Abnormal Detection Device, Central Device, and Abnormal Detection Method

Abstract

To provide a technique capable of more easily detecting abnormality in a railway facility operated by a motor.SOLUTION: A data collection device 3 calculates time series data of a torque equivalent value during switching operation by calculating the torque equivalent value based on mutual inductance of an induction motor from motor voltage and motor current of the induction motor of a switch machine 9 and calculates and transmits a feature value of each feature amount as a parameter value to a central device 5 while using a plurality of feature amounts based on the time series data as parameters. The central device 5 obtains a plot position corresponding to the received parameter value in a coordinate system having each parameter as a coordinate axis, with a predetermined reference point where the switch machine 9 is regarded as normal, and determines the presence / absence of an abnormality sign in the switch machine 9 based on a positional relation between the reference point and the plot position.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to an abnormality detection device for railway facilities and the like.

Background Art

[0002] Various methods have been proposed for detecting abnormalities or signs of abnormalities in a turntable, which is one of the railway facilities. For example, a method of determining the presence or absence of a sign of abnormality in a turntable by comparing the motor current waveform during the conversion operation of the turntable with a standard waveform, which is the motor current waveform of a normal turntable, is well known (for example, Patent Document 1).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In recent years, the development of a facility monitoring system equipped with a function for detecting signs of abnormalities in turntables as described above has been underway. In this facility monitoring system, a central device collects conversion data (such as motor current values) related to the conversion operations of each turntable from a remote location and determines the presence or absence of signs of abnormalities. Then, when any sign of abnormality is detected, a maintenance worker goes to the site where the turntable is installed to perform inspections, maintenance, etc.

[0005] However, there are a large number of turntables to be installed, and collecting conversion data for each conversion operation and determining the presence or absence of signs of abnormalities has a problem that the processing load on the central device becomes large. In particular, the method of comparing the motor current waveform with the standard waveform as described above requires a large amount of data and a large amount of calculation for determining the presence or absence of signs of abnormalities, which has been a cause of the large processing load. Such problems are not limited to turntables, but are the same for other railway facilities that operate by motors, such as level crossing barriers and home fences.

[0006] The problem to be solved by the present invention is to provide a technology that enables easier detection of abnormalities in railway facilities operated by a motor.

Means for Solving the Problem

[0007] A first invention for solving the above problems is Based on the motor voltage and motor current of the induction motor for a railway facility (for example, the turntable 9 in FIG. 1) that performs a specified operation from a stopped state and then returns to a stopped state by controlling the power supply to the induction motor, a torque equivalent value based on the mutual inductance of the induction motor is calculated, and torque equivalent value calculation means (for example, the torque equivalent value calculation unit 204 in FIG. 12) for calculating time-series data of the torque equivalent value during the specified operation; Parameter value calculation means (for example, the parameter value calculation unit 206 in FIG. 12) that uses a plurality of feature amounts based on the time-series data as parameters and calculates the feature values of each feature amount as parameter values; A coordinate system with each of the above parameters as coordinate axes, in a coordinate system in which a reference point for considering the railway facility to be normal is predetermined, determines a plot position corresponding to the parameter value calculated by the parameter value calculation means, and based on the positional relationship between the reference point and the plot position, determination means (for example, the determination unit 502 in FIG. 12) for determining the presence or absence of an abnormality omen in the railway facility; An abnormality detection device comprising the above.

[0008] As another invention, The central device of an abnormality detection device comprising a data collection device and a central device, The data collection device calculates a torque equivalent value based on the mutual inductance of the induction motor from the motor voltage and motor current of the induction motor related to railway equipment that performs a specified operation from a stopped state and then returns to the stopped state by controlling the power supply to the induction motor, thereby calculating time-series data of the torque equivalent value during the specified operation (for example, the torque equivalent value calculation unit 204 in FIG. 12), and parameter value calculation means that uses a plurality of feature amounts based on the time-series data as parameters and calculates the feature values of the respective feature amounts as parameter values (for example, the parameter value calculation unit 206 in FIG. 12). Receiving means (for example, the communication unit 406 in FIG. 12) for receiving the parameter values calculated by the parameter value calculation means from the data collection device. A coordinate system with each of the parameters as coordinate axes, in a coordinate system in which a reference point for considering the railway equipment to be normal is predetermined, obtains a plot position corresponding to the parameter value received by the receiving means, and based on the positional relationship between the reference point and the plot position, determination means for determining the presence or absence of an abnormal sign of the railway equipment (for example, the determination unit 502 in FIG. 12). A central device including these may be configured.

[0009] Also, as another invention,[[]] Calculating time-series data of the torque equivalent value during the specified operation by calculating a torque equivalent value based on the mutual inductance of the induction motor from the motor voltage and motor current of the induction motor related to railway equipment that performs a specified operation from a stopped state and then returns to the stopped state by controlling the power supply to the induction motor (for example, step S3 in FIG. 15). Using a plurality of feature amounts based on the time-series data as parameters and calculating the feature values of the respective feature amounts as parameter values (for example, steps S5 and S7 in FIG. 15). A coordinate system with the respective parameters as coordinate axes, in a coordinate system in which a reference point considering the railway facilities to be normal is predetermined, obtains a plot position corresponding to the calculated parameter value, and based on the positional relationship between the reference point and the plot position, determines the presence or absence of an abnormal sign of the railway facilities (for example, steps S11 and S13 in FIG. 15), and An abnormality detection method including the above may be configured.

[0010] According to the first invention or the like, it is possible to more easily realize the abnormality detection of railway facilities operated by an induction motor. That is, any abnormality that occurs during the operation of the railway facilities appears as a change in the torque of the induction motor that operates the railway facilities. Therefore, based on the time-series data of the torque equivalent value of the induction motor, it is possible to determine the abnormal sign that has occurred in the railway facilities. Since the parameter values of each parameter calculated based on the time-series data of the torque equivalent value represent the characteristics of the time-series data, the positional relationship between the reference point and the plot position in the coordinate system with each parameter as the coordinate axis corresponds to the difference between the time-series data of the torque equivalent value during normal operation of the railway facilities and the time-series data of the torque equivalent value to be determined. As a result, it is possible to determine the presence or absence of an abnormal sign by a relatively simple process based on the positional relationship between the plot position obtained by plotting the parameter values of a plurality of parameters and the reference point, so that the amount of calculation required for abnormality detection can be reduced and the processing load can be reduced. In addition, since a plurality of characteristic values are comprehensively determined, it is possible to accurately determine the presence or absence of an abnormal sign.

[0011] The second invention is the same as the first invention, The parameter value calculation means uses at least two of the maximum value of the torque equivalent value in a predetermined period after the motor start period, the average value of the torque equivalent value in the predetermined period, and the operation time related to the specified operation as the characteristic values. It is an abnormality detection device.

[0012] According to the second invention, by using at least two of the maximum value of the torque equivalent value in a predetermined period after the motor starting period of the induction motor, the average value of the torque equivalent value in the predetermined period, and the operation time related to the specified operation as characteristic values, it is possible to determine whether there is a good-precision abnormal omen. That is, any abnormality that occurs during the operation of railway facilities appears as a change in the torque of the induction motor. However, the torque fluctuates greatly during the motor starting period when the induction motor starts operating. Therefore, for example, the maximum value of the torque equivalent value in a predetermined period after the motor starting period and the average value of the torque equivalent value in the predetermined period can be used as characteristic values suitable for determining the presence or absence of an abnormal omen in railway facilities. Also, if the railway facilities are normal, the operation time related to the specified operation is almost constant, but when some abnormality occurs, the operation time related to the specified operation often changes. Therefore, the operation time related to the specified operation of railway facilities can also be used as a characteristic value suitable for determining the presence or absence of an abnormal omen in railway facilities.

