JP4900591B2 - Rate sensor correction method for measuring the yaw angular velocity of railway vehicles - Google Patents

Description

The present invention relates to the correction how the rate sensor for measuring the yaw rate of the rail vehicle, and particularly to a technique for correcting the zero drift of the rate sensor.

  Conventionally, in a railway vehicle, a yaw angular velocity is measured using a rate sensor. The measured yaw angular velocity is used, for example, for verification of curve detection (see Patent Document 1).

  Patent Document 1 determines that the railway vehicle has reached a curved section from the current position of the railway vehicle and a route map stored in advance, and forcibly pushes the vehicle body according to the curve direction of the route in the curved section. A tilt control method for tilting is disclosed.

  In this tilt control method, the determination result is verified using the yaw angular velocity measured by the rate sensor. That is, when the turning direction of the route indicated by the yaw angular velocity does not coincide with the turning direction of the route indicated by the route map, the vehicle body tilt control is suspended. This avoids a very dangerous situation where the vehicle body is given a reverse tilt.

As seen in this example, it is useful and important to measure the yaw angular velocity using a rate sensor in a railway vehicle.
Japanese Patent No. 3340283

  However, in general, a zero point drift occurs in the output signal of the rate sensor due to fluctuations in supply voltage or temperature changes. Therefore, when a rate sensor is used in a railway vehicle, for example, there is a long-period zero point drift that is difficult to predict due to fluctuations in power supply voltage due to power regeneration from other trains and temperature fluctuations depending on the travel point. May occur in the output signal.

  FIG. 7A is a diagram illustrating an example of an output signal of a rate sensor on which a long-period rising zero point drift rides. In order to eliminate the zero point drift as shown in FIG. 7A, conventionally, the output signal of the rate sensor is often subjected to high-pass filtering.

  FIG. 7B is a diagram illustrating an example of a signal after high-pass filter processing and a true signal. As shown in FIG. 7B, in the signal after the high-pass filter processing, the zero point drift disappears, but the constant value signal for a long time also attenuates.

  The long-time constant value signal indicates, for example, that the railway vehicle is traveling in a long constant curve section. The attenuation of the constant value signal becomes a very serious problem that leads to malfunction of various processes referring to the output of the rate sensor, for example, verification of curve detection as described above.

The present invention has been made in view of the above problems, it aims to provide a correction how to correct without the output signal of the rate sensor for measuring the yaw rate of the railway car, to attenuate the constant value signal And

In order to solve the above problems, a correction method of the present invention is a correction method of a rate sensor that measures a yaw angular velocity of a railway vehicle, and the correction method is performed using route data indicating the position of a curved section of a track. The railway vehicle is curved by comparing the obtained current position with the route data, a position obtaining step for obtaining the current position of the railway vehicle, a speed obtaining step for obtaining a traveling speed of the railway vehicle, and the route data. If it is determined that the railway vehicle is in a curved section, it is determined that the railway vehicle is in a state where it can have a yaw angular velocity . If it is determined that the railway vehicle is not in a curved section, the railway vehicle is average output of the rate sensor when the determining step of determining that there is no state that may have an angular velocity, the rail vehicle is determined not to state that may have a yaw rate A holding step of holding a value as an offset value, look including a calculation step of outputting the difference between the output value and the retained offset value of the rate sensor as a correction value, in the determination step, the railcar curve Even if it is specified that the vehicle is not in the section, if the acquired traveling speed is smaller than a predetermined threshold speed, it is determined that the yaw angular speed can be obtained .

  According to this configuration, when it is determined that the railway vehicle is not in a state where it can have a yaw angular velocity, the average output value of the rate sensor considered to vibrate near the zero point is held as an offset value. In addition, since the output value of the rate sensor is corrected so that the held offset value becomes zero, compared to the case where correction is performed using a conventional high-pass filter, the output time of the rate sensor is long. The constant value signal is not attenuated. Then, by periodically holding a new offset value, it becomes possible to satisfactorily correct the output signal of the rate sensor following the long-period zero point drift which is difficult to predict.

