Switch machine status monitoring device, switch machine status monitoring method, and program

Abstract

The motor current is measured to detect the switching operation process and determine the state at each stage of the switching operation. [Solution] The switch machine status monitoring device 31 includes a teacher information acquisition unit 43 that acquires the motor current waveform when a reference electric switch machine is driven and the switching operation process corresponding to this current waveform; a current waveform acquisition unit 41 that acquires a monitoring current waveform, which is the motor current waveform when the electric switch machine to be monitored is driven; a current waveform comparison processing unit 44 that compares the two current waveforms and detects corresponding points in each; a switching operation process identification processing unit 45 that identifies the switching operation process in the monitoring current waveform based on the corresponding points detected by the current waveform comparison processing unit 44; and a status determination processing unit 46 that determines the status of the electric switch machine to be monitored for each switching operation process identified by the switching operation process identification processing unit 45.

Description

[Technical Field]

[0001] The present invention relates to a switch machine condition monitoring apparatus, a switch machine condition monitoring method, and a program. [Background Art]

[0002] Devices that monitor the operating state of electric switch machines are used to detect signs before the operation stops and becomes unable to switch, and to estimate the cause after a failure to switch occurs, when the electric switch machine drives to switch the movable part of a switch (such as tongue rails). This device is important for preventing impacts on scheduled train operation, such as disruption to train operation caused by failure to switch and prolonged recovery time.

[0003] As a method for monitoring the condition of an electric switch machine and detecting abnormalities, a method of measuring and monitoring the current and voltage of the motor that drives the electric switch machine (see, for example, Patent Document 1) is widely prevalent. [Prior Art Documents] [Patent Documents]

[0004] [Patent Document 1] Japanese Patent No. 2612759 [Summary of the Invention] [Problem to be Solved by the Invention]

[0005] By the way, the switching operation process of a point apparatus using an electric switch machine is broadly divided into three phases: (1) an unlocking phase, (2) a switching phase, and (3) a locking phase. Different contents of abnormal conditions and failure factors affect different switching operation processes. For example, if a foreign object is present during movement of the switch, an impact appears in (2) the switching phase, and if the locking mechanism cannot be seated due to lock displacement during locking, an impact appears in (3) the locking phase.

[0006] However, the technology described in Patent Document 1 above compares the peak value, peak position, average value, operating time, and number of operations of the motor current with threshold values ​​for determining abnormalities. If there are no sensors that measure other data, such as operating deviation, in addition to measuring the motor current and voltage, it is not possible to determine which phase the abnormal condition is occurring in.

[0007] The present invention has been made in view of the above circumstances, and aims to provide a switch machine status monitoring device, a switch machine status monitoring method, and a program that can detect the switching operation process by measuring the motor current and determine the state for each switching operation process. [Means for solving the problem]

[0008] One aspect of the switch machine status monitoring device according to the present invention is characterized by comprising: a first acquisition unit that acquires a first current waveform of the motor of a first electric switch machine and a switching operation process of the first electric switch machine corresponding to the first current waveform when the first electric switch machine, which is a reference, is driven; a second acquisition unit that acquires a second current waveform, which is the motor current waveform when the second electric switch machine, which is the target of monitoring, is driven; a current waveform comparison processing unit that compares the first current waveform and the second current waveform and detects corresponding points in the first current waveform and the second current waveform, respectively; an identification unit that identifies a switching operation process in the second current waveform based on the corresponding points in the first current waveform and the second current waveform, respectively, detected by the current waveform comparison processing unit; and a determination unit that determines the state of the second electric switch machine for each switching operation process identified by the identification unit.