[0013] The third invention is the invention according to the first or second invention, The determination means determines the presence or absence of an abnormal omen in the railway facilities based on the direction of the plot position with respect to the reference point in the coordinate system. It is an abnormality detection device.

[0014] According to the third invention, it is possible to determine that there is an abnormal omen in the railway facilities based on the direction of the plot position with respect to the reference point in the coordinate system. That is, the direction in the coordinate system represents the magnitude relationship including the positive and negative with respect to the reference point of the parameter values of each parameter that is a coordinate axis. Therefore, according to the type of characteristic value of each parameter, by associating and determining the combination of the magnitude relationship of the parameter values of each parameter with respect to the reference point and the presence or absence of an abnormal omen, it is possible to determine the presence or absence of an abnormal omen in the railway facilities from the direction of the plot position with respect to the reference point.

[0015] The fourth invention is the invention according to any one of the first to third inventions, Based on the positional relationship, there is a warning level determination means (for example, the determination unit 502 in FIG. 12) for determining the warning level of the abnormal omen. It is an abnormality detection device further comprising

[0016] According to the fourth invention, based on the positional relationship between the reference point and the plot position in the coordinate system, the attention - calling level of the abnormality omen can be determined. For example, it can be determined that the longer the distance between the reference point and the plot position in the coordinate system, the higher the attention - calling level.

[0017] The fifth invention is based on the fourth invention, The attention - calling level determination means determines the attention - calling level based on the transition of the plot positions corresponding to a plurality of the specified operations in chronological order. It is an abnormality detection device.

[0018] According to the fifth invention, the attention - calling level of the abnormality omen can be determined based on the transition of the plot positions corresponding to a plurality of the specified operations in chronological order. For example, when the plot position moves away from the reference point, it can be determined that the attention - calling level is increased.

[0019] The sixth invention is based on the fourth invention, The attention - calling level determination means determines the attention - calling level based on the change in the interval between the plot positions corresponding to a plurality of the specified operations in chronological order. It is an abnormality detection device.

[0020] According to the sixth invention, the attention - calling level of the abnormality omen can be determined based on the change in the interval of the plot positions in chronological order. The interval of the plot positions in the coordinate system represents the degree of change in the characteristic value of the time - series data of the torque - equivalent value corresponding to the plot position. Therefore, for example, when the change in the interval of the plot positions is large, it can be determined that the attention - calling level is increased.

[0021] The seventh invention is based on the sixth invention, The attention arousal level determination means obtains the distance between the plot positions using the Mahalanobis distance based on the plot positions in the coordinate system, and determines the attention arousal level based on the change in the distance. It is an abnormality detection device.

[0022] According to the seventh invention, the distance between the plot positions can be obtained using the Mahalanobis distance.

[0023] The eighth invention is in any one of the first to seventh inventions, The railway facility is any one of a turntable, a switch machine, and a platform fence. It is an abnormality detection device.

[0024] According to the eighth invention, abnormality detection for any one of a turntable, a switch machine, and a platform fence can be realized with a small processing load.

[0025] The ninth invention is in any one of the first to eighth inventions, A data collection device having the torque equivalent value calculation means and the parameter value calculation means, and acquiring the motor voltage and the motor current from the railway facility; A central device communicatively connected to the data collection device via a communication network, the central device having the determination means; It is an abnormality detection device comprising:

[0026] According to the ninth invention, the data collection device calculates the parameter values of the respective parameters based on the motor drive information related to the specified operation of the railway facility and transmits them to the central device. The central device determines the presence or absence of an abnormality omen of the railway facility based on the parameter values of the respective parameters received. Therefore, the amount of data communicated between the data collection device and the central device can be reduced, and the processing load of the central device can be reduced. As a result, it becomes easy to apply to a system for centrally monitoring a large number of railway facilities.

Brief Description of the Drawings

[0027]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15

Figure 16

Figure 17

Embodiments for Carrying Out the Invention

[0028] Hereinafter, preferred embodiments of the present invention will be described with reference to the drawings. Note that the applicable forms of the present invention are not limited to the following embodiments. Also, in the description of the drawings, the same reference numerals are given to the same elements.

[0029] [Overall Configuration] FIG. 1 is a diagram showing a configuration example of the abnormality detection device 1 according to the present embodiment. The abnormality detection device 1 is a device that detects an abnormality in a turntable 9, which is an example of railway equipment, and includes a data collection device 3 associated with each of a plurality of turntables 9, and a central device 5 communicably connected to each of the data collection devices 3 via a communication network N.

[0030] The turntable 9 includes an induction motor 7 as a drive source, and performs a switching operation by controlling power supply to the induction motor 7. That is, after the turntable 9 performs a switching operation, which is a specified operation of switching the tong rail to the normal position or the reverse position by supplying power to the induction motor 7 that drives the turntable 9 and causing the induction motor 7 to rotate, the power supply to the induction motor 7 is stopped and the rotation operation stops, so that the turntable 9 returns to the stopped state again.

[0031] The data collection device 3 is installed inside the corresponding turntable 9 or in an instrument box installed together. The data collection device 3 acquires motor drive information, which is the motor voltage and motor current of the induction motor 7 of the corresponding turntable 9. Specifically, during the switching operation of the turntable 9, for example, the motor voltage measured by a voltage sensor and the motor current measured by a current sensor at a predetermined interval of 50 milliseconds are input and sampled, and A / D converted to obtain digital values. That is, the motor drive information is time-series data of the motor voltage and the motor current. The start and end of the switching operation of the turntable 9 can be determined, for example, by whether the motor current is equal to or greater than a predetermined threshold value, a switching command from an external device, or the like.

[0032] Then, based on the motor drive information related to one conversion operation acquired, the data collection device 3 calculates N (N≥2) feature amounts as parameters and calculates the feature values of each feature amount as parameter values, and transmits (outputs) the calculated parameter values to the central device 5. In the present embodiment, the maximum value of the torque equivalent value (maximum torque equivalent value) during the stroke period and the conversion time of the conversion operation, which is the operation time related to the specified operation of the tumbler 9, are used as the feature values to calculate the parameter values of two parameters (N = 2). In the present embodiment, this parameter is appropriately referred to as a "forecast parameter".

[0033] The central device 5 is installed in the station building, command post, etc., but may also be realized on a cloud system. Furthermore, it may be realized as an on-vehicle device mounted on an operating train, maintenance vehicle, etc. The central device 5 determines the presence or absence of an abnormality forecast of the corresponding tumbler 9 by using the parameter values of the forecast parameters received from the data collection device 3.

[0034] [Determination of the presence or absence of an abnormality forecast] (A) Calculation of parameter values of forecast parameters When calculating the parameter values of the forecast parameters, first, based on the time-series data of the motor voltage and motor current, which are the motor drive information for one conversion operation of the tumbler 9, the time-series data of the torque equivalent value is calculated. The torque equivalent value is a value corresponding to the torque of the induction motor 7. In the present embodiment, the torque equivalent value k is calculated from the motor voltage V1 and motor current I1, which are the motor drive information, according to the following formula (1).

Equation

[0035] In Equation (1), "R1" is the resistance (real part) of the primary side (motor connection side: stator) of the induction motor 7, and "ω" is the angular frequency of the AC power supplied to the induction motor 7. This torque equivalent value k is a value based on the mutual inductance M of the induction motor 7, and in this embodiment, it is the reciprocal (1 / M) of the mutual inductance M. Then, based on the time-series data of the torque equivalent value for one conversion operation of the transfer machine 9 calculated, the parameter value of the omen parameter is calculated.

