Drive device and method for rotating electric machine, and railway vehicle

Abstract

To detect slippage or skidding early while stabilizing a control system even during slippage or skidding in a railway vehicle equipped with a rotating electric machine.SOLUTION: A drive device for a rotating electric machine includes a voltage output device that outputs a drive voltage to a rotating electric machine, a current detector that detects the current flowing through the rotating electric machine, and a control device that estimates and calculates the angular velocity of the rotor of the rotating electric machine. The control device calculates a compensation value that corrects a state quantity that serves as an input for estimating and calculating the angular velocity during acceleration and deceleration of the rotating electric machine in order to suppress to approximately zero a steady-state error that occurs in the state quantity, which is an input for estimating and calculating the angular velocity, and detects slippage or skidding of the drive wheels driven by the rotating electric machine when the compensation value exceeds a predetermined value.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a drive device and a drive method for a rotating electric machine, and a railway vehicle equipped with the rotating electric machine. [Background technology]

[0002] In railway vehicles, the wheels, which are drive wheels, are rotated by the torque of a rotating electric machine, and the vehicle is accelerated by the tangential force generated in the wheels as a reaction force that the wheel tread receives from the rail. This tangential force varies depending on the tangential force coefficient μ, which indicates the state of adhesion between the wheel and the rail. If the wheel torque exceeds the tangential force, the force that accelerates the vehicle remains small, while the force that rotates the wheel increases, resulting in wheel spin or skid (hereinafter referred to as "wheel skid"). In particular, in rainy or snowy weather, the adhesion coefficient drops significantly, making wheel spin and skid more likely to occur.

[0003] If this slippage occurs more frequently, it can cause problems such as a decline in the vehicle's acceleration and deceleration performance, and large fluctuations in tangential force can cause vehicle body vibration, reducing ride comfort. For this reason, a slippage readhesion control system is widely used, which quickly detects slippage that occurs between the wheels and rails and reduces the torque of the rotating electric machine to allow the wheels to readhere to the rails.

[0004] When a wheel skid occurs, the angular velocity of the rotor of the rotating electrical machine changes suddenly, which can easily destabilize sensorless control. Therefore, there is a need for technology that can detect wheel skid early while stabilizing the control system. Various improved technologies have been proposed to address this issue.

[0005] For example, Patent Document 1 discloses a technology that enables detection of slippage or skidding of the drive shaft of an induction motor without using a speed sensor and even when the relative speed of each induction motor is small. This technology calculates an estimated acceleration value αi from the time change (differential value) of the frequency command value of the control device, and determines that slippage or skidding has occurred if it is determined that the value exceeds a predetermined threshold.

[0006] Patent Document 2 discloses a technique for calculating an acceleration estimation value by adding a term proportional to a position error estimation value of a permanent magnet synchronous motor and a derivative of the term proportional to the position error estimation value, in order to reduce delays in acceleration estimation caused by γ- and δ-axis currents and vibrations of the mechanical system, and thus delays in slippage control.

[0007] Patent Document 3 discloses a technology that varies the sensorless control gain according to the rotation phase angle error in order to achieve two goals at the same time: to prevent control instability caused by a control gain that is too large during steady state in response to a sudden change in speed, such as during idling, and to improve tolerance to step-out during transient states. [Prior art documents] [Patent documents]

[0008] [Patent Document 1] Japanese Patent Application Laid-Open No. 2004-80936 [Patent Document 2] Patent No. 5515885 [Patent Document 3] Patent No. 5908205 Summary of the Invention [Problem to be solved by the invention]

[0009] The inventors of the present application have conducted extensive research with the aim of further increasing the speed of a means for detecting wheel slip, and have come to the following findings. The technology described in Patent Document 1 calculates an estimated acceleration value αi from the time change (differential value) of a frequency command value of a control device that drives multiple induction motors in parallel without a speed sensor. However, as described in paragraph

[0027] of Patent Document 1, in order to remove noise, the estimated acceleration value αi is calculated by performing a low-pass filter (LPF) calculation with a passband of 3 Hz or less on the calculated output frequency, and then performing a first-order differentiation. This low-pass filter calculation poses a problem in that it delays the detection of skid.

[0010] The technology described in Patent Document 2 proposes a method for reducing the delay in detecting wheel skid by differentiating a position error estimate of a permanent magnet synchronous motor to calculate an acceleration estimate. This method, like the commonly used method of calculating an acceleration estimate from a frequency derivative, calculates acceleration from a differentiation process, and requires processing (such as a low-pass filter) to smooth vibrations in the derivative. Therefore, as described in paragraphs

[0028] to

[0031] of Patent Document 2, low-pass filters with time constants Ta and Tb are configured as vibration reduction countermeasures units 54 and 55, and the influence of these low-pass filters poses a problem in that the delay in detecting wheel skid cannot be sufficiently improved.

[0011] The technology described in Patent Document 3 proposes a method of changing the sensorless control gain more as the absolute value of the estimated phase shift increases in order to stabilize sensorless control when the speed changes suddenly. However, even if it is possible to operate with a stable gain during steady state operation, there is still the possibility of instability when the sensorless control gain is increased when the phase shift is large. Ideally, a control system would be one that can stably track position sensorless control even when the speed changes with high acceleration, such as during idling.

[0012] Based on the above, when considering early detection of wheel slip while stabilizing the control system, if differentiation processing is used on frequency estimates or position error estimates as in Patent Documents 1 and 2, the differential values ​​will become oscillatory, making it necessary to also provide low-pass filter processing to sufficiently attenuate the vibration components contained in the differential values, making it difficult to improve the delay in wheel slip detection. Furthermore, a method of transiently increasing the sensorless control gain as in Patent Document 3 may result in unstable control.

