directions. Are respectively approximated by polynomials and transplanted to the correction processing unit 20. Then, the polynomial for correction is properly used depending on whether the controlled direction is an open direction or a closed direction.

  By doing in this way, the output error of the 2nd position detector 6 resulting from the play of the power transmission mechanism is also suppressed, and the accuracy of the valve shaft rotation angle output from the correction processing unit 20 can be further improved. .

  Furthermore, all embodiments of the present invention can be applied not only to electronic throttle valves, exhaust gas recirculation valves, waste heat recovery valves, VG (Variable Geometry) actuators, and in-vehicle devices such as electric pumps. Any device can be used as long as it is driven by a brushless DC motor, and application to FA devices such as robots and machine tools is also possible.

Claims (11)

Control of an electronically controlled valve that controls opening and closing of the valve mechanism by controlling a drive source that applies a drive torque against the return torque for a valve mechanism in which a return torque acts in a valve opening direction or a valve closing direction A device,
A brushless DC motor as the drive source;
A first position detector for detecting a motor shaft rotation angle of the brushless DC motor and outputting a discrete signal in a pulse format;
A second position detector for detecting a valve shaft rotation angle of the valve mechanism and outputting a continuous signal as an analog voltage;
Based on a discrete motor shaft rotation angle output from the first position detector and an analog voltage output from the second position detector, driven by a reference command input from the outside, A calibration processing unit for obtaining a relationship between the motor shaft rotation angle and the analog voltage;
A correction processing unit that converts an analog voltage output from the second position detector into a valve shaft rotation angle based on the relationship between the motor shaft rotation angle and the analog voltage obtained by the calibration processing unit;
A control unit that generates a drive command to the brushless DC motor based on a valve shaft rotation angle output from the correction processing unit and a valve shaft position command input from the outside;
A motor shaft rotation angle output from the first position detector, on the basis on the valve shaft rotation angle output from the correction processing unit, detects the leading SL malfunction of the second position detector, said first A failure detection processing unit for determining whether to apply the output of the position detector ;
An electronically controlled valve control apparatus , wherein the processing in the calibration processing unit is periodically executed after the valve is assembled or mounted on the vehicle . The electronically controlled valve according to claim 1, wherein the first position detector has a resolution of 60 deg, 30 deg, or 15 deg in electrical angle, and the second position detector is a non-contact type of magnetic detection. Control device. When the calibration processing unit is driven by a reference command input from the outside, a motor shaft rotation angle output from the first position detector and an analog voltage output from the second position detector based on, Ru obtain an approximate relationship of the motor shaft rotation angle as a polynomial of the voltage
Electronically controlled control valve according to 請 Motomeko 1 or claim 2. The correction processing unit calculates the output from the second position detector, the coefficient of the transmission mechanism system between the motor shaft and the valve shaft, and the motor shaft rotation angle obtained by the calibration processing unit as a voltage polynomial. The electronically controlled valve control device according to any one of claims 1 to 3, wherein a nonlinear characteristic of the second position detector is corrected and output based on a relationship approximated as: The controller is
The position command, the valve shaft rotation angle and the bus voltage of the inverter are input, and a first control unit and a second control unit that output a drive command are configured.
The first controller is
A current command is output as a result of feedback control calculation so that a deviation obtained by a difference calculation between the position command and the valve shaft rotation angle becomes zero,
The second controller is
Based on the current command, the valve shaft rotation angle, and the bus voltage of the inverter, a drive command is output as a result of feedback control calculation without a current sensor so that the actual phase current follows the current command.
The electronically controlled valve control device according to any one of claims 1 to 4. The controller is
A first controller that outputs the current command with the position command and the valve shaft rotation angle as inputs;
A second control unit that outputs the drive command with the current command, the valve shaft rotation angle, and a bus voltage of an inverter that drives the brushless DC motor as inputs, and
The first controller is
With a deviation between the position command and the valve shaft rotation angle as an input, a dead zone for outputting a correction deviation,
A PID controller that receives the corrected deviation and the valve shaft rotation angle as input, and performs a PID calculation by a proportional device, an integrator and a differentiator, and outputs the current command;
A first saturator for limiting the current command;
And functions as a position control system and a speed control system,
The second controller is
A phase corrector that outputs a phase correction amount that corrects the phase delay of the electrical angle delayed by the influence of the control cycle;
A current command distributor that distributes the current command to a phase current command of each phase of the brushless DC motor based on the phase correction amount, the electrical angle, and the angular velocity;
A current controller that outputs a phase voltage command by inputting a current deviation obtained by a difference calculation between the phase current command and the estimated current of each phase of the fed back brushless DC motor;
A voltage corrector that corrects the phase current command in a ratio of a reference voltage and a bus voltage of the inverter with respect to the phase current command output from the current controller;
A second saturator for limiting the output from the voltage corrector within the actual operating voltage range;
A filter for smoothing the electrical angle and obtaining an angular velocity of the brushless DC motor;
An induced voltage estimator that estimates the induced voltage of each phase of the brushless DC motor based on the phase correction amount, the electrical angle, the angular velocity, and the previously measured induced voltage;
Voltage per motor phase for obtaining an estimated current based on a voltage deviation obtained by a difference calculation between an estimated induced voltage output from the induced voltage estimator and a phase voltage command output from the saturator Current estimator that models the relationship between current and current with a first-order lag element;
5. The electronically controlled valve control device according to claim 1, wherein the electronic control valve control device has a function as a current control system. The controller is
Based on the output of the second saturator, it has a dead zone corrector that calculates and outputs a corrected phase voltage,
The dead zone corrector is δ which represents a dead zone of voltage obtained from a relationship between a phase voltage command obtained by measuring a winding current of a brushless DC motor when a DC phase voltage command is given and a winding current. The output of the two saturators is V ^* _{j_comp} , the coefficient is a, the output of the dead zone corrector is V ^* _{j_comp_c} , g (V ^* _{j_comp} ) and g ′ (V ^* _{j_comp} ) are monotonic with V ^* _{j_comp} as a variable Given as an increasing function