[0036] (B) Torque equivalent value The reason for using the torque equivalent value in calculating the parameter value of the omen parameter is that slip (motor slip) occurs in the induction motor 7 due to fluctuations in the motor voltage. That is, when the load of the induction motor 7 changes (increases) due to some abnormality during the conversion operation of the transfer machine 9, this change (increase) in the load results in a change (increase) in torque, which appears as a change (increase) in the motor current. Therefore, it is possible to determine the presence or absence of an abnormality omen of the transfer machine 9 from the change (increase) in the motor current. However, in the induction motor 7, due to its construction principle, even if the load is constant, when the motor voltage fluctuates, slip occurs due to the fluctuation of the motor voltage, and the motor current also fluctuates due to this slip. That is, it is impossible to distinguish whether the fluctuation of the motor current is due to the fluctuation of the load or due to the motor slip. Therefore, in this embodiment, the torque equivalent value calculated by Equation (1), which takes into account the motor voltage V1, a factor causing motor slip, is used in calculating the parameter value of the omen parameter.

[0037] The fluctuations in the motor voltage V1 that cause the motor slip of the induction motor 7 mainly occur for the following reasons. Fig. 2 is an example of the wiring diagram of the power supply to the induction motors 7 of a plurality of transfer machines 9 in the station building. Usually, a plurality of transfer machines 9 are installed in the station building. AC power is supplied from the power supply line to the induction motors 7 of each of these plurality of transfer machines 9, and the power supply line is a cross-connection wiring from a common power supply AC. Note that the induction motor 7 of the present embodiment is a single-phase motor. Also, the transfer machine 9 can be individually controlled whether to perform the conversion operation or not, and the supply of driving power to the induction motor 7 of each transfer machine 9 is performed by on / off control of a switch SW provided between the power supply line and the induction motor 7. That is, driving power is supplied only to the induction motor 7 of the transfer machine 9 that performs the conversion operation, and control is performed so that driving power is not supplied to the induction motor 7 of the transfer machine 9 that does not perform the conversion operation.

[0038] Since the induction motor 7 has a low internal impedance, it requires excessive power at startup. Therefore, depending on the number and combination of transfer machines 9 that perform the conversion operation simultaneously, that is, the number and combination of induction motors 7 to which driving power is supplied simultaneously, the motor voltage V1 of the induction motor 7 fluctuates due to the voltage division of the cable impedance of the cross-connection wiring. This fluctuation in the motor voltage V1 causes the slip of the induction motor 7 (motor slip), and this motor slip causes the motor current of the induction motor 7 to fluctuate.

[0039] Fig. 3 is an example of the waveforms of the motor voltage and motor current of the induction motor 7 related to the conversion operation of the transfer machine 9, and schematically shows the waveforms in which the motor current and motor voltage, which are discrete values, are continuous. In Fig. 3, the horizontal axis represents the common time, the waveform of the motor current (current waveform) is shown on the upper side, and the waveform of the motor voltage (voltage waveform) is shown on the lower side. The time is set to "0 seconds" at the start point of the conversion operation (the start point of startup of the induction motor 7).

[0040] Two types of operation controls were performed: when the transfer machine 9 was made to perform conversion operations a plurality of times alone (i.e., when no other transfer machine 9 performed a conversion operation), and when the transfer machine 9 was made to perform conversion operations a plurality of times simultaneously with another transfer machine 9. When the transfer machine 9 was operated for conversion alone, no motor slip occurred. When the transfer machine 9 was operated for conversion simultaneously with another transfer machine 9, motor slip occurred. Fig. 3 shows two types of waveform patterns of conversion operations for a plurality of times where no motor slip occurred (hereinafter referred to as "waveform pattern without motor slip" or "without motor slip") and waveform patterns of conversion operations for a plurality of times where motor slip occurred (hereinafter referred to as "waveform pattern with motor slip" or "with motor slip") superimposed on each other.

[0041] When the transfer machine 9 performs a conversion operation simultaneously with another transfer machine 9, it means that the two machines, the transfer machine 9 and another transfer machine 9, are supplied with driving power from a common power source and perform a conversion operation simultaneously. Also, each waveform pattern shown in Fig. 3 is a waveform of a conversion operation with the same conversion direction (positioning or reverse positioning). Also, both the transfer machine 9 and the other transfer machine 9 are in a normal state.

[0042] According to the voltage waveform in Fig. 3, the motor voltage of the transfer machine 9 differs depending on whether another transfer machine 9 is performing a conversion operation simultaneously, that is, whether driving power is also supplied to the induction motor 7 of another transfer machine 9. And as shown by the current waveform in Fig. 3, due to the difference in motor voltage, the motor current during the stroke period fluctuates. The stroke period is the period in one conversion operation during which the operation rod is linearly moved to switch the tong rail left and right. In Fig. 3, during this stroke period, there is a time shift between the case where the transfer machine 9 performs a conversion operation alone and the case where the transfer machine 9 performs a conversion operation simultaneously with another transfer machine 9.

[0043] Incidentally, the conversion operation of the transfer machine 9 is divided into three periods: an unlocking period for unlocking the locking mechanism, a stroke period for linearly moving the operating rod to convert the tong rail left and right, and a locking period for locking the locking mechanism. In Fig. 3, the point in time when the motor current rapidly decreases to zero (which can also be said to be the point in time when the motor voltage rapidly increases to reach 110V) is the end point of the conversion operation. Also, when there is no motor slip, the point in time when "about 1.7 seconds" has elapsed since the start of the conversion operation is the start point of the stroke period. Further, the unlocking period is a period during which a load is applied only to the drive mechanism inside the transfer machine 9 including gears, bearings, clutches, etc., and furthermore, it is divided into a motor starting period in which power supply to the induction motor 7 is started and the number of revolutions per unit time increases, and a stable period in which the number of revolutions per unit time stabilizes.

[0044] Also, as shown by the voltage waveform and the current waveform, the length of the motor starting period, which is the period immediately after the start of the induction motor 7 and until just before the motor voltage and the motor current stabilize, is different between the case of no motor slip (when the transfer machine 9 performs the conversion operation alone) and the case of motor slip (when the transfer machine 9 performs the conversion operation simultaneously with another transfer machine 9). This difference in the length of the motor starting period appears as a difference in the conversion time of the conversion operation.

[0045] The difference in the motor voltage and the difference in the length of the motor starting period are due to motor slip. In the waveform pattern with motor slip, compared to the waveform pattern without motor slip, the motor voltage decreases and the length of the motor starting period becomes longer. Along with this, a time shift occurs in the stroke period.

[0046] Figure 4 is an equivalent circuit diagram of the induction motor 7. As shown in Figure 4, the induction motor 7 can be represented as an equivalent circuit equivalent to a transformer. In this equivalent circuit, AC drive power is supplied from the power source AC to the primary side (motor connection side: stator), and the load, the rolling machine 9, is connected to the secondary side (rotor). The induced current I2 flowing through the secondary side (rotor) becomes a value corresponding to the load torque. Consider the case where the motor voltage V1 (voltage on the primary side) of the induction motor 7 changes in a state where the rolling machine 9 is normal, that is, the load of the induction motor 7 does not change. In this case, in order not to change the induced current I2 on the secondary side (rotor), slip (motor slip) s occurs in the induction motor 7 and the load resistance R2 changes. The motor slip becomes a change in the electromagnetic coupling between the primary side and the secondary side, that is, a change in the mutual inductance M, and this change in the mutual inductance M causes a change in the motor current I1 on the primary side.