[0013] Therefore, the present invention has been made in consideration of the above points, and aims to provide a technology that achieves faster skid detection while stabilizing the control system by adding a term to the speed estimate of sensorless control to compensate for the deviation due to the acceleration component so that the sensorless control does not become unstable during skid, and further by detecting skid based on the magnitude of this compensation amount without using differential processing. [Means for solving the problem]

[0014] In order to solve this problem, one representative synchronous motor driving device according to the present invention comprises a voltage output device that outputs a driving voltage to the synchronous motor, and a control device that outputs an estimated value of the rotor angular velocity of the synchronous motor, and the control device uses, as an input quantity, phase deviation information that indicates the difference between the estimated value of the phase of the synchronous motor and the rotor phase of the synchronous motor, and narrows down the torque command value for the synchronous motor based on the phase deviation information. [Effects of the Invention]

[0015] According to the present invention, for example, it is possible to stabilize the control system when the drive wheels driven by the rotating electric motor mounted on a railway vehicle skid, while realizing faster detection of skid, and to quickly reduce the torque command value when skid is detected to bring the skid to a halt. Problems, configurations, and effects other than those described above will become apparent from the following description of the preferred embodiments. [Brief explanation of the drawings]

[0016] [Figure 1] 1 is a diagram showing an example of functional blocks of a driving device for a rotating electric machine according to a first embodiment. [Figure 2] FIG. 1 is a diagram showing a schematic configuration of a bogie for a railway vehicle. [Figure 3] FIG. 1 is a diagram showing an example of functional blocks of equations of motion of a vehicle and a driving wheel set for a railway vehicle. [Figure 4] FIG. 10 is a diagram showing the relationship between the slip velocity between the wheel and the rail and the tangential force coefficient. [Figure 5]FIG. 1 is a diagram showing wheel slip during power running and regeneration of a railway vehicle. [Figure 6] FIG. 10 is a diagram illustrating a phase deviation that occurs during acceleration and deceleration of a railway vehicle. [Figure 7] FIG. 10 is a diagram showing current vectors when a phase deviation occurs between the rotor phase and the phase estimate value for control. [Figure 8] FIG. 2 is a diagram illustrating an example of functional blocks of a phase synchronization control unit according to the first embodiment. [Figure 9] FIG. 10 is a diagram illustrating an example of a functional block of a constant speed phase deviation convergence unit. [Figure 10] FIG. 10 is a diagram illustrating an example of a functional block of an acceleration / deceleration phase deviation convergence unit. [Figure 11] 10A and 10B are diagrams illustrating an example of the effect of adding compensation by an acceleration / deceleration phase deviation convergence unit. [Figure 12] 10 is a diagram showing a method for detecting a slip based on the output of a phase deviation convergence unit during acceleration / deceleration when a slip occurs. FIG. [Figure 13] FIG. 2 is a diagram illustrating an example of a functional block of a skid detection / determination unit. [Figure 14] FIG. 1 is a diagram showing a comparison between a conventional configuration for detecting wheel slip and a configuration of the present invention. [Figure 15] FIG. 10 is a diagram illustrating an example of functional blocks of a phase synchronization control unit according to a modified example of the first embodiment. [Figure 16] FIG. 10 is a diagram showing an example of functional blocks of a constant speed phase deviation convergence unit which is a modified example of the first embodiment. [Figure 17] FIG. 10 is a diagram illustrating an example of functional blocks of an acceleration / deceleration phase deviation convergence unit that is a modified example of the first embodiment. [Figure 18] 3 is a diagram illustrating an example of functional blocks of a frequency estimation control unit when the first embodiment is applied to a driving device for an induction machine. FIG. [Figure 19] FIG. 4 is a diagram illustrating an example of a functional block of a q-axis current deviation convergence unit at a constant speed in a frequency estimation control unit. [Figure 20] FIG. 2 is a diagram illustrating an example of a functional block of a q-axis current deviation convergence unit during acceleration / deceleration in a frequency estimation control unit. [Figure 21]FIG. 10 is a diagram showing an example of functional blocks of a driving device for a rotating electric machine according to a second embodiment. [Figure 22] FIG. 10 is a diagram showing an example of a functional block of a skid / slide detection / determination unit according to a second embodiment. [Figure 23] FIG. 10 is a diagram showing an example of functional blocks of a driving device for a rotating electric machine according to a third embodiment. [Figure 24] FIG. 11 is a diagram showing an example of a functional block of a skid detection / determination unit according to a third embodiment. [Figure 25] FIG. 11 is a diagram illustrating an example of functional blocks of an acceleration / deceleration phase deviation convergence unit according to a third embodiment. [Figure 26] 10A and 10B are diagrams illustrating a comparison of the effects of the presence or absence of upper and lower limiters provided in the acceleration / deceleration phase deviation convergence unit. DETAILED DESCRIPTION OF THE INVENTION

[0017] Hereinafter, as modes for carrying out the present invention, first to third embodiments will be described in detail with reference to the drawings. In each embodiment, the same reference numerals are used to denote the same components or components having similar functions. In addition, in a later embodiment, the description of the same or similar components as in a previous embodiment may be omitted. [Embodiment 1]

[0018] FIG. 1 is a diagram showing an example of functional blocks of a driving device for a rotating electric machine according to a first embodiment of the present invention. FIG. 1 shows only the minimum functional blocks required for the first embodiment, and the power converter composed of driving transistors such as IGBTs (Insulated Gate Bipolar Transistors) and power devices such as diodes, and the control configuration for this power converter are shown in a block diagram as a voltage output device 3, and detailed illustrations are omitted.

[0019] Furthermore, although the rotating electric machine in FIG. 1 is configured to use a synchronous machine 4, it may be an induction machine or the like other than a synchronous machine, as shown in FIGS. 18 to 20 described later.