[0009] One aspect of the switch machine status monitoring method according to the present invention is characterized by including: a first acquisition step of acquiring a first current waveform of the motor of a first electric switch machine and a switching operation process of the first electric switch machine corresponding to the first current waveform when a reference first electric switch machine is driven; a second acquisition step of acquiring a second current waveform, which is the motor current waveform when a second electric switch machine to be monitored is driven; a current waveform comparison step of comparing the first current waveform and the second current waveform and detecting corresponding points in the first current waveform and the second current waveform, a specification step of identifying a switching operation process in the second current waveform based on the corresponding points in the first current waveform and the second current waveform detected by the processing of the current waveform comparison step, and a determination step of determining the state of the second electric switch machine for each switching operation process identified by the processing of the specification step.

[0010] One aspect of the program according to the present invention is a process in which a computer is made to execute a process that includes: a first acquisition step of acquiring a first current waveform of the motor of a first electric point machine and a switching operation process of the first electric point machine corresponding to the first current waveform when a reference first electric point machine is driven; a second acquisition step of acquiring a second current waveform, which is the motor current waveform when a second electric point machine that is under monitoring is driven; a current waveform comparison step of comparing the first current waveform and the second current waveform and detecting corresponding points in the first current waveform and the second current waveform, respectively; an identification step of identifying a switching operation process in the second current waveform based on the corresponding points in the first current waveform and the second current waveform detected by the processing of the current waveform comparison step; and a determination step of determining the state of the second electric point machine for each switching operation process identified by the processing of the identification step. [Effects of the Invention]

[0011] According to the present invention, it is possible to provide a switch machine status monitoring device, a switch machine status monitoring method, and a program that can detect the switching operation process by measuring the motor current and determine the state for each switching operation process. [Brief explanation of the drawing]

[0012] [Figure 1] Figure 1 is a plan view showing the structure of a railway switch equipped with an electric switch. [Figure 2] Figure 2 is a schematic diagram illustrating the basic structure of an electric point machine. [Figure 3] Figure 3 is a diagram illustrating the operation of the cam mechanism of an electric point machine. [Figure 4] Figure 4 is a diagram illustrating the operation of the cam mechanism of an electric point machine. [Figure 5] Figure 5 is a diagram illustrating the operation of the cam mechanism of an electric point machine. [Figure 6] Figure 6 is a diagram illustrating the operation of the cam mechanism of an electric point machine. [Figure 7] Figure 7 is a functional block diagram showing the functional configuration of the switch machine status monitoring device 31 according to this embodiment. [Figure 8] Figure 8 is a diagram illustrating the correct answer data. [Figure 9] Figure 9 is a diagram illustrating the monitoring current waveform. [Figure 10] Figure 10 is a diagram illustrating the comparison of waveforms. [Figure 11] Figure 11 is a diagram illustrating the detection of the transition process. [Figure 12] Figure 12 is a flowchart illustrating the electric point machine monitoring process. [Figure 13] Figure 13 is a flowchart illustrating the monitoring process for electric point machines. [Modes for carrying out the invention]

[0013] The switch machine condition monitoring device according to this embodiment will be described below with reference to the drawings.

[0014] [Turn switch] Fig. 1 is a plan view showing the structure of a switch 1. The switch 1 is a device that branches one track into two or more tracks to change the traveling route of a vehicle. The switch 1 includes an electric point machine whose state is monitored by the point machine state monitoring device according to the present embodiment.

[0015] That is, the switch 1 is configured to include a tongue rail 2, a stock rail 3, a floor plate 4, a switch rod 5, a front rod 6, a connecting rod 7, a switch adjuster 8, a jaw pin 8a, a connecting plate 9, and an electric point machine 10.

[0016] Strictly speaking, the constituent members of the switch 1 are the tongue rail 2, the stock rail 3, the floor plate 4, the switch rod 5, and the connecting plate 9. The remaining front rod 6, connecting rod 7, switch adjuster 8, and jaw pin 8a are signal members used in combination with the electric point machine 10. The present embodiment aims to determine the state of the electric point machine 10, and external faults are highly related to the switch 1, so for convenience, the signal members will be described as part of the switch 1.