[0047] Figure 5 is a diagram showing the impedance on the primary side of the induction motor 7 in the complex plane. The horizontal axis is the real axis and the vertical axis is the imaginary axis. The combined impedance of the real number component R1 and the imaginary number component Mω becomes the impedance on the primary side. And the following equation (2) holds between the voltage (motor voltage) V1 and the current (motor current) I1 on the primary side and the combined impedance.

Equation

[0048] When this equation (2) is transformed, it becomes equation (1). That is, equation (1), which is the reciprocal of the mutual inductance M, is the torque equivalent value k. In other words, the torque equivalent value k is a value based on the mutual inductance M of the induction motor 7.

[0049] FIG. 6 shows waveforms obtained by converting the voltage waveform and the current waveform shown in FIG. 3 into torque equivalent values (hereinafter, appropriately referred to as “converted waveforms”). Similar to the voltage waveform and the current waveform, the torque equivalent values, which are discrete values, are schematically shown as continuous waveforms. That is, for each of the multiple conversion operations in which the voltage waveform and the current waveform are shown in FIG. 3, each waveform of the torque equivalent value calculated according to Equation (1) using the motor voltage V1 and the motor current I1, which are motor drive information. The horizontal axis is the same time axis as the horizontal axis in FIG. 3, and the start point of the conversion operation (the start point of the induction motor 7) is set to “0 seconds”. The vertical axis is the torque equivalent value. By using the torque equivalent value, the amount of information (data amount) can be approximately halved from the amount of information of the original information, the motor voltage V1 and the motor current I1.

[0050] Also in the waveform of the torque equivalent value (converted waveform), similar to the voltage waveform and the current waveform shown in FIG. 3, there are differences in the length of the period of the motor starting period of the induction motor 7 due to motor slip. It can be seen that there are two types: a waveform pattern without motor slip (when the transfer device 9 performs a conversion operation alone) and a waveform pattern with motor slip (when the transfer device 9 performs a conversion operation simultaneously with another transfer device 9). And because there are differences in the length of the period of the motor starting period, there are also differences in the conversion time of the conversion operation, and there are also temporal differences in the start point of the stroke period in the conversion operation. These points are the same as the voltage waveform and the current waveform shown in FIG. 3. Since the maximum torque equivalent value, which is one of the characteristic values of the prediction parameter, is the maximum value of the torque equivalent value in the stroke period, which is the period after the motor starting period, it is necessary to accurately determine the stroke period in the conversion operation also in the waveform of the torque equivalent value (converted waveform).

[0051] FIG. 7 shows waveforms after moving the converted waveforms of the torque equivalent value for each conversion operation shown in FIG. 6 in the direction along the time axis so as to align the start point of the stable period (which can also be said to be the end point of the motor starting period). Specifically, the conversion waveform without motor slip is left as it is, and the conversion waveform with motor slip is moved so as to move backward in time.

[0052] As shown in FIG. 7, by aligning the start points of the stable periods of the respective conversion waveforms (the end points of the motor starting period), it can be seen that for each of the stable period, the stroke period, and the locking period of the respective conversion waveforms, regardless of the presence or absence of motor slip, their lengths and the start / end timings are almost the same. That is, by aligning the start points of the stable periods of the respective conversion waveforms, the periods after the start points of the stable periods of the respective conversion waveforms, that is, the torque equivalent values sampled at a predetermined period in each of the stable period, the stroke period, and the locking period, become data without a temporal deviation from which the influence of motor slip has been removed. Therefore, by aligning the start points of the stable periods of the respective conversion waveforms, the stroke period can be accurately determined, and the presence or absence of an abnormal sign of the pantograph 9 can be accurately determined without being affected by motor slip.

[0053] A method for aligning the start points of the stable periods of the respective conversion waveforms will be described. The start point of the stable period is the P point shown in FIG. 7. This P point is a characteristic waveform part that becomes a constant value after the torque equivalent value sharply decreases. The P point can be determined, for example, as a point where the amount of change (time differential value) between the data before and after in time is almost zero in the time-series torque equivalent value data which is the conversion waveform data.

[0054] In order to align the P points (start point of the steady state) of each conversion waveform, for the conversion waveform without motor slip, the conversion waveform with motor slip is moved backward in time. The discrimination between the conversion waveform without motor slip and the conversion waveform with motor slip may be determined by whether the conversion operation starts and ends simultaneously with other commutation machines 9 based on the commutation information of the commutation machine 9 managed by an external device, or may be determined from the minimum voltage level in the motor starting period of each conversion waveform. That is, as shown in the voltage waveform of FIG. 3, the minimum voltage level in the motor starting period is lower for the case with motor slip than for the case without motor slip. The difference in the minimum voltage level is caused by the voltage drop due to the supply of driving power to other commutation machines 9 that perform the commutation operation simultaneously, as shown in FIG. 2. The larger the number of other commutation machines 9 that perform the commutation operation simultaneously, the greater the decrease in the motor voltage, so the degree of influence of motor slip increases and the minimum voltage level in the motor starting period decreases. Therefore, by comparing the minimum voltage levels in the motor starting period of each conversion waveform and moving the conversion waveform with the lower minimum voltage level in the motor starting period backward in time so that the P point of the conversion waveform coincides with the P point of the conversion waveform with the higher minimum voltage level in the motor starting period, the P points of each conversion waveform can be aligned.

[0055] As a method of aligning the P points of each conversion waveform, for example, it can be performed by collating within a collation range including several tens of data before and after in time centered on the P point of each conversion waveform. Specifically, it can be realized by moving the conversion waveform data with motor slip backward in time so that the error becomes minimum by the least squares method for the data in the collation range of each conversion waveform. Alternatively, a conversion waveform image obtained by imaging the waveform portion corresponding to the data in the collation range of each conversion waveform may be generated, and the conversion waveform with motor slip may be shifted until the image matching degree between the conversion waveforms is high, and the time shifted in time may be calculated from the number of shifted pixels. In any case, by providing a certain collation range, the start points of the steady state can be accurately aligned.

[0056] Thus, the determination of the start point of the steady state of the conversion waveform (which can also be said to be the end point of the motor starting period) is hereinafter appropriately referred to as "slip correction". By performing slip correction, the start point of the steady state of the conversion waveform can be made to coincide with the start point of the steady state of the conversion waveform without motor slip.

[0057] According to the conversion waveforms shown in FIG. 7, it can be seen that regardless of the presence or absence of motor slip, the lengths of the steady state periods of the respective conversion waveforms are substantially the same. And since the start points of the steady states of the respective conversion waveforms (which can also be said to be the end points of the motor starting periods) are substantially the same, it can be seen that the start points of the stroke periods are also substantially the same.

[0058] (C) Conversion waveform FIG. 8 is a diagram showing a conversion waveform of a torque equivalent value obtained by one conversion operation of the transfer press 9, and schematically shows the types of conversion waveforms that can be obtained. The conversion waveform schematically shows the time series data of the torque equivalent value calculated from the motor drive information as a waveform in which the torque equivalent values, which are discrete values, are continuous, with the horizontal axis being time and the vertical axis being the torque equivalent value. The time is set to "0 seconds" at the start point of the conversion operation.