[0020] As shown in FIG. 1, in the drive device 1, the voltage output device 3 applies a three-phase AC voltage to a synchronous machine 4 via a drive circuit and a main circuit (not shown, but included in the voltage output device 3) based on a switching command from the control device 2.

[0021] A drive current flows through the synchronous machine 4 when a three-phase AC voltage is applied from the voltage output device 3, and a rotational torque is generated.

[0022] The current detector 5 is composed of a Hall CT (Current Transformer) or the like, and detects the three-phase current i of U phase, V phase, and W phase flowing through the synchronous machine 4. u , i v and i w However, it is not necessary to detect the currents of all three phases by the current detector 5, and it is also possible to configure the system so that any two of the three phases are detected and the remaining phase is found by calculation, assuming that the three-phase currents are in a balanced state.

[0023] A control program for driving and controlling a synchronous machine 4 connected as a load is installed in the control device 2. The functions and aspects of each of the components included in the control device 2 will be described below.

[0024] The torque command calculation unit 11 calculates a torque command value τ m * Output.

[0025] The current command calculation unit 10 calculates the torque command value τ m * The current command value i on the dq axis to obtain a specified torque is d * , i q * Output.

[0026] The current detection coordinate conversion unit 8 converts the three-phase current i of the synchronous machine 4 detected by the current detector 5 into u , i v and i wThe d-axis estimated phase θ recognized by the control device 2 dc Convert it into dq coordinates of the rotating coordinate system using df and i qf ) to the current control unit 9.

[0027] The current control unit 9 controls the dq voltage command value v by PI (Proportional-Integral) control or the like so that the current deviation between the dq axis current detection value output by the current detection coordinate conversion unit 8 and the dq axis current command value output by the current command calculation unit 10 becomes zero. d * , v q * Generate and output.

[0028] The voltage command coordinate conversion unit 12 converts the dq-axis voltage command value output by the current control unit 9 into the d-axis estimated phase θ dc and the three-phase AC voltage command value v u * , v v * and v w * Output.

[0029] The PWM control unit 7 converts the three-phase AC voltage command value v output from the voltage command coordinate conversion unit 12 into u * , v v * and v w * Based on this, a switching command for PWM voltage (Pulse Width Modulation) is output to the voltage output device 3.

[0030] The phase synchronization control unit 14 receives the phase deviation information Δθ c Based on this phase deviation information Δθ c The angular velocity estimate ω1 is set to zero. ^ At the same time, the acceleration / deceleration phase deviation compensation amount Δθ is output, which will be described in detail later. α Output.

[0031] In addition, the phase deviation information Δθ c is the phase estimate θ dcand the rotor phase θ of synchronous machine 4 d Here, the phase deviation information Δθ c is the estimated value of the phase deviation by sensorless control, Δθ est Although not shown, the phase detection value θ r Using the information of dc The difference between these is calculated to obtain the phase deviation information Δθ c Either of the above configurations may be used.

[0032] In addition, the phase deviation estimate Δθ est As a method for obtaining this, for example, a method for estimating based on the high frequency current detection value when a high frequency voltage is superimposed in the low speed range, or a method for estimating using the induced voltage of the rotating electric machine in the high speed range is used.

[0033] The skid detection / determination unit 15 detects the acceleration / deceleration phase deviation compensation amount Δθ output from the phase synchronization control unit 14. α The torque command calculation unit 11 detects and judges whether a skid has occurred based on the detected skid and outputs a skid detection signal to the torque command calculation unit 11. When the torque command calculation unit 11 receives the skid and skid detection signal, it immediately calculates the torque command value τ m * By narrowing the gap, the wheel slippage that occurs between the drive wheels and the rail is reduced, allowing the wheels to re-adhere to the rail.

[0034] The phase calculation unit 13 calculates the angular velocity estimated value ω1 output from the phase synchronization control unit 14. ^ is integrated to obtain the phase estimate θ dc Output.

[0035] The following describes slippage and readhesion control in railway vehicles. FIG. 2 is a diagram showing a schematic configuration of a bogie for a railway vehicle. A synchronous machine 4 is mounted on the bogie 31, and a rotor shaft 30 of the synchronous machine 4 transmits power to a pinion gear 32 via a coupling 34. A reduction gear consisting of the pinion gear 32 and a gear 33 rotates an axle 35, which in turn rotates the wheels 27 connected to the axle 35. The wheels 27 transmit force to the bogie 31 by a tangential force generated in the wheels 27 as a reaction force received by the wheel treads from the rails 36, thereby accelerating the vehicle.

[0036] FIG. 3 is a diagram showing functional blocks of the equations of motion of a vehicle and one driving wheelset. The equation of motion for the car body and motor is shown in a block diagram, taking into account the adhesion between the drive wheels (wheels) and the rails due to the tangential force coefficient μ caused by the motor torque. The force that rotates the wheels is determined by the difference between the wheel torque and the tangential force torque, and the force that accelerates the vehicle is determined by the difference between the tangential force and running resistance, and these change depending on the tangential force coefficient μ.

[0037] FIG. 4 is a diagram showing the relationship between the slip velocity between the wheel and the rail and the tangential force coefficient μ. The tangential force coefficient μ changes depending on the slip speed, which is the difference between the wheel peripheral speed and the vehicle speed. Furthermore, in rainy weather, the tangential force coefficient μ decreases compared to sunny weather, making wheel slippage more likely to occur. It is also known that the tangential force coefficient μ changes not only in rainy weather, but also due to oil, fallen leaves, and dust on the rail surface, as well as vehicle speed.