  According to this configuration, an average output value of the rate sensor when it is specified that the railway vehicle is not in a curved section can be held as an offset value. Whether the railway vehicle is in a curved section or not is based on a comparison between the current position of the railway vehicle and the route data, and the output signal of the rate sensor recorded when traveling in advance on the same route Can be specified by pattern matching with the rate sensor output signal in the current travel, or by determining whether or not the bogie angle formed by the body of the railway vehicle and the bogie shows a zero value. You can also.

According to this configuration, when the railway vehicle decelerates, for example, there is a possibility of passing through a crossover that cannot be known in advance in order to enter a station, a vehicle base, etc. By not holding the value, a situation in which an erroneous offset value is held is avoided, and as a result, the correction accuracy is improved.

Further, the correction method of the present invention is a rate sensor correction method for measuring a yaw angular velocity of a railway vehicle, wherein a filter step for removing high-frequency noise contained in an output signal of the rate sensor, and the railway vehicle has a yaw angular velocity. And a result value of the filter step only when it is determined that the railway vehicle does not have a yaw angular velocity continuously for a predetermined threshold time. Holding as the offset value, a calculating step for outputting the difference between the output value of the rate sensor and the held offset value as a correction value, and a time for the railway vehicle to pass through the straight section The lower the specified time, the higher the cutoff frequency for removing the high frequency noise in the filter step. And, and and a constant changing step when shortening the threshold time for holding the offset value in the holding step.

  According to this configuration, the signal value after high frequency noise is removed from the output signal of the rate sensor is held as an offset value. At this time, by setting the threshold time to the convergence time of the output signal in the filter step, the offset value sufficiently converged to the zero value of the rate sensor can be held, and as a result, the accuracy of the correction Helps improve.

  According to this configuration, even when the railway vehicle travels in a continuous curve section, the cut-off function can be used even in a situation where there is no opportunity to hold an offset value using a fixed threshold time. By increasing the frequency, the output signal of the filter step can be quickly converged and the threshold time can be shortened to hold a new offset value at a certain frequency.

  Conversely, when the railway vehicle travels in a very long straight section, the signal after high frequency noise is more reliably removed by the filter step is held as an offset value by lowering the cutoff frequency. Is possible.

  Moreover, the present invention can be realized not only as such a correction method but also as a correction device that executes such a correction method, and causes a computer to execute the steps included in such a correction method. It can also be realized as a program. Needless to say, such a program can be distributed via a recording medium such as a CD-ROM or a transmission medium such as the Internet.

  As described above, according to the correction method of the present invention, the average output value of the rate sensor when it is determined that the railway vehicle is not in a state where it can have the yaw angular velocity is held as an offset value. Since the output value of the rate sensor is corrected so that the offset value becomes zero, the constant value signal for a long time included in the output signal of the rate sensor is compared with the case of correcting using the conventional high-pass filter. There is no attenuation. By periodically holding a new offset value, it becomes possible to satisfactorily follow a long-period zero point drift that is difficult to predict and correct the output signal of the rate sensor.

  A correction method and a correction apparatus according to an embodiment of the present invention will be described in detail with reference to the drawings.

  FIG. 1 is a diagram schematically illustrating an example of a configuration of a railway vehicle on which the correction device according to the present embodiment is mounted.

  The railway vehicle 100 corrects the yaw angular velocity measured by the rate sensor 10 using the correction unit 20 and refers to the corrected yaw angular velocity to perform various controls, for example, verification of curve detection in the above-described vehicle body tilt control. This is an example of a vehicle, and includes a rate sensor 10, a correction unit 20, a position signal receiving unit 30, an axle sensor 40, wheels 50, a carriage 60, and a vehicle body 70.

  Note that the content of the control performed with reference to the corrected yaw angular velocity is not a feature of the present invention, and therefore illustration and detailed description thereof are omitted.

  The rate sensor 10 is, for example, a rate gyro sensor, is attached to the railway vehicle 100 in a direction in which the yaw angular velocity is measured, and outputs a signal representing the measured yaw angular velocity.

  The correction unit 20 is an example of a correction device according to the present invention, and corrects the output signal from the rate sensor 10 so as to remove the zero point drift. The correction unit 20 may be realized as, for example, a hardware circuit, or may be realized as a software function that is exhibited when a DSP (Digital Signal Processor) (not shown) executes a predetermined program.