[0059] Figure 8 shows two conversion waveforms (normal conversion waveforms) when the transfer machine 9 is normal, along with three conversion waveforms (abnormality omen conversion waveforms) when an abnormality omen has occurred in the transfer machine 9. The two normal conversion waveforms are a no-slip normal conversion waveform and a with-slip normal conversion waveform. The no-slip normal conversion waveform is the conversion waveform when the transfer machine 9 performs a conversion operation alone, that is, the conversion waveform when there is no motor slip due to voltage fluctuation. The with-slip normal conversion waveform is the conversion waveform when motor slip occurs due to voltage fluctuation. The three abnormality omen conversion waveforms are a light-load conversion waveform, a heavy-load conversion waveform, and a maintenance failure conversion waveform. The light-load conversion waveform is the conversion waveform when an increase in load occurs due to an abnormality in the transfer machine 9, such as an increase in the sliding resistance between the tong rail and the floor plate due to foreign matter intrusion or insufficient lubricating oil. The heavy-load conversion waveform is the conversion waveform when a further increase in load occurs compared to the light-load conversion waveform. The maintenance failure conversion waveform is the conversion waveform when the transfer machine 9 is artificially operated during a conversion test during maintenance or when devices such as a voltage sensor or a current sensor fail. Note that for all the abnormality omen conversion waveforms, it is assumed that there is no motor slip due to voltage fluctuation.

[0060] The conversion operation of the transfer machine 9 is divided into three periods: an unlocking period in which the locking mechanism is unlocked in the stopped state, a stroke period in which the operating rod is linearly operated to turn the tong rail left and right, and a locking period in which the locking mechanism is locked. The unlocking period is further divided into a motor starting period in which power supply to the induction motor 7 is started and the number of revolutions per unit time increases, and a stable period in which the number of revolutions per unit time stabilizes. Figure 8 shows each period for the no-slip normal conversion waveform.

[0061] Comparing the five conversion waveforms in Fig. 8, the following characteristics can be observed. First, when comparing the no-slip normal conversion waveform (which is a normal conversion waveform) with the normal conversion waveform with slip, since motor slip due to voltage fluctuations mainly occurs during the starting period of the induction motor 7, the length of the starting period in the normal conversion waveform with slip is longer than that in the no-slip normal conversion waveform. As a result, the conversion time of the conversion waveform with slip is longer than that of the no-slip conversion waveform. Also, the torque equivalent values are approximately the same magnitude.

[0062] When comparing the no-slip normal conversion waveform with the light-load conversion waveform and the heavy-load conversion waveform, which are abnormal omen waveforms, the torque equivalent value associated with the conversion of the tong rail due to the increase in load during the stroke period is increased in the light-load conversion waveform and the heavy-load conversion waveform compared to the no-slip normal conversion waveform, and the length of the stroke period is longer. Also, compared to the light-load conversion waveform, the heavy-load conversion waveform has a greater degree of increase in the torque equivalent value and a longer stroke period. It can be seen that the greater the degree of increase in load, the greater the increase in the torque equivalent value during the stroke period and the longer the length of that period.

[0063] When comparing the no-slip normal conversion waveform with the maintenance failure conversion waveform, which is an abnormal omen waveform, the magnitudes of the torque equivalent values are approximately equal. However, it can be seen that due to the artificial operation of the conversion operation, the lengths of the stroke period and the conversion time are much shorter in the maintenance failure conversion waveform than in the no-slip normal conversion waveform.

[0064] Since the conversion waveform has such characteristics, in this embodiment, two prognostic parameters using the maximum torque equivalent value and the conversion time during the stroke period as characteristic quantities are used. Note that the number of prognostic parameters is not limited to two, and may be three or more. For example, prognostic parameters having as characteristic quantities the maximum value or average value of the torque equivalent value in the stable period, the average value of the torque equivalent value in the stroke period, and the maximum value of the torque equivalent value in the motor starting period can be considered. Furthermore, environmental factors such as the temperature and humidity during the conversion operation may be used as prognostic parameters having as characteristic quantities.

[0065] (D) Determination of the presence or absence of an abnormal omen The determination of the presence or absence of an abnormal omen of the tumbler 9 based on the parameter values of the prognostic parameters is performed based on the positional relationship between the plotted position of the parameter value of each calculated prognostic parameter and the reference point that regards the tumbler 9 as normal in an N-dimensional coordinate system with each prognostic parameter as a coordinate axis. In this embodiment, since it is based on two prognostic parameters, a two-dimensional coordinate system with these two prognostic parameters as coordinate axes is used.

[0066] FIG. 9 is a diagram showing an example of a two-dimensional coordinate system. In the two-dimensional coordinate system shown in FIG. 9, the horizontal axis is the conversion time, which is one of the prognostic parameters, and the vertical axis is the maximum torque equivalent value, which is the other prognostic parameter. In this two-dimensional coordinate system, the presence or absence of an abnormal omen of the tumbler 9 is determined based on the positional relationship between the plotted position where the parameter values of the two prognostic parameters are plotted and the reference position. Specifically, the presence or absence of an abnormal omen of the tumbler 9 is determined based on the direction of the plotted position with respect to the reference point.

[0067] The reference point can be defined as the coordinate position corresponding to the plot position of the parameter value of the prognostic parameter obtained from the conversion waveform (normal conversion waveform without slip, see Fig. 8) when, for example, the transfer machine 9 is normal and no motor slip occurs. And the normal range can be defined as a range that includes the reference point and includes the plot position of the parameter value of the prognostic parameter obtained from the conversion waveform (normal conversion waveform with slip, see Fig. 8) when the transfer machine 9 is normal but motor slip occurs. That is, the normal range is a range in which the transfer machine 9 is regarded as normal. Also, the abnormal prognostic range can be defined as a range that includes the plot position of the parameter value of the prognostic parameter obtained from the conversion waveform (abnormal prognostic conversion waveform, see Fig. 8) when an abnormal prognosis occurs in the transfer machine 9.

[0068] In this embodiment, with respect to the reference point, a range where the conversion time is long and the value corresponding to the maximum torque is large, that is, a range where the parameter values of each prognostic parameter are large, is defined as the abnormal prognostic range. This is because, as shown in Fig. 8, the light-load conversion waveform and the heavy-load conversion waveform have a longer conversion time and a larger value corresponding to the maximum torque compared to the normal conversion waveform. That is, the abnormal prognostic range is a range corresponding to the direction in which both the conversion time, which is a characteristic value of the prognostic parameter, and the value corresponding to the maximum torque are large with respect to the reference point. And the range other than this normal range and abnormal prognostic range is defined as the maintenance / failure range.

[0069] The state of the transfer machine 9 is determined by the range in which the plot position is located among the ranges defined in the two-dimensional coordinate system. That is, if the plot position is located within the abnormal prognostic range, it is determined that there is an abnormal prognosis in the transfer machine 9, and if it is located within the normal range, it is determined that there is no abnormal prognosis (normal) in the transfer machine 9.

[0070] Also, by dividing the abnormal prognostic range in the two-dimensional coordinate system into a plurality of ranges and setting them, it is also possible to determine the attention level of the abnormal prognosis. This attention level represents, for example, that the higher the level, the higher the severity of the abnormal prognosis.

[0071] FIG. 10 is a diagram showing an example in which an abnormal sign range is divided into a plurality of ranges in a two-dimensional coordinate system. In the example of FIG. 10, the abnormal sign range is divided into two ranges, and two levels of warning levels are associated with each range. As can be seen from the light load conversion waveform and the heavy load conversion waveform shown in FIG. 8, the higher the degree of abnormality of the transfer machine 9, the higher the load on the induction motor 7, so the conversion time and the maximum torque equivalent value increase during conversion. Therefore, in the present embodiment, as shown in FIG. 10, the range is defined such that the warning level of the abnormal sign becomes higher as the conversion time becomes longer or the maximum torque equivalent value becomes larger with respect to the reference point. The warning level can be determined from the direction and distance of the plot position with respect to the reference point. Needless to say, the warning level is not limited to two levels, and may be three levels or more.