[0038] FIG. 5 is a diagram showing wheel slip during power running and regenerative braking of a railway vehicle. As shown in Figure 4, when the tangential force coefficient μ decreases due to rain or other factors, the force accelerating the vehicle in Figure 3 remains small, while only the force rotating the wheels increases, causing the slip speed to increase. If this occurs and the inflection point of the tangential force coefficient μ shown in Figure 4 is exceeded, wheel slip occurs. Because the tangential force coefficient μ has a negative slope with respect to the slip speed, once wheel slip occurs, the wheel slip will continue to increase in an expanding direction. The above applies when the vehicle is powered, but when regenerating, wheel slip also occurs, as shown in Figure 5.

[0039] If this wheel slippage expands, it can cause problems such as a decline in the vehicle's acceleration and deceleration performance, and large fluctuations in the tangential force can cause large fluctuations in the force that accelerates the vehicle, as shown in Figure 3, which can cause vibrations in the vehicle body and reduce ride comfort. For this reason, it is necessary to quickly detect wheel slippage that occurs between the wheels and rails and reduce the torque of the rotating electric machine to allow the wheels to re-adhere to the rails.

[0040] The phase deviation that occurs during acceleration and deceleration of a railway vehicle will be explained below. FIG. 6 is a diagram showing a phase deviation that occurs during acceleration or deceleration of a vehicle. First, the phase deviation Δθ that occurs during acceleration / deceleration will be described. d and the control phase estimate θ dc The deviation from is defined by the following equation (Equation 1): where s is the Laplace operator and ω r is the rotor angular speed of the synchronous machine 4.

number

[0041] Next, an example will be described in which the phase synchronization control unit 14 is configured using a general PI control. The proportional gain and integral gain in the phase synchronization control unit 14 are respectively K P , K. I Then, the phase deviation information Δθ c Based on this, the angular velocity estimate ω1 ^ can be calculated using the following equation (Equation 2).

number

[0042] Rotor angular speed ω of synchronous machine 4 r is the motor torque τ m and the moment of inertia J can be calculated using the following equation (Equation 3). mis the number of pole pairs of the synchronous machine 4. Also, to simplify the explanation, the moment of inertia J is treated as an equivalent moment of inertia that combines the vehicle mass and adhesion due to the tangential force coefficient μ when viewed from the motor shaft shown in Fig. 3. When there is adhesion between the wheel and rail, the moment of inertia J is large, and when skidding occurs between the wheel and rail, the moment of inertia J is treated as appearing to be smaller when viewed from the motor shaft.

number

[0043] From (Equation 1) to (Equation 3), the phase deviation Δθ can be expressed by the following equation (Equation 4).

number

[0044] Phase deviation information Δθ recognized by the control device 2 c Assuming that coincides with the phase deviation Δθ, the following equation (Equation 5) is obtained.

number

[0045] The motor torque τ m When accelerating and decelerating the motor with the output of , the phase deviation Δθ is steadily expressed as follows (Equation 6).

number

[0046] That is, when the phase synchronization control unit 14 is configured with a general PI control, as explained using (Equation 1) to (Equation 6), the torque τ m , moment of inertia J and integral gain K I In accordance with this, the estimated phase deviation Δθ(=-(τ m P m ) / (J·K I )) will remain.

[0047] As is clear from (Equation 6), the integral gain K I If is sufficiently large and acceleration / deceleration is gradual, the phase deviation Δθ is also small. However, under conditions such as skidding, the equivalent moment of inertia J becomes small, and unless the torque is reduced, a large phase deviation Δθ occurs.

[0048] Figure 7 shows the rotor phase (θ d ) and the control phase estimate (θ dc 10 is a diagram showing current vectors when a phase deviation Δθ occurs between the first and second inverters. The reference phase of the rotor of the synchronous machine 4 is the d-axis, and the reference phase of the rotor estimated by the control device 2 is the dc-axis. When a phase deviation Δθ occurs, the current command value i d * , i q * The current is controlled so that the actual current is i d , i q and the torque command value τ m * The actual torque output by the motor is τ m has a large error.

[0049] As such, it is desirable not to generate a phase deviation Δθ according to acceleration, not only in situations where speed changes with high acceleration such as during a skid, but also during normal acceleration and deceleration in an adhesive state, since this not only leads to destabilization of control but also to a decrease in torque accuracy.

[0050] Below, we will explain the key feature of this invention, which is to stabilize the control system even when speed suddenly changes due to skid, while also speeding up skid detection.This involves providing a term in sensorless control that compensates for deviations due to acceleration components, and using the output value of this term that compensates for acceleration components to detect skid without using differential processing.

[0051] FIG. 8 is a diagram illustrating an example of functional blocks of the phase synchronization control unit 14 according to the first embodiment. The phase synchronization control unit 14 comprises a constant speed phase deviation convergence unit 20 that can converge the phase deviation to zero at constant speed, and an acceleration / deceleration phase deviation convergence unit 21 that can converge the phase deviation during acceleration / deceleration.

[0052] The acceleration / deceleration phase deviation convergence unit 21 calculates the phase deviation information Δθ c and the phase deviation target value (before compensation) Δθ c0 is used as input, and the acceleration / deceleration phase deviation compensation amount Δθ α Output.

[0053] The constant speed phase deviation convergence unit 20 calculates the phase deviation amount (before compensation) Δθ c0 and the acceleration / deceleration phase deviation compensation amount Δθ α The phase deviation (after compensation) Δθ is the sum of c ’ is used as input, and the angular velocity estimate ω1 ^ Output.