[0017] When switching the track, the tongue rail 2 changes its position in the direction of arrow A or arrow B in Fig. 1, which is a direction perpendicular to the traveling direction of the vehicle. The stock rail 3 is a main rail continuously connected from the track outside the switch 1, and when the tongue rail 2 switches position, the tip end 2a of the tongue rail 2 comes into close contact with or separates from the stock rail 3. The floor plate 4 switchably supports the tongue rail 2 and supports the load of a vehicle traveling on the stopped tongue rail 2.

[0018] The switch rod 5 interlocks with the operating rod 15 of the electric point machine 10 to switch the position of the tongue rail 2 and press the tongue rail 2 against the stock rail 3. The front rod 6 is fixed between the tip ends 2a of the tongue rail 2, and fixes the spacing between the tip ends 2a. The connecting rod 7 is connected between the front rod 6 and the locking rod 16 of the electric point machine 10, and operates in conjunction with the locking rod 16.

[0019] The switch adjuster 8 is connected to the operating rod 15 and is an adjuster that sets the contact force when the tongue rail 2 is switched and comes into contact with the base rail 3. The jaw pin 8a is a pin that connects the operating rod 15 and the switch adjuster 8. The connecting plate 9 is a plate-shaped member that connects the tongue rail 2 and the turn bar 5 at the position where they intersect.

[0020] The electric point machine 10 switches the tongue rail 2 using the operating lever 15 and checks the opening direction with the locking lever 16. The structure of the electric point machine 10 is described below.

[0021] [Electric points machine] Figure 2 is a schematic diagram illustrating the structure of the electric point machine 10. In Figure 2, to clarify the drawing, the components that perform rotational motion are shown as perspective views at the top of the page, and the components that perform linear motion are shown as top views from above the axial direction of the switching roller 18, which will be described later, at the bottom of the page (within the dashed lines).

[0022] The electric point machine 10 comprises a motor 11, a clutch 12, a reduction mechanism 13, a cam mechanism 14, an operating lever 15, a locking lever 16, and a motor control unit 17.

[0023] Motor 11 is the driving force source for the electric point machine 10. The current value of motor 11 is measured by an ammeter (not shown), and the current waveform is supplied to the point machine condition monitoring device, which will be described later.

[0024] The clutch 12 is interposed between the motor 11 and the reduction mechanism 13. When excessive torque is input from the motor 11 or the rotational load on the downstream reduction mechanism 13 becomes excessive, multiple sliding plates slide against each other to protect the reduction mechanism 13. Here, a friction clutch is used as an example of the clutch 12, but other types of clutch 12 may be used as long as they can prevent excessive torque input from the motor 11.

[0025] The reduction mechanism 13 increases the rotational driving force by reducing the rotational speed of the motor 11 by continuously transmitting the rotation of the motor 11 to multiple gears with different gear ratios. In Figure 2, reduction is performed by switching between two speeds, but instead, a reduction mechanism 13 that switches between one speed or three or more speeds may be used.

[0026] The cam mechanism 14 comprises a switching roller 18 provided on the final gear 20 of the reduction mechanism 13, and a pair of lock pieces 19 that move in accordance with the movement of the switching roller 18. The switching roller 18 is fitted into an axially parallel hole provided on the final gear 20 and protrudes cylindrically from the lower gear surface. The lock pieces 19 move linearly in accordance with the movement of the switching roller 18, which rotates with the rotation of the final gear 20, so that the cam surface 19a provided on the inner circumference does not move away from the switching roller 18. Thus, the rotational motion of the final gear 20 is converted into linear motion of the lock pieces 19 by the cam mechanism 14. The final gear 20 rotates and moves the switching roller 18 to switch the operating lever 15, and is therefore hereinafter referred to as the "switching gear 20".

[0027] The operating lever 15 is a rod-shaped member having a notch 15a that forms its inner circumferential surface. The switching roller 18 slides along the notch 15a, moving over a predetermined stroke and driving the tongue rail 2 in the turnout 1 relative to the base rail 3 in the direction of arrow A or arrow B in Figure 1. When the operating lever 15 is driven, it engages with the locking portion 19c of the lock piece 19 (described later) and functions as the main locking mechanism.