  The position signal receiving unit 30 is, for example, a wireless receiver, and receives a position signal that informs the railway vehicle 100 of the position. This position signal may be, for example, a signal transmitted indicating the installation position from a transmitter installed at a key point (before moving from a straight section to a curved section). Positioning System) radio signal may be used.

  The axle sensor 40 is a rotary encoder, for example, and detects the amount of rotation of the wheel. The amount of rotation detected per unit time represents the traveling speed of the railway vehicle 100. The detected integrated value of the rotation amount represents the current position of the railway vehicle 100 as a distance from a predetermined base point (for example, the starting station).

  FIG. 2 is a block diagram illustrating an example of a functional configuration of the correction unit 20.

  The correction unit 20 includes a route data storage unit 21, a position and speed specification unit 22, a state determination unit 23, a filter unit 24, a holding unit 25, and a calculation unit 26.

  The route data storage unit 21 stores in advance route data indicating the position of a curved section on a route on which the railway vehicle 100 travels.

  The position and speed specifying unit 22 acquires a signal indicating the amount of rotation of the wheel from the axle sensor 40, and specifies the traveling speed of the railway vehicle 100 from the amount of rotation per unit time represented by the acquired signal. Further, the current position of the railway vehicle 100 is specified by using the integrated value obtained by integrating the rotation amount as a distance from a predetermined base point. And a position signal is acquired from the position signal receiving part 30, and the said integrated value is calibrated so that the position represented by the acquired position signal may be shown.

  The position and speed specifying unit 22 outputs the specified current position and traveling speed of the railway vehicle 100 to the state determining unit 23.

  The state determination unit 23 obtains the current position and the traveling speed from the position and speed specifying unit 22 and refers to the route data from the route data storage unit 21 so that the railway vehicle 100 can have a yaw angular velocity. And the determination result is output to the holding unit 25.

  This determination is made by, for example, identifying whether or not the railway vehicle 100 is in a curved section by comparing the current position acquired from the position and speed specifying unit 22 with the route data stored in the route data storage unit 21. Can be done. That is, it is determined that the railway vehicle 100 is in a state where it can have a yaw angular velocity when it is specified that it is in a curved section, and it is determined that it is not in a state where it can have a yaw angular speed when it is specified that it is not in a curved section.

  Here, even if it is a case where it is specified that the railway vehicle 100 is in a curve section from the current position, the state determination unit 23 is a case where the traveling speed acquired from the position and speed specifying unit 22 is a zero value. It may be determined that the yaw angular velocity is not in a ready state. This is because the railway vehicle 100 cannot naturally have a yaw angular velocity while it is stopped.

  Even if it is determined that the railway vehicle 100 is not in a curved section from the current position, if the acquired traveling speed is smaller than the predetermined threshold speed, the yaw angular speed can be obtained. You may judge. This is because when the railway vehicle 100 decelerates, for example, it may have an unexpected yaw angular velocity by passing through a crossover that is a curve that cannot be known in advance in order to enter a station, a vehicle base, or the like. It is.

  This threshold speed may be, for example, 200 km / h in a high-speed railway system (for example, a Japanese Shinkansen system) operated at a maximum speed of about 300 km / h.

  The low-pass filter unit 24 acquires an output signal from the rate sensor 10 and outputs a signal indicating an average value obtained by removing high-frequency noise included in the acquired signal to the holding unit 25. Here, the output signal from the rate sensor 10 acquired by the filter unit 24 may be numerical data digitized by an analog-digital converter (not shown). The filter unit 24 may be realized by a secondary Butterworth filter having a cutoff frequency of 0.1 Hz, for example.

  The cutoff frequency of the low-pass filter unit 24 is as low as possible so that the output signal of the low-pass filter unit 24 can follow the zero point drift of the rate sensor 10, and the low-pass filter unit when the railway vehicle 100 enters the straight section from the curved section. The 24 output signals should be determined so high that they return (or are referred to as convergence) to the zero point of the rate sensor 10 within a practical time. This cut-off frequency is desirably 0.01 Hz to 0.1 Hz in the high-speed railway system described above.