[0072] Further, instead of the plot position for one conversion operation, the presence or absence of an abnormal sign of the transfer machine 9 may be determined based on the plot positions for each of a plurality of conversion operations. FIG. 11 is a diagram showing an example of the plot positions for each of a plurality of conversion operations in a two-dimensional coordinate system. In this case, for example, the presence or absence of an abnormal sign can be determined based on the ratio of the plot positions located within the abnormal sign range among the most recent M times (M>2) of plot positions. For example, it can be said that when the ratio is equal to or higher than a predetermined ratio, it is determined that there is an abnormal sign of the transfer machine 9.

[0073] By basing on the plot positions for a plurality of times, it is possible to eliminate the mistake of determining a temporary increase in the torque equivalent value of the induction motor 7 that is not caused by the abnormality of the transfer machine 9 as an abnormal sign. Thereby, the accuracy of the determination of the abnormal sign can be improved. Since the conversion waveform may differ depending on the conversion direction even for the same transfer machine 9, it is preferable to base on the plot positions for a plurality of conversion operations having the same conversion direction.

[0074] Further, based on the time-series transition of the plot positions corresponding to each of the multiple conversion operations, it is also possible to determine the attention - calling level of the abnormal sign. Specifically, in FIG. 11, the time-series transition of the plot positions is indicated by arrows between the plot positions. In this case, the attention - calling level can be determined according to the direction in which the plot positions transition in time series as viewed from the reference point. For example, if the direction of the transition of each plot position is away from the reference point, the attention - calling level can be set high, and if it is approaching, the attention - calling level can be set low. In this embodiment, the direction in which the plot position transitions away from the reference point corresponds to the direction in which the conversion time and the value equivalent to the maximum torque increase. Note that the determination of the attention - calling level based on the time - series transition of these plot positions may be made for one transition direction between two plot positions, or for the average of a plurality of transition directions between three or more plot positions.

[0075] Alternatively, based on the change in the interval between the plot positions in time series, the attention - calling level of the abnormal sign can also be determined. Specifically, since the distance, which is the interval between the plot positions, corresponds to the amount of change in the conversion time and the value equivalent to the maximum torque during the time elapsed between the corresponding conversion operations, the longer the distance between the plot positions, the greater the change in the degree of the abnormal sign of the transfer mechanism 9 between the conversion operations. Therefore, a determination can be made such that the longer the distance between the plot positions, the higher the attention - calling level. And if the distance between the plot positions is changing in the direction of increasing, it indicates that the degree of the abnormal sign is increasing, so a determination can be made to increase the attention - calling level.

[0076] Furthermore, the attention - calling level of the abnormal sign may be determined based on both the time - series transition of the plot positions corresponding to each of the multiple conversion operations and the change in the interval between the plot positions in time series. For example, among the attention - calling levels determined based on each, the higher attention - calling level can be adopted.

[0077] Note that the determination of the warning level based on the distance between plot positions may be made based on the distance between two plot positions, or may be made based on the average of the distances between three or more plot positions. Also, the distance between plot positions may be the Euclidean distance or the Mahalanobis distance. It is more preferable to use more prognostic parameters using the Mahalanobis distance. In that case, in addition to the two prognostic parameters characterized by the maximum torque equivalent value and the conversion time during the stroke period, for example, it can be based on prognostic parameters characterized by environmental factors such as the average value of the torque equivalent value during the stroke period and the temperature and humidity during the conversion operation.

[0078] [Functional Configuration] FIG. 12 is a diagram showing an example of the functional configuration of the data collection device 3 and the central device 5 that constitute the abnormality detection device 1. According to FIG. 12, the data collection device 3 includes an operation unit 102, a display unit 104, a communication unit 106, a processing unit 200, and a storage unit 300, and can be configured as a kind of computer system.

[0079] The operation unit 102 is realized by an input device such as a button switch, a touch panel, or a keyboard, and outputs an operation signal corresponding to the performed operation to the processing unit 200. The display unit 104 is realized by a display device such as an LCD (Liquid Crystal Display) or a touch panel, and performs various displays according to the display signal from the processing unit 200. The communication unit 106 is realized by a wired or wireless communication device, and communicates with an external device such as the central device 5 via the communication network N.

[0080] The processing unit 200 is implemented by an arithmetic unit such as a CPU (Central Processing Unit), and based on programs, data, etc. stored in the storage unit 300, it gives instructions to and transfers data to each part constituting the data collection device 3, and performs overall control of the data collection device 3. Further, the processing unit 200 includes a voltage / current acquisition unit 202, a torque equivalent value calculation unit 204, and a parameter value calculation unit 206 as functional units according to this embodiment. However, these functional units can also be configured as independent arithmetic circuits by an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or the like.

[0081] The voltage / current acquisition unit 202 acquires motor drive information, which is the motor voltage and motor current of the induction motor 7 of the corresponding turntable 9. Specifically, during the conversion operation of the corresponding turntable 9, the motor voltage measured by a voltage sensor and the motor current measured by a current sensor at a predetermined interval are sampled, A / D converted, and acquired as digital values, so that the time-series data of the motor voltage and motor current are acquired as motor drive information.

[0082] The torque equivalent value calculation unit 204 calculates the torque equivalent value based on the mutual inductance of the induction motor 7 from the motor voltage and motor current of the induction motor 7 related to the turntable 9, which is a railway facility that returns to the stopped state again after performing a conversion operation, which is a specified operation, from the stopped state by performing power supply control to the induction motor 7, thereby calculating the time-series data of the torque equivalent value during the specified operation.

[0083] Specifically, the time-series data of the torque equivalent value during the conversion operation is calculated from the time-series data of the motor voltage and motor current, which are the motor drive information during the conversion operation acquired by the voltage / current acquisition unit 202, according to Equation (1).

[0084] The parameter value calculation unit 206 calculates, as parameters, N (N ≧ 2) feature amounts based on the time-series data, and calculates the feature values of the respective feature amounts as parameter values. Further, the parameter values of each parameter are calculated using, as the feature values, at least two of the maximum value of the torque equivalent value in a predetermined period after the motor start period, the average value of the torque equivalent value in the predetermined period, and the operation time related to the conversion operation.

[0085] Specifically, based on the time-series data of the torque equivalent value during the conversion operation of the tumbler 9 calculated by the torque equivalent value calculation unit 204, the conversion time of the conversion operation is calculated as one of the omen parameters. Further, the stroke period during the conversion operation, which is a predetermined period after the motor start period, is determined, and the maximum value (maximum torque equivalent value) of the torque equivalent value in this stroke period is calculated as another omen parameter (see FIG. 8). Note that the feature values used as the omen parameters are not limited to the two combinations of the conversion time of the conversion operation and the maximum value of the torque equivalent value in the stroke period. For example, the two combinations may be the conversion time of the conversion operation and the average value of the torque equivalent value in the stroke period, or the two combinations may be the maximum value of the torque equivalent value in the stroke period and the average value of the torque equivalent value in the stroke period, or the three combinations may be the conversion time of the conversion operation, the maximum value, and the average value of the torque equivalent value in the stroke period.

[0086] The storage unit 300 is realized by a storage device such as a hard disk, a ROM (Read Only Memory), or a RAM (Random Access Memory), and stores programs, data, etc. for the processing unit 200 to integrally control the data collection device 3. Further, the storage unit 300 is used as a work area for the processing unit 200, and temporarily stores calculation results executed by the processing unit 200 according to various programs, input data via the operation unit 102 or the communication unit 106, etc. In the present embodiment, the tumbler management data 310 related to the corresponding tumbler 9 is stored in the storage unit 300.