[0028] The locking pin 16 has a notch 16a on the side opposite to the side where the cam surface 19a of the lock piece 19 is provided, through which the lock piece 19 is inserted and removed. After the tongue rail 2 is rotated, the lock piece 19 is inserted into the notch 16a to check the tip position of the tongue rail 2. In addition, the locking pin 16 functions as a backup lock to prevent the displacement of the operating pin 15 when the primary locking function of the operating pin 15 is lost.

[0029] The motor control unit 17 controls the rotational driving force of the motor 11 by supplying a predetermined current to the motor 11 when the electric point machine 10 is in operation.

[0030] [Cam mechanism] Next, the operation of the unlocking phase P1, the conversion phase P2, and the locking phase P3 of the cam mechanism 14 will be explained with reference to Figures 3 to 6. Figures 3 to 6 are explanatory diagrams showing the state of the part of Figure 2 excluding the locking lever 16, indicated by the dashed line, rotated 90 degrees to the right. Each figure also includes an explanatory diagram showing the corresponding operating state of the operating lever 15 and the lock piece 19.

[0031] As shown in Figures 3 to 6, the pair of lock pieces 19 are laid flat so as to be slidable in the vertical direction of the paper. The lock pieces 19 slide as the conversion roller 18 slides against the cam surface 19a as the conversion gear 20 (Figure 2) rotates. An operating lever 15 that is slidable in the horizontal direction of the paper is positioned on the vertical upper surface (towards the front of the paper) of the lock piece 19.

[0032] The operating lever 15 slides as the conversion roller 18 slides along the notch 15a in conjunction with the rotation of the conversion gear 20 (Figure 2). The lock piece 19 has a locking portion 19c on its upper surface that prevents the operating lever 15 from sliding. When the lock piece 19 slides and comes into contact with the lock receiver 15b of the operating lever 15, the operating lever 15 is prevented from sliding any further.

[0033] The locking portion 19c is provided on the upper surfaces of both of the pair of locking pieces 19. When the locking piece 19 on the left side of the page slides, the locking portion 19c prevents the operating rod 15 from sliding to the reverse side, resulting in a locked state. When the locking piece 19 on the right side of the page slides, the locking portion 19c prevents the operating rod 15 from sliding to the reverse side, resulting in a locked state.

[0034] Figure 3 shows the electric point machine 10 before switching, in the locked position state. In this state, the switching roller 18 is stationary, and the operating lever 15 is locked to the locking portion 19c of the lock piece 19. Therefore, as indicated by the arrows in the figure, both the operating lever 15 and the lock piece 19 are in the stopped state.

[0035] Figure 4 shows the unlocking operation of the electric point machine 10 on the normal position side, i.e., the operation in unlocking phase P1. During the unlocking operation, the switching gear 20 rotates, causing the switching roller 18 to slide on the cam surface 19b of the lock piece 19, and the lock piece 19 slides downward from the plane of the paper. As the lock piece 19 slides, the locking part 19c disengages from the lock receiver 15b of the operating lever 15.

[0036] At this time, the switching roller 18 slides along the escape surface 15c on the normal position side of the notch 15a of the operating lever 15, so the operating lever 15 does not slide. Therefore, in the unlocking operation, as shown by the arrow in the figure, the operating lever 15 remains stationary, but the lock piece 19 slides.

[0037] Figure 5 shows the switching operation of the electric point machine 10, that is, the operation in switching phase P2. In Figure 4, the switching roller 18 slides against the cam surface 19b of the lock piece 19, causing the lock piece 19 to slide downwards from the plane of the paper, and the lock part 19c completely detaches from the lock receiver 15b. In this state, as the switching gear 20 (Figure 2) rotates, the switching roller 18 fits into the switching surface 15e of the notch 15a of the operating rod 15. The switching roller 18 on the switching surface 15e slides the operating rod 15 as it moves. Therefore, in the switching operation, as shown by the arrows in the figure, the operating rod 15 slides, but the lock piece 19 does not.