[0054] 9 is a diagram showing an example of a functional block of the constant speed phase deviation convergence unit 20. P and integral gain K I It is a PI control using

[0055] 10 is a diagram showing an example of a functional block of the acceleration / deceleration phase deviation convergence unit 21. II The integration is performed by

[0056] In order to prevent the steady-state deviation of the phase during acceleration / deceleration shown in (Equation 6) from occurring, the phase deviation Δθ c From the angular velocity estimate ω1 ^ The order of the transfer function that calculates the above may be configured to be second order or higher as shown in the following equation (Equation 7).

number

[0057] Similarly to (Equation 5), rearranging the equation for the phase deviation Δθ′ in the first embodiment gives the following equation (Equation 8).

number

[0058] When the vehicle is accelerating or decelerating, the phase deviation Δθ′ that is steadily generated can be converged to zero as shown in the following equation (Equation 9).

number

[0059] At this time, in the acceleration / deceleration phase deviation convergence unit 21, K II / s operates to output a compensation amount so that the steady-state deviation of the phase deviation Δθ in (Equation 6) becomes zero, so the output (acceleration / deceleration phase deviation compensation amount) Δθ α is related by the following equation (Equation 10).

number

[0060] FIG. 11 is a diagram showing an example of the effect of adding compensation by the acceleration / deceleration phase deviation convergence unit 21. The acceleration / deceleration phase deviation compensation amount Δθ, which is the output of the acceleration / deceleration phase deviation convergence unit 21 α This allows the steady-state phase deviation during acceleration / deceleration to converge to zero, as shown in Fig. 11. Even if wheel skid occurs and the acceleration / deceleration rate increases, the phase deviation can be prevented from increasing, preventing control instability and a decrease in torque accuracy.

[0061] The acceleration / deceleration phase deviation compensation amount Δθ α is the torque τ m , integral gain K I is known, it will change depending on the equivalent moment of inertia J, which includes the vehicle mass and the tangential force coefficient μ, as viewed from the motor shaft.

[0062] The inventors have determined that the acceleration / deceleration phase deviation compensation amount Δθ αBy focusing on this, we discovered that it is possible to obtain information equivalent to acceleration from phase deviation information without using differential processing, and devised a method to detect wheel slip using this information.

[0063] FIG. 12 is a diagram showing a method for detecting a slip based on the output of the acceleration / deceleration phase deviation convergence unit when a slip occurs. In comparison with the phase deviation Δθ in a normal adhesion state, the moment of inertia J appears to be suddenly small during slippage, so the Δθ of the output value of the phase deviation convergence unit 21 during acceleration / deceleration expressed by (Equation 10) α also increases.

[0064] FIG. 13 is a diagram showing an example of a functional block of the skid detection / determination unit 15. The acceleration / deceleration phase deviation compensation amount Δθ output by the phase synchronization control unit 14 α However, if the acceleration component of the phase deviation exceeds the judgment value (slip judgment threshold or slide judgment threshold) expected in a normal adhesion state, it is assumed that slippage has occurred, and a slippage detection signal is output to the torque command calculation unit 11.

[0065] Here, the slippage determination threshold or skid determination threshold is set to be larger than the compensation amount during acceleration / deceleration in an adhesive state, and to a degree that does not cause erroneous detection according to the vibration component contained in the phase deviation information.

[0066] FIG. 14 is a diagram showing a comparison between a conventional configuration for detecting wheel slip and a configuration of the present invention. The differences in configuration and effect between the prior art and the present invention will be described below with reference to FIG.

[0067] <Conventional configuration> In the conventional method described in Patent Document 1, a configuration is generally used in which an integral process is performed using PI control or the like in the conventional phase synchronization control unit 16 to calculate a frequency estimate, and then this frequency estimate is differentiated using a differential process to calculate an acceleration estimate (dfr / dt).

[0068] In this type of method, since differential processing (s) is performed after integral processing (1 / s) is performed using PI control, it is clear that there is room for improvement in terms of the calculation processing itself, and it can be seen that acceleration components can be extracted without using differential processing.

[0069] Furthermore, the problem of delayed detection of skid / slide is unavoidable because the conventional phase synchronization control unit 16 has a cutoff frequency. Furthermore, when differential processing is used to calculate acceleration, it becomes necessary to implement low-pass filtering as in Patent Document 1 and Patent Document 2 as a noise countermeasure, and it is clear that this will delay the detection of skid / slide.

[0070] <Configuration and effects of the present invention> On the other hand, in the present invention, the phase deviation estimated value Δθ before calculating the frequency estimated value is est This configuration detects wheel slip by extracting the acceleration component of the wheel. Therefore, wheel slip can be detected earlier than with conventional configurations.

[0071] Even when a skid occurs and the speed changes at a high acceleration, the phase deviation Δθ due to the acceleration is calculated by the output Δθ of the acceleration / deceleration phase deviation convergence unit 21. α This not only speeds up the detection of wheel slip, but also stabilizes sensorless control when wheel slip occurs.

[0072] Furthermore, even in an adhesion state where no wheel slip occurs, the phase deviation Δθ shown in (Equation 6) is suppressed, so torque errors due to phase deviations such as those described with reference to FIG. 7 are not generated, and the effects of stabilizing sensorless control and improving torque accuracy are also obtained.

[0073] On the other hand, if the sensorless control gain is increased as in Patent Document 3, the phase deviation Δθ can be reduced, as can be seen from Equation 6. However, there is a limit to how much the gain can be increased due to the calculation cycle of the control device, the delay in voltage output, and the delay in current detection, and simply increasing the gain may result in unstable control.

[0074] Furthermore, it is possible to eliminate the need for differential processing (acceleration calculation) used only for detecting wheel slippage, as in the conventional configuration shown in Figure 14 and as described in Patent Documents 1 and 2, and this also has the effect of reducing the calculation load on the microcomputer installed in the control device 2. The above is the difference in configuration and effect between the prior art and the present invention.