[0087] FIG. 13 is a diagram showing an example of the tumbler machine management data 310. According to FIG. 13, the tumbler machine management data 310 stores the tumbler machine ID of the corresponding tumbler machine 9 and the conversion data for each conversion operation. The conversion data includes motor drive information, torque equivalent value data, the conversion direction of the conversion operation, and parameter value data. The motor drive information is the time-series data of the motor voltage and motor current during the conversion operation acquired by the voltage / current acquisition unit 202. The torque equivalent value data is the time-series data of the torque equivalent value during the conversion operation calculated by the torque equivalent value calculation unit 204. The parameter value data is the parameter of each omen parameter calculated by the parameter value calculation unit 206, and includes the conversion time of the conversion operation and the maximum torque equivalent value during the stroke period.

[0088] Returning to FIG. 12, the central device 5 includes an operation unit 402, a display unit 404, a communication unit 406, a processing unit 500, and a storage unit 600, and can be configured as a kind of computer system.

[0089] The operation unit 402 is realized by an input device such as a button switch, a touch panel, a keyboard, etc., and outputs an operation signal corresponding to the performed operation to the processing unit 500. The display unit 404 is realized by a display device such as an LCD (Liquid Crystal Display) or a touch panel, and performs various displays according to the display signal from the processing unit 500. The communication unit 406 is realized by a wired or wireless communication device, and communicates with an external device such as the data collection device 3 via the communication network N.

[0090] The processing unit 500 is implemented by an arithmetic unit such as a CPU (Central Processing Unit), and based on programs, data, etc. stored in the storage unit 600, it gives instructions to and transfers data to each part constituting the central unit 5, and performs overall control of the central unit 5. Further, the processing unit 500 has a determination unit 502 as a functional unit according to this embodiment. However, these functional units can also be configured as independent arithmetic circuits by an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or the like.

[0091] The determination unit 502 is an N-dimensional coordinate system with each parameter as a coordinate axis, and in a coordinate system in which a reference point for regarding the turntable 9, which is railway equipment, as normal is predetermined, it obtains the plot position corresponding to the parameter value calculated by the parameter value calculation unit 206 of the data collection device 3, and based on the positional relationship between the reference point and the plot position, determines the presence or absence of an abnormal sign of the turntable 9. For example, based on the direction of the plot position with respect to the reference point in the coordinate system, it is determined that there is an abnormal sign in the turntable 9. Also, based on the positional relationship between the reference point and the plot position in the coordinate system, it determines the attention call level of the abnormal sign. For example, based on the transition of the plot positions corresponding to each of a plurality of specified operations in chronological order, it determines the attention call level. Also, based on the change in the interval between the plot positions corresponding to each of a plurality of specified operations in chronological order, it determines the attention call level.

[0092] Specifically, in a two-dimensional coordinate system with the conversion time and the value equivalent to the maximum torque, which are omen parameters, as the coordinate axes, based on the positional relationship between the plot position where each parameter value received from the data collection device 3 is plotted and the reference position, it is determined whether there is an abnormal omen in the turning machine 9. In the two-dimensional coordinate system, a normal range including a reference point considering the turning machine 9 to be normal, an abnormal omen range considering that an abnormal omen has occurred in the turning machine 9, and a maintenance / failure range considering the turning machine 9 to be in a maintenance or failure state are defined, and the state represented by the range where the plot position is located is determined as the state of the turning machine 9. That is, if the plot position is within the abnormal omen range, it is determined that there is an abnormal omen in the turning machine 9, and if it is within the normal range, it is determined that there is no (normal) abnormal omen in the turning machine 9 (see Fig. 9).

[0093] Also, by dividing and setting the abnormal omen range in the two-dimensional coordinate system into a plurality of ranges, it is also possible to determine the attention level of the abnormal omen (see Fig. 10).

[0094] Further, based on the plot positions for each of a plurality of conversion operations, it is also possible to determine whether there is an abnormal omen in the turning machine 9. For example, the presence or absence of an abnormal omen is determined based on the ratio of the plot positions located within the abnormal omen range among a plurality of plot positions, such as determining that there is an abnormal omen in the turning machine 9 when the ratio is equal to or greater than a predetermined ratio (see Fig. 11).

[0095] Further, it is also possible to determine the attention - calling level of an abnormality sign based on the transition of plot positions corresponding to a plurality of conversion operations in chronological order. For example, if the direction of the transition of each plot position is away from the reference point, the attention - calling level is increased, and if it is approaching, the attention - calling level is decreased. That is, the attention - calling level is determined according to the direction in which the plot position transitions in chronological order as viewed from the reference point. Also, for example, it is also possible to determine the attention - calling level of an abnormality sign based on the change in the interval between plot positions corresponding to each of the plurality of conversion operations in chronological order, such as increasing the attention - calling level as the distance between plot positions becomes longer (see FIG. 11). In this case, the interval between plot positions can be obtained using the Mahalanobis distance based on the plot positions in the N - dimensional coordinate system.

[0096] The storage unit 600 is realized by a storage device such as a hard disk, ROM (Read Only Memory), or RAM (Random Access Memory), and stores programs, data, etc. for the processing unit 500 to integrally control the central device 5. Also, the storage unit 600 is used as a work area for the processing unit 500, and temporarily stores calculation results executed by the processing unit 500 according to various programs, input data via the operation unit 402 and the communication unit 406, etc. In the present embodiment, the storage unit 600 stores abnormality - detection management data 610 for managing the abnormality detection of each tilting machine 9.

[0097] FIG. 14 is a diagram showing an example of the abnormality - detection management data 610. According to FIG. 14, the abnormality - detection management data 610 is prepared for each tilting machine 9, and one abnormality - detection management data 610 stores abnormality - sign determination data for each conversion operation in association with the tilting - machine ID of the corresponding tilting machine 9. One abnormality - sign determination data includes the conversion direction of the conversion operation received from the data - collection device 3, parameter - value data, an abnormality - sign determination result, and motor - drive information. The abnormality - sign determination result is the determination result of the presence or absence of an abnormality sign by the determination unit 502. The motor - drive information is stored only when received from the data - collection device 3.

[0098] [Process Flow] FIG. 15 is a flowchart showing an example of the process flow of the abnormality detection process performed by the abnormality detection device 1. This process is related to one conversion operation of the tilting machine 9. Also, the process of the data collection device 3 is shown on the left side, and the process of the central device 5 is shown on the right side.

[0099] First, in the data collection device 3, the voltage / current acquisition unit 202 acquires motor drive information, which is time-series data of the motor voltage and motor current related to one conversion operation of the tilting machine 9 (step S1). Next, the torque equivalent value calculation unit 204 calculates time-series data of the torque equivalent value according to formula (1) based on the acquired motor drive information (step S3). Subsequently, the parameter value calculation unit 206 calculates the conversion time of the conversion operation of the tilting machine 9 based on the calculated time-series data of the torque equivalent value (step S5), and also calculates the maximum value (maximum torque equivalent value) of the torque equivalent value during the stroke period (step S7). Then, the calculated conversion time and maximum torque equivalent value, which are the parameter values of the predicted parameter, are transmitted to the central device 5 (step S9).

[0100] Then, in the central device 5, the determination unit 502 calculates the plot position in the coordinate system of the parameter value received from the data collection device 3 (step S11), and determines the presence or absence of an abnormality prediction of the corresponding tilting machine 9 based on the positional relationship between the reference point and the plot position in the coordinate system (step S13). As a result, if it is determined that there is an abnormality prediction (step S15: YES), a request is sent to the data collection device 3 to transmit the motor drive information for the corresponding conversion operation (step S17). Then, the data collection device 3 transmits the requested motor drive information to the central device 5 (step S19). The abnormality detection process is performed in this way.

[0101] [Function and Effect] According to the present embodiment, it is possible to more easily detect an abnormality in the turntable 9, which is railway equipment operated by the induction motor 7. That is, any abnormality that occurs during the switching operation of the turntable 9 appears as a change in the torque of the induction motor 7 that operates the turntable 9. Therefore, it is possible to determine a sign of an abnormality occurring in the turntable 9 based on the time-series data of the torque equivalent value of the induction motor 7.