The electronically controlled valve control device according to claim 6 . The electronically controlled valve control device according to claim 6 or 7 , wherein a phase correction amount output from the phase corrector is a fixed value or a variable value. The failure detection processing unit
The absolute value of the deviation between the valve shaft rotation angle calculated based on the electrical angle output from the first position detector and the valve shaft rotation angle output from the correction processing unit, and the predetermined threshold value A condition determination unit for determining the relationship;
The electronic control type valve control device according to any one of claims 1 to 8, wherein a determination result in the condition determination unit is output as a drive availability determination flag. The drive control is continued based on the output of the first position detector when it is determined by the failure determination in the failure detection processing unit that the second position detector has failed. Item 10. The electronically controlled valve control device according to any one of Items 9 to 9 . Control of an electronically controlled valve that controls opening and closing of the valve mechanism by controlling a drive source that applies a drive torque against the return torque for a valve mechanism in which a return torque acts in a valve opening direction or a valve closing direction An electronically controlled valve control method executed by an apparatus,
The control device for the electronically controlled valve is:
A brushless DC motor as the drive source;
A first position detector for detecting a motor shaft rotation angle of the brushless DC motor and outputting a discrete signal in a pulse format;
A second position detector that detects a valve shaft rotation angle of the valve mechanism and outputs a continuous signal as an analog voltage;
Based on a discrete motor shaft rotation angle output from the first position detector and an analog voltage output from the second position detector, driven by a reference command input from the outside, A calibration processing step for obtaining a relationship between the motor shaft rotation angle and the analog voltage;
A correction processing step of converting an analog voltage output from the second position detector into a valve shaft rotation angle based on the relationship between the motor shaft rotation angle and the analog voltage obtained in the calibration processing step;
A control step of generating a drive command to the brushless DC motor based on the valve shaft rotation angle converted in the correction processing step and a valve shaft position command input from the outside;
Wherein the motor shaft rotation angle outputted from the first position detector, the correction processing based on the converted and the valve shaft rotation angle in step detects the leading SL malfunction of the second position detector, said first A failure detection processing step for determining whether to apply the output of the position detector ;
An electronically controlled valve control method , wherein the calibration processing step is periodically executed after the valve is assembled or mounted on the vehicle .

Images