[0075] Next, a modification of the first embodiment will be described. FIG. 15 is a diagram showing an example of a functional block of a phase synchronization control unit 14a which is a modified example of embodiment 1, FIG. 16 is a diagram showing an example of a functional block of a constant speed phase deviation convergence unit 20a in the phase synchronization control unit 14a, and FIG. 17 is a diagram showing an example of a functional block of an acceleration / deceleration phase deviation convergence unit 21a in the phase synchronization control unit 14a.

[0076] The phase synchronization control section 14a differs from the phase synchronization control section 14 shown in FIG. 8 in that the phase deviation convergence section 20a at constant speed and the phase deviation convergence section 21a at acceleration / deceleration provide a phase deviation amount (before compensation) Δθ c0 At this time, unlike FIG. 10, the acceleration / deceleration phase deviation convergence unit 21a provides a square term of 1 / s as shown in FIG. 17 to reduce the phase deviation Δθ c From the angular velocity estimate ω1 ^ The order of the denominator of the transfer function that calculates the above formula is set to second order. This has the effect of suppressing the phase deviation Δθ during acceleration / deceleration.

[0077] Since the same effect can be obtained by setting the order of the denominator of the transfer function to a second-order or higher order, the order can be any number as long as it includes a second-order or higher term, and the phase synchronization control unit 14 is not necessarily limited to the configurations shown in Figures 8 to 10 or Figures 15 to 17.

[0078] In addition, the phase detection value θ by a resolver, etc. r Using the information of dc The difference between these is calculated to obtain the phase deviation information Δθ cEven if the phase synchronization control unit 14 is configured as above, steady-state deviations similarly remain during acceleration and deceleration of the vehicle, so this can also be applied to configurations with resolver control or speed sensor control.

[0079] In the first embodiment described above, the synchronous machine 4 is used as an example of a rotating electric machine, but the present invention is not limited to a synchronous machine and can also be applied to other rotating electric machines such as an induction machine.

[0080] Figure 18 is a diagram showing an example of a functional block of the frequency estimation control unit 40 when embodiment 1 is applied to a driving device for an induction machine, Figure 19 is a diagram showing an example of a functional block of the q-axis current deviation convergence unit 41 at constant speed in the frequency estimation control unit 40, and Figure 20 is a diagram showing an example of a functional block of the q-axis current deviation convergence unit 42 at acceleration / deceleration in the frequency estimation control unit 40.

[0081] For example, even in a configuration in which the angular velocity of the rotor is estimated from the deviation between the q-axis current command value and the q-axis current detection value, the q-axis current deviation convergence unit 42 during acceleration / deceleration is configured to compensate for the q-axis current deviation due to acceleration. That is, the acceleration / deceleration q-axis current deviation compensation amount Δi q_α If the wheel slip is detected based on the above, the speed estimation control can be stabilized and wheel slip can be detected early.

[0082] In the configurations shown in Figures 18 to 20, as with (Equation 6), when no compensation for acceleration is added, the deviation amount of the following equation (Equation 11) remains, and the compensation amount becomes the following equation (Equation 12), as with (Equation 10).

number

number

[0083] In other words, regardless of the type of rotating electric machine, the effects of the present invention can be obtained by outputting a compensation value that corrects a state quantity (estimated rotor phase in the case of a synchronous machine, or current flowing through the induction machine in the case of an induction machine) used as input for calculating an angular velocity estimation value during acceleration or deceleration of the rotating electric machine in order to suppress the deviation amount that occurs in the state quantity to approximately zero, and by detecting slippage when the compensation value exceeds a predetermined value.

[0084] As described above, the driving device for the rotating electric machine according to the first embodiment includes the voltage output device 3 that outputs a voltage to the synchronous machine 4, and the torque command value τ m * The control device 2 controls the output voltage of the voltage output device 3 based on the angular velocity estimated value ω1 from the rotor phase deviation information. ^ The phase synchronization control unit 14 has an acceleration / deceleration phase deviation convergence unit 21 that suppresses the steady-state deviation of the estimated phase during acceleration / deceleration, and a slip / slide detection / determination unit 15 that detects a slip / slide state of the wheels driven by the synchronous machine 4 based on the output of this acceleration / deceleration phase deviation convergence unit 21. This makes it possible to stabilize the control system even during a slip / slide and to achieve high-speed slip / slide detection. Also, in the case of an induction machine, a similar effect can be achieved by replacing the phase deviation with the q-axis current deviation. [Embodiment 2]

[0085] The second embodiment of the present invention differs from the first embodiment in that the determination threshold for detecting a skid in the skid detection / determination unit 15 is variable depending on the operating state of the vehicle and the surrounding environment.

[0086] This makes it possible to set an appropriate skid / slide judgment threshold even when the operation of the rotating electric machine or the surrounding environment (torque, occupancy rate, weather, train formation status, and running resistance) changes, and makes it possible to detect skid / slide even faster than in embodiment 1 while preventing false detection of skid / slide.

[0087] FIG. 21 is a diagram illustrating an example of functional blocks of a driving device for a rotating electric machine according to the second embodiment. In the second embodiment, the slippage detection and determination unit 15 detects the torque command value τm * , occupancy information, weather information, position information within the train, and estimated angular velocity value ω1 ^ Enter the following to make the skid / slide detection threshold variable.

[0088] FIG. 22 is a diagram showing an example of functional blocks of the skid detection / determination unit 15 according to the second embodiment.

[0089] (1) Torque (torque command value τ m * ) As can be seen from (Equation 10), for example, torque τ m If is smaller, the acceleration / deceleration also becomes lower, and the compensation amount Δθ α In other words, the torque command value τ m * When the threshold value is increased, the torque command value τ m * By adjusting the judgment threshold to be lower when the torque becomes smaller, it is possible to set an appropriate judgment threshold for the torque state, thereby preventing false detection of skid and achieving even faster detection.