  When the determination unit 23 obtains a determination result from the state determination unit 23 that the railway vehicle 100 is not in a state where the yaw angular velocity can be maintained for a predetermined threshold time, the output signal from the filter unit 24 is obtained. A register that holds a value as an offset value.

  This threshold time is provided to wait for the output signal from the filter unit 24 to return to the zero point of the rate sensor 10 when the railway vehicle 100 enters the straight section from the curved section. This threshold time is determined from the cut-off frequency of the filter unit 24 and the high-frequency cutoff performance.

  When the filter unit 24 is realized by a Butterworth filter having a cutoff frequency of 0.1 Hz as described above, for example, the step response converges to a deviation of 2% or less after 10 seconds. 15 seconds is desirable.

  Lowering the cut-off frequency of the filter unit 24 and increasing the threshold time of the holding unit 25 is effective for reliably removing high-frequency noise riding on the output signal of the rate sensor 10, but the threshold is reduced. If the value time is excessively large, there is a possibility that the opportunity to hold the offset value may be lost when the railway vehicle 100 travels in a curve continuous section.

  The calculation unit 26 outputs the difference between the value of the output signal of the rate sensor 10 and the offset value held in the holding unit 25 as a correction value. In the correction value output from the calculation unit 26, the value of the output signal of the rate sensor 10 is corrected so that the offset value held in the holding unit 25 becomes zero.

  FIG. 3A is a diagram illustrating an example of a route, and FIG. 3B is a diagram illustrating an example of route data stored in the route data storage unit 21 corresponding to the route.

  FIG. 3A shows an example of a route provided with a straight section, a curved section, a straight section, and a station from left to right.

  The route data shown in FIG. 3 (B) represents the distance Pn from the base point (for example, the starting station) and the description of the points in association with each other for a plurality of points. As shown in this example, the route data represents the positions of the start point and end point of the curve.

  FIG. 4 is a flowchart illustrating an example of the operation performed by the correction unit 20. This process is started and executed periodically (for example, every 1 mS) according to a timer (not shown). In FIG. 4, V is a running speed specified by the position and speed specifying unit 22, V0 is a predetermined threshold speed, T is a counter value, and T0 is a predetermined threshold time. is there.

  The low-pass filter unit 24 outputs the value of the signal after removing high-frequency noise from the output signal of the rate sensor 10 (S01).

  If the traveling speed V is equal to or higher than the threshold speed V0 (YES in S02), the state determining unit 23 compares the current position acquired from the position and speed specifying unit 22 with the route data in the route data storage unit 21. Thus, it is specified whether or not the current position is included in the curved section, that is, whether or not the railway vehicle 100 is in the curved section.

  When the traveling speed V is equal to or higher than the threshold speed V0 and the current position is not included in the curved section (a straight section in S03), and when the traveling speed V is 0 regardless of the current position (in S04). YES), it is determined that the railway vehicle 100 is not in a state where it can have the yaw angular velocity, and the counter value T is incremented by 1 (S05).

  If the counter value T is larger than the time threshold value T0 (YES in S06), the output value of the filter unit 24 is held in the holding unit 25 as an offset value (S07), and the counter value T is reset to 0. (S08).

  On the other hand, when the current position is included in the curve section (curve section in S03), and when the traveling speed V is less than the threshold speed V0 and not 0 regardless of the current position (NO in S04), the railway vehicle It is determined that 100 is in a state where it can have a yaw angular velocity, and the counter value T is reset to 0 (S08).

  In this way, the offset value is held in the holding unit 25 every time it is determined that the railway vehicle 100 is not in a state where it can have the yaw angular velocity for the threshold time T0.

  The calculation unit 26 outputs a value obtained by subtracting the offset value held in the holding unit 25 from the actual measurement value that is the value of the output signal of the rate sensor 10 as a correction value (S09).

  Here, in the flowchart shown in FIG. 4, step S01 is an example of the filter step described in the claims, and steps S02 to S04 are examples of the determination step described in the claims. Steps S05 to S08 are examples of the holding step described in the claims, and step S09 is an example of the calculation steps described in the claims.