[0102] Since the parameter values of the respective parameters calculated based on the time-series data of the torque equivalent value represent the characteristics of the time-series data, the positional relationship between the reference point and the plot position in the coordinate system corresponds to the difference between the time-series data of the torque equivalent value during normal operation of the railway equipment and the time-series data of the torque equivalent value to be determined. As a result, it is possible to determine the presence or absence of an abnormality sign by a relatively simple process based on the positional relationship between the plot position where the parameter values of a plurality of sign parameters are plotted and the reference point. Therefore, the amount of calculation required for abnormality detection can be reduced, and the processing load can be reduced. Further, since a plurality of characteristic values are comprehensively determined, it is possible to accurately determine the presence or absence of an abnormality sign.

[0103] In addition, from the data collection device 3 to the central device 5, it is only necessary to transmit the parameter values of a plurality of sign parameters as data related to the switching operation. Therefore, the amount of communication data between the central device 5 and each data collection device 3 can be reduced. Then, the central device 5 can determine the presence or absence of an abnormality sign of the turntable 9 with a light processing load based on the received parameter values. For this reason, the abnormality detection device 1 can be suitably realized as a system in which the central device 5 remotely monitors a large number of turntables 9 in a centralized manner.

[0104] It should be noted that the applicable embodiments of the present invention are not limited to the above-described embodiments, and it goes without saying that they can be appropriately changed without departing from the spirit of the present invention.

[0105] (A) Torque equivalent value In the above-described embodiment, the torque equivalent value of the induction motor 7 is set to the reciprocal of the mutual inductance M (see Equation (1)), but the mutual inductance M itself may be used as the torque equivalent value. Further, if the motor characteristics of the induction motor 7 (the relationship between torque with respect to motor voltage and motor current) are known, the torque equivalent value may be converted into the actual torque value and used by comparing the torque based on the motor characteristics with the torque equivalent value.

[0106] (B) Configuration of the abnormality detection device 1 Instead of the data collection device 3, the central device 5 may be configured to calculate the parameter values of the omen parameters. Specifically, as shown in FIG. 16, the central device 5A is configured to have the function of the parameter value calculation unit 206.

[0107] Also, in the above-described embodiment, the abnormality detection device 1 is configured such that the data collection device 3 and the central device 5 are communicatively connected via the communication network N, but it may be configured by a single device installed in association with or in the vicinity of the tumbler 9. Specifically, as shown in FIG. 17, the abnormality detection device 1C is realized as a single computer system, and has the functions of the voltage / current acquisition unit 202, the torque equivalent value calculation unit 204, and the parameter value calculation unit 206 included in the data collection device 3, and the function of the determination unit 502 included in the central device 5.

[0108] (C) Railway facilities In the above-described embodiment, the railway facilities have been described as the tumbler 9, but the above embodiment is also applicable to other railway facilities. For example, railway facilities such as a level crossing barrier and a home gate that perform a specified operation from a stopped state and then return to the stopped state by controlling the power supply to the induction motor 7 can be applied in the same manner as the above-described embodiment.

Explanation of reference numerals

[0109] 1, 1A, 1C... Abnormality detection device 3, 3A... Data collection device 200, 200A... Processing unit 202…Voltage and current acquisition unit 204…Torque equivalent value calculation unit 206…Parameter value calculation unit 300…Memory unit 310…Rotary converter management data 5,5A…Central unit 500,500A…Processing unit 502…Judgment unit 600…Memory unit 610…Abnormality detection management data 7…Induction motor 9…Rotary converter

Claims

1. By performing power supply control to an induction motor, a torque equivalent value based on the mutual inductance of the induction motor is calculated from the motor voltage and motor current of the induction motor in railway equipment that returns to a stopped state after performing a specified operation from a stopped state, thereby calculating time-series data of the torque equivalent value during the specified operation; a torque equivalent value calculating means; Parameter value calculating means for calculating, as parameters, a plurality of feature amounts based on the time-series data and calculating, as parameter values, the feature values of the respective feature amounts; A coordinate system having the respective parameters as coordinate axes, in a coordinate system in which a reference point assuming the railway equipment to be normal is predetermined, obtaining a plot position corresponding to the parameter value calculated by the parameter value calculating means, and based on the positional relationship between the reference point and the plot position, determining means for determining the presence or absence of an abnormal sign of the railway equipment; An abnormality detection device comprising:

2. The parameter value calculating means uses, as the feature values, at least two of the maximum value of the torque equivalent value during a predetermined period after the motor start period, the average value of the torque equivalent value during the predetermined period, and the operation time related to the specified operation. The abnormality detection device according to claim 1.

3. The determination means determines the presence or absence of an abnormal sign of the railway equipment based on the direction of the plot position with respect to the reference point in the coordinate system. The abnormality detection device according to claim 1 or 2.

4. Warning level determination means for determining a warning level of an abnormal sign based on the positional relationship; The abnormality detection device according to any one of claims 1 to 3, further comprising:

5. The warning level determination means determines the warning level based on the transition of the plot positions corresponding to a plurality of times of the specified operation in time series order. The abnormality detection device according to claim 4.

6. The warning level determination means determines the warning level based on a change in the interval between the plot positions corresponding to a plurality of times of the specified operation in time series order. The abnormality detection device according to claim 4.

7. The warning level determination means obtains the interval between the plot positions using the Mahalanobis distance based on the plot positions on the coordinate system, and determines the warning level based on the change in the interval. The abnormality detection device according to claim 6.

8. The railway equipment is any one of a tumbler, a switch disconnector, and a home gate. The abnormality detection device according to any one of claims 1 to 7.

9. A data collection device having the torque equivalent value calculation means and the parameter value calculation means, and acquiring the motor voltage and motor current from the railway facility; A central device communicatively connected to the data collection device via a communication network, the central device having the determination means; The abnormality detection device according to any one of claims 1 to 8, comprising:

10. The central device of the abnormality detection device including a data collection device and a central device, The data collection device calculates a torque equivalent value based on the mutual inductance of the induction motor from the motor voltage and motor current of the induction motor related to railway facilities that perform a specified operation from a stopped state and then return to a stopped state by controlling power supply to the induction motor, thereby calculating time-series data of the torque equivalent value during the specified operation, a parameter value calculation means for calculating, as parameters, a plurality of feature amounts based on the time-series data and calculating, as parameter values, the feature values of the respective feature amounts, Receiving means for receiving the parameter values calculated by the parameter value calculation means from the data collection device; In a coordinate system with each parameter as a coordinate axis, in a coordinate system in which a reference point for considering the railway facility to be normal is predetermined, obtaining a plot position corresponding to the parameter value received by the receiving means, and determining the presence or absence of an abnormality sign of the railway facility based on the positional relationship between the reference point and the plot position; A central device comprising:

11. Calculating time-series data of the torque equivalent value during the specified operation by calculating a torque equivalent value based on the mutual inductance of the induction motor from the motor voltage and motor current of the induction motor related to railway facilities that perform a specified operation from a stopped state and then return to a stopped state by controlling power supply to the induction motor; Calculating, as parameters, a plurality of feature amounts based on the time-series data and calculating, as parameter values, the feature values of the respective feature amounts; In a coordinate system with each parameter as a coordinate axis, in a coordinate system in which a reference point for considering the railway facility to be normal is predetermined, obtaining a plot position corresponding to the calculated parameter value, and determining the presence or absence of an abnormality sign of the railway facility based on the positional relationship between the reference point and the plot position; An abnormality detection method including:

Images