[0090] (2) Regarding occupancy rate information As can be seen from Figure 3, the tangential force (= μ·W·g) is proportional to the weight W applied per axle. The weight W changes depending on the passenger occupancy rate, so for example, if the occupancy rate decreases, skid sliding becomes more likely, so the judgment threshold is lowered, and if the occupancy rate increases, skid sliding becomes less likely, so the judgment threshold is raised. In this way, an appropriate judgment threshold can be set according to the occupancy rate, preventing false detection of skid sliding and achieving even faster detection.

[0091] In FIG. 22, the torque command value τ m * is the integral gain K I The skid / slide determination threshold is the value obtained by dividing the load factor information by the moment of inertia J. Here, the equivalent moment of inertia calculation unit 50 calculates the equivalent moment of inertia J from the load factor information.

[0092] (3) Regarding weather information As can be seen from Figure 4, in rainy weather the tangential force coefficient μ is smaller than in sunny weather, and the tangential force (= μ·W·g) is also smaller. Therefore, for example, when information about rain or snowfall is detected from an external device, such as wiper operation or weather information, the judgment threshold is lowered because skidding becomes more likely. By doing this, even faster detection can be achieved even when the road surface (rail surface) becomes slippery due to changes in weather such as rain or snowfall.

[0093] In FIG. 22, a weather information correction gain calculation unit 51 receives weather information as an input, calculates a weather information correction gain, and outputs it as one of the correction gains for the skid / slide determination threshold.

[0094] (4) Regarding location information within the train 4, similar to the principle that the tangential force coefficient μ changes depending on the weather, i.e., the condition of the road surface (rail surface), when the axle 35 to which the rotating electric machine is connected is located at the front in the direction of travel, raindrops, dust, etc. on the rail surface are not cleared away, and skid sliding tends to occur easily. On the other hand, when the axle 35 to which the rotating electric machine is connected is located at the rear in the direction of travel, skid sliding tends to occur relatively less, even in rainy weather.

[0095] Therefore, if it is possible to distinguish between the direction of travel of the train itself and the car number on which the axle 35 to which the rotating electric machine is connected is mounted, it is possible to adjust the threshold value for determining whether a wheel slippage has occurred appropriately. For example, if the train is located closer to the front car in the direction of travel, the threshold value is lowered because wheel slippage is more likely to occur, and if the train is located closer to the rear car in the direction of travel, the threshold value is higher because wheel slippage is less likely to occur. In this way, it is possible to set an appropriate threshold value according to the position of the car with the axle within the train, thereby preventing false detection of wheel slippage and achieving even faster detection.

[0096] In FIG. 22, an intra-formation position information correction gain calculation unit 52 receives intra-formation position information as input and calculates an intra-formation position information correction gain, which is output as one of the correction gains for the slip / slide detection threshold.

[0097] (5) Angular velocity estimate ω1 ^ Regarding The tangential force coefficient μ is dependent on the vehicle speed (see the tangential force coefficient table shown in Figure 3), and the higher the vehicle speed, the smaller the tangential force coefficient μ. In addition, the running resistance also increases as the speed increases, making wheel slippage more likely to occur. Therefore, for example, when the angular velocity estimated value ω1 ^ When the value of the angular velocity estimation value ω1 becomes high, the wheel becomes prone to skidding, so the judgment threshold is lowered. ^ When the angular velocity estimate ω1 is low, the determination threshold is set high because it becomes difficult for the vehicle to skid. ^ Instead of the above, the frequency of the rotating electric machine or the vehicle speed may be directly used. In this way, an appropriate determination threshold can be set according to the vehicle speed, and even faster detection can be achieved while preventing erroneous detection of wheel slip.

[0098] In Figure 22, the estimated angular velocity ω1 ^ The speed correction gain calculation unit 53 receives the above as input, calculates the running resistance correction gain, and outputs it as one of the correction gains for the skid / slide determination threshold.

[0099] Furthermore, the means for adjusting these determination thresholds may be any method that changes the skid / slide determination threshold in accordance with each parameter, such as an arithmetic formula or a table for each parameter.

[0100] Furthermore, in FIG. 22, values ​​with inverted signs are used as the determination thresholds for slip and slide, but a configuration may be adopted in which correction gain calculations are provided separately for slip and slide.

[0101] As described above, in the second embodiment, the slippage detection and determination unit 15 detects the torque command value τ m * , occupancy information, weather information, position information within the train, and estimated angular velocity value ω1 ^The system incorporates these information and varies the threshold for detecting wheel slippage depending on the information. This makes it possible to prevent false detection of wheel slippage while further increasing the speed of wheel slippage detection. [Embodiment 3]

[0102] The third embodiment according to the present invention differs from the previous second embodiment in that upper and lower limiters are provided for the output of the acceleration / deceleration phase deviation convergence section 21.

[0103] This allows for the detection of slippage to be delayed, or for the rotor acceleration to fluctuate significantly in both positive and negative directions, resulting in the phase deviation information Δθ c The output Δθ of the acceleration / deceleration phase deviation convergence unit 21 α Even if the phase difference becomes excessively large, the control system can be stabilized without reducing the tracking ability of the phase synchronization control section 14.

[0104] FIG. 23 is a diagram illustrating an example of functional blocks of a driving device for a rotating electric machine according to the third embodiment. In embodiment 3, the slip / slide judgment threshold is output from the slip / slide detection judgment unit 15, input to the phase synchronization control unit 14, and used to calculate upper and lower limit values ​​for the output of the acceleration / deceleration phase deviation convergence unit 21.

[0105] FIG. 24 is a diagram showing an example of functional blocks of the skid detection / determination unit 15 according to the third embodiment. A function of outputting a determination threshold value for detecting a skid is added to the skid / slide detection / determination unit 15 according to the second embodiment shown in FIG.