  FIG. 5A is a diagram illustrating an example of an output signal of the rate sensor 10 and an example of an output signal of the filter unit 24. The output signal of the rate sensor 10 is loaded with short-period noise and zero-point drift that rises with a long period. In FIG. 5A, for the sake of explanation, the magnitude of the zero point drift is enlarged compared to the magnitude at which the signal actually changes upon receiving the yaw angular velocity.

  FIG. 5B is a diagram illustrating an example of an output signal from the calculation unit 26, that is, a correction result by the correction unit 20.

  In FIG. 5 (A), the state determination unit 23 determines that the current position acquired from the position and speed specifying unit 22 is not included in the curve section indicated by the route data in the route data storage unit 21. It is shown as a straight section. During this period, the output value of the filter unit 24 is held in the holding unit 25 as an offset value every threshold time T0. The point in time when the offset value is held is indicated by a triangular mark.

  Further, a period in which the current position is determined to be included in the curve section by the state determination unit 23 is shown as a curve section in FIG. During this period, the offset value is not held, and the output signal of the rate sensor 10 rises according to the actual yaw angular velocity.

  Furthermore, when entering the straight section from the curved section, a period of waiting for the output value of the filter unit 24 to return to the zero point of the rate sensor 10 while the threshold time T0 elapses is shown in FIG. ) Indicates waiting for return. After the return waiting period elapses, the offset value is periodically held again.

  As described above, according to the correction unit 20, the high-frequency noise included in the output signal of the rate sensor 10 is removed by the filter unit 24, and it is determined that the railway vehicle 100 is not in a state where it can have a yaw angular velocity. Since the output value of the filter unit 24 is held in the holding unit 25 as an offset value, and the difference between the value of the output signal of the rate sensor 10 and the offset value of the holding unit 25 is output as a correction value, a conventional high-pass filter is used. Compared to the correction, the long-time constant value signal included in the output signal of the rate sensor 10 is not attenuated.

  By periodically holding a new offset value, it is possible to satisfactorily follow a long-period zero point drift that is difficult to predict and correct the output signal of the rate sensor 10.

  So far, the present invention has been described based on the embodiments, but the present invention is not limited to the above-described embodiments. The following cases are also included in the present invention.

  In the embodiment, the state determination unit 23 specifies whether or not the railway vehicle 100 is in a curve section by comparing the current position acquired from the position and speed specifying unit 22 and the route data in the route data storage unit 21. However, in addition to this, it may be specified by pattern matching between the output signal of the rate sensor 10 recorded when traveling on the same route in advance and the output signal of the rate sensor 10 in the current traveling. It can also be specified by whether or not the bogie angle, which is the angle formed between the vehicle body 70 and the carriage 60 of the railway vehicle 100, shows a zero value. The bogie angle can be known by expansion and contraction of a yaw damper generally provided between the vehicle body 70 and the carriage 60.

  In the embodiment, the route data represents the positions of the start point and the end point of the curve, but may further represent the positions of the start point and the end point of the station. That is, the distance P2 from the starting point to the starting point of the station and the distance P3 from the starting point to the ending point of the station may be added to FIG. In this way, it is possible to determine that the railway vehicle 100 can have a yaw angular velocity in the station area as well as the curved section, and an error occurs when passing through a crossover that cannot be known in advance in the station area. It is possible to reliably avoid the situation where the offset value is held.

  In the embodiment, the filter unit 24 holds the cutoff frequency for removing high-frequency noise contained in the output signal from the rate sensor 10 and the holding unit 25 holds the output value of the filter unit 24 as an offset value. The threshold times are described as fixed values, but it is also possible to consider a modification in which these values can be changed.

  FIG. 6 is a block diagram illustrating an example of a functional configuration of the correction unit 20a according to such a modification.

  The correction unit 20a is configured by adding a time constant changing unit 27 and changing the filter unit 24a and the holding unit 25a as compared with the correction unit 20 shown in FIG.

  The time constant changing unit 27 refers to the current position and traveling speed acquired from the position and speed specifying unit 22 and the route data in the route data storage unit 21 so that the railway vehicle 100 passes through the straight section ahead. Estimate time. This time may be obtained, for example, by dividing the distance from the end point of the first curve section after the current position to the start point of the next curve section by the traveling speed.