[0038] Figure 6 shows the locking operation of the reverse side of the electric point machine 10, i.e., the operation in locking phase P3. During the locking operation, the switching gear 20 rotates, causing the switching roller 18 to slide on the cam surface 19a of the lock piece 19, and the lock piece 19 slides upward on the plane of the paper. As the lock piece 19 slides, the locking portion 19c advances toward the lock receiver 15b of the operating rod 15.

[0039] At this time, the switching roller 18 slides along the escape surface 15d on the opposite side of the notch 15a of the operating lever 15, so the operating lever 15 does not slide. Therefore, in the locking operation, as shown by the arrow in the figure, the operating lever 15 remains stationary, but the lock piece 19 slides.

[0040] As shown in Figures 3 to 6 above, the operating lever 15 is initially in a non-sliding state, but the rotation of the switching gear 20 (Figure 2) causes the switching roller 18 to move, which unlocks the lock piece 19 and enables the switching operation. Subsequently, further rotation of the switching gear 20 moves the switching roller 18, causing the lock piece 19 to lock and preventing the operating lever 15 from switching.

[0041] Furthermore, although the locking lever 16 is omitted in Figures 3 to 6, when the operating lever 15 is detached from the locking portion 19c of the lock piece 19 and in the unlocked state, the lock piece 19 is detached from the notch 16a of the locking lever 16. When the operating lever 15 is engaged with the locking portion 19c of the lock piece 19 and in the locked state, the lock piece 19 is in a state where it can be inserted into the notch 16a of the locking lever 16. In other words, when the notch 16a of the locking lever 16 is in a predetermined position, the locking by the lock piece 19 and the verification are synchronized.

[0042] [Point machine status monitoring device] Next, Figure 7 is a functional block diagram showing the functional configuration of the switch machine condition monitoring device 31.

[0043] The switch machine status monitoring device 31 consists of a CPU (Central Processing Unit), memory units (ROM (Read Only Memory), RAM (Random Access Memory), non-volatile memory, etc.), hardware, and other elements. The switch machine status monitoring device 31 functions as a current waveform acquisition unit 41, a teacher information storage unit 42, a teacher information acquisition unit 43, a current waveform comparison processing unit 44, a switching operation process identification processing unit 45, a status determination processing unit 46, and an output processing unit 47 by executing a control application program stored in a memory unit (not shown).

[0044] The current waveform acquisition unit 41 acquires the current waveform of the motor 11 of the electric switch machine 10 whose status is being monitored, and supplies it to the current waveform comparison processing unit 44. The current waveform of the motor 11 acquired by the current waveform acquisition unit 41 will be referred to below as the monitored current waveform.

[0045] The training information storage unit 42 stores in advance a set of data (hereinafter referred to as "correct answer data") which includes information showing the current waveform of the motor 11 of an electric point machine 10 that has successfully determined the switching operation process, and information indicating which phase of the switching operation process the current waveform represents. Note that the waveform of the motor drive current during the switching operation process of an electric point machine takes a similar shape even between different electric point machines. The correct answer data will be described later using Figure 8.

[0046] The teacher information acquisition unit 43 acquires correct answer data or characteristic current waveforms during a malfunction stored in the teacher information storage unit 42 and supplies them to the current waveform comparison processing unit 44.

[0047] The current waveform comparison processing unit 44 compares the monitored current waveform of the motor 11 of the electric switch machine 10, which is the subject of state monitoring and supplied from the current waveform acquisition unit 41, with the correct data supplied from the training information acquisition unit 43, detects the correspondence between points on the corresponding waveforms, and calculates the similarity between the two waveforms.

[0048] Specifically, the current waveform comparison processing unit 44 performs calculations using an algorithm that can detect corresponding points on two waveforms even if the length and period of the time series are different, and calculate the similarity of the waveforms. Examples of algorithms that can calculate similarity even if the length and period of the time series are different include DTW (Dynamic Time Warping) and D-DTW (Derivative-DTW).

[0049] DTW is a method that calculates the distance (absolute value of error) at each point in the time series of two waveforms by brute force, and then finds the path that results in the shortest cumulative distance between the two time series. D-DTW is a method derived from DTW that calculates similarity by focusing on the left and right slopes of measurement points on the waveform, i.e., the difference in the amount of change. However, DTW may perform an alignment that is contrary to intuition for time series data that has locally increasing or decreasing parts with respect to the time axis. For this reason, when the current waveform comparison processing unit 44 performs calculations on the monitored waveform as is, it is preferable to use D-DTW.

[0050] The current waveform comparison processing unit 44 determines that the similarity between the current waveform of the ground truth data and the monitored current waveform is greater than or equal to a predetermined value, i.e., the cost calculated by DTW or D-DTW is lower than a predetermined value, and that the similarity between these waveforms is sufficient for determining the conversion operation process. It then supplies the detection result of the correspondence between each point in the two waveforms, along with the monitored current waveform and the ground truth data, to the conversion operation process identification processing unit 45. The similarity threshold is a value that can be appropriately designed experimentally and empirically.

[0051] Furthermore, if the similarity between the current waveform of the correct data and the monitored current waveform is less than or equal to a predetermined value, that is, if the cost calculated by DTW or D-DTW is higher than a predetermined value, the current waveform comparison processing unit 44 will determine that the conversion operation process cannot be identified from the monitored current waveform using the current waveform of the correct data, and will supply this result to the conversion operation process identification processing unit 45.

[0052] The conversion operation process identification processing unit 45 identifies three conversion operation processes in the monitored current waveform based on the detection results of the correspondence between each point in the current waveform of the correct data and the monitored current waveform, the monitored current waveform, and the correct data, which are supplied from the current waveform comparison processing unit 44 when the similarity between the current waveform of the correct data and the monitored current waveform is greater than or equal to a predetermined value, and supplies these to the state determination processing unit 46. Specific examples of the processing by the current waveform comparison processing unit 44 and the conversion operation process identification processing unit 45 will be described later with reference to Figures 8 to 11.

[0053] The state determination processing unit 46 determines the state in each switching operation process based on the information supplied by the switching operation process identification processing unit 45, using the monitored current waveform. For example, the state determination processing unit 46 compares the duration, average current, and peak voltage in each switching operation process with those in the current waveform of the correct data, and if there is a difference greater than a predetermined value, it can determine that the state of the corresponding switching operation process is abnormal. Specifically, if the numerical difference is large in the switching phase P2, for example, it is possible that a foreign object is interposed during the movement of the turnout 1, and if the numerical difference is large in the locking phase P3, for example, it is possible that the locking mechanism does not settle due to lock displacement during locking.

[0054] The output processing unit 47 performs processing to output the state determination result supplied from the state determination processing unit 46 to a predetermined external output device such as a display device, a printing device, or a speaker.

[0055] [Specific example] Next, with reference to Figures 8 to 11, specific examples of processing by the current waveform comparison processing unit 44 and the conversion operation process identification processing unit 45 will be described. Here, the current waveform comparison processing unit 44 will be described as using D-DTW as an algorithm that can detect corresponding points in each waveform and calculate the similarity even if the length and period of the time series of the two waveforms are different.

[0056] Figure 8 shows an example of correct answer data stored in the teacher information storage unit 42. The correct answer data is a set of data that includes the current waveform α of the electric switch machine motor when it is switched normally, and the determination results of the three switching operation processes: unlocking phase P1, switching phase P2, and locking phase P3 at that time. The current waveform comparison processing unit 44 acquires the correct answer data shown in Figure 8 from the teacher information acquisition unit 43.

[0057] Figure 9 