[0106] FIG. 25 is a diagram illustrating an example of functional blocks of the acceleration / deceleration phase deviation convergence unit 21 according to the third embodiment. In the third embodiment, the acceleration / deceleration phase deviation compensation amount Δθ α An upper and lower limiter 22 is provided at the output of the limiter 22. The set value of the upper and lower limiter 22 is set to a value slightly larger than the level at which it is determined that a skid has been detected, by multiplying the skid / slide determination threshold by the determination threshold adjustment gain.

[0107] The upper limit value of the upper / lower limiter 22 is set based on the determination threshold value for slippage, and the lower limit value of the upper / lower limiter 22 is set based on the determination threshold value for sliding.

[0108] FIG. 26 is a diagram showing a comparison of the effects of the presence or absence of the upper and lower limiter 22 provided in the acceleration / deceleration phase deviation convergence unit 21.

[0109] Without a limiter, after detecting slippage and starting to reduce the torque, the acceleration / deceleration phase deviation compensation amount Δθ α Therefore, even after re-adhesion, the acceleration / deceleration phase deviation compensation amount Δθ α It takes time for the voltage to return to a steady state value.

[0110] On the other hand, in the configuration with a limiter according to the third embodiment, when the limit is set to a value slightly larger than the slip / slide detection level, the acceleration / deceleration phase deviation compensation amount Δθ α Prevents excessive acceleration / deceleration phase deviation compensation amount Δθ after re-adhesion α This has the effect of speeding up the convergence of the constant to a steady value.

[0111] In addition, by preventing overshoot, the ON time for detecting wheel slippage is closer to the actual time of wheel slippage, which has the effect of optimizing the timing, and also reduces the risk of falsely detecting wheel slippage after readhesion.

[0112] As described above, in the third embodiment, the upper and lower limiter 22 is provided to the output of the acceleration / deceleration phase deviation convergence unit 21, so that the acceleration / deceleration phase deviation compensation amount Δθ α This accelerates the convergence of the wheel speed to a steady value, thereby preventing false detection of wheel slip after readhesion.

[0113] Although the embodiments of the present invention have been described above, the present invention is not limited to the above-described embodiments, and various modifications are possible within the scope of the gist of the present invention. [Explanation of symbols]

[0114] 1... drive device, 2... control device, 3... voltage output device, 4... synchronous machine, 5... current detector 7... PWM control unit, 8... current detection coordinate conversion unit, 9... current control unit, 10... current command calculation unit, 11...torque command calculation unit, 12...voltage command coordinate conversion unit, 13...phase calculation unit, 14, 14a...phase synchronization control unit, 15...slip / slide detection / determination unit, 16...conventional phase synchronization control unit, 20, 20a...constant speed phase deviation convergence unit, 21, 21a... Phase deviation convergence unit during acceleration / deceleration, 22... Upper and lower limiter, 27... Wheel, 30...rotor shaft, 31...carriage, 32...pinion gear, 33...gearwheel, 34...coupling, 35... axle, 36... rail, 40... frequency estimation control unit, 41... q-axis current deviation convergence section at constant speed, 42... q-axis current deviation convergence section at acceleration / deceleration, 50... equivalent moment of inertia calculation unit, 51... weather information correction gain calculation unit, 52...Intra-organization position information correction gain calculation unit, 53...Speed ​​correction gain calculation unit

Claims

1. A drive device for a synchronous motor mounted on a railway vehicle, a voltage output device that outputs a drive voltage to the synchronous motor; a control device that outputs an estimated angular velocity of the rotor of the synchronous motor; Equipped with The control device uses, as an input quantity, phase deviation information indicating a difference between a phase estimate value of the synchronous motor and a rotor phase of the synchronous motor, and narrows down a torque command value for the synchronous motor based on the phase deviation information. A driving device for a synchronous motor.

2. 2. The synchronous motor drive device according to claim 1, The phase deviation information is an estimated value of the phase deviation by sensorless control. A driving device for a synchronous motor.

3. 2. The synchronous motor drive device according to claim 1, The control device calculates the angular velocity estimate using compensation including a proportionality regulator and an integrator. A driving device for a synchronous motor.

4. 4. The synchronous motor drive device according to claim 3, The control device includes a first compensator, and the first compensator causes a steady-state deviation of the phase deviation information to converge to zero when the rotation speed of the synchronous motor is constant. A driving device for a synchronous motor.

5. 5. The synchronous motor drive device according to claim 4, The control device further includes a second compensator, the second compensator including an integrator, and suppressing the phase deviation information when the steady-state deviation cannot be converged to zero by the first compensator. A driving device for a synchronous motor.

6. 6. A synchronous motor drive device according to claim 5, The control device narrows down a torque command value for the synchronous motor based on the compensation amount output from the second compensation unit. A driving device for a synchronous motor.

7. 6. A synchronous motor drive device according to claim 5, The first compensation unit receives the compensated value by the second compensation unit as an input. A driving device for a synchronous motor.

8. A railway vehicle equipped with a drive device for a synchronous motor according to any one of claims 1 to 7.

9. A method for driving a synchronous motor mounted on a railway vehicle, comprising: a step of controlling an output voltage of a voltage output device that outputs a drive voltage to the synchronous motor; The controlling step includes: calculating an estimated angular velocity of a rotor of the synchronous motor; narrowing down a torque command value for the synchronous motor based on phase deviation information indicating a difference between a phase estimate value of the synchronous motor and a rotor phase of the synchronous motor, A method for driving a synchronous motor.

10. 10. A method for driving a synchronous motor according to claim 9, comprising: The phase deviation information is an estimated value of the phase deviation by sensorless control. A method for driving a synchronous motor.

Images