  As the estimated time is shorter, a higher cutoff frequency is instructed to the filter unit 24a, and a shorter threshold time is instructed to the holding unit 25a.

  The filter unit 24a performs the filtering process according to the cutoff frequency instructed from the time constant changing unit 27, and the holding unit 25a holds the offset value according to the threshold time instructed from the time constant changing unit 27.

  According to this configuration, even when the railway vehicle travels in a continuous curve section, even when the fixed cutoff frequency and threshold time are used, there is no chance of holding the offset value. By raising the cut-off frequency, the output signal of the filter unit 24a can be quickly converged, and by shortening the threshold time, a new offset value can be held in the holding unit 25a with a certain frequency.

  Conversely, when the railway vehicle travels in a very long straight section, the high frequency noise is more reliably removed in the filter unit 24a by lowering the cutoff frequency, and the threshold time is increased. The sufficiently converged output value from the unit 24a can be held in the holding unit 25a.

  The correction method and the correction apparatus according to the present invention can be widely used for correcting the output value of a rate sensor that measures the yaw angular velocity of a railway vehicle.

The figure which shows typically an example of a structure of the railway vehicle carrying the correction apparatus in embodiment The block diagram which shows an example of a functional structure of the correction | amendment part 20 (A) The figure which shows an example of a route, (B) The figure which shows an example of the route data memorize | stored in the route data storage part 21 corresponding to the route The flowchart showing an example of the operation | movement which the correction | amendment part 20 performs. (A) The figure which shows an example of the output signal of the rate sensor 10, and the example of the output value of the filter part 24, (B) The figure which shows an example of the output signal of the correction | amendment part 20 The block diagram which shows an example of a functional structure of the correction | amendment part 20a which concerns on a modification. (A) A figure showing an example of an output signal of a rate sensor on which a long-point rising zero point drift rides, (B) A figure showing an example of a signal after high-pass filter processing and a true signal

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Rate sensor 20, 20a Correction | amendment part 21 Route data memory | storage part 22 Position and speed specific | specification part 23 State judgment part 24, 24a Filter part 25, 25a Holding part 26 Calculation part 27 Time constant change part 30 Position signal receiving part 40 Axle sensor 50 Wheel 60 bogie 70 car body 100 railway vehicle

Claims (2)

A rate sensor correction method for measuring the yaw angular velocity of a railway vehicle,
The correction method is performed using route data indicating the position of the curve section of the track,
A position acquisition step of acquiring a current position of the railway vehicle;
A speed acquisition step of acquiring a traveling speed of the railway vehicle ;
By comparing the acquired current position with the route data, it is determined whether or not the railway vehicle is in a curved section, and when it is specified that the railway vehicle is in a curved section, the railway vehicle can have a yaw angular velocity. A determination step of determining that the railway vehicle is not in a state in which the rail vehicle can have a yaw angular velocity when it is determined that there is no curve section;
A holding step of holding an average output value of the rate sensor as an offset value when it is determined that the railway vehicle is not in a state where it can have a yaw angular velocity;
A calculation step of outputting a difference between the output value of the rate sensor and the held offset value as a correction value ,
Even if it is determined in the determination step that the railway vehicle is not in a curved section, if the acquired traveling speed is lower than a predetermined threshold speed, the yaw angular speed can be obtained. A correction method characterized by determining that there is. A rate sensor correction method for measuring the yaw angular velocity of a railway vehicle,
A filter step for removing high-frequency noise contained in the output signal of the rate sensor;
A determination step of determining whether or not the railway vehicle is in a state where it can have a yaw angular velocity;
A holding step of holding the result value of the filter step as an offset value only when it is determined that the railway vehicle is not in a state capable of having a yaw angular velocity continuously for a predetermined threshold time;
A calculation step of outputting a difference between the output value of the rate sensor and the held offset value as a correction value;
The time for the railway vehicle to pass through a straight section is specified, and the shorter the specified time, the higher the cutoff frequency for removing the high-frequency noise in the filter step, and the holding step correcting method characterized by including the constant changing step when shortening the threshold time for holding an offset value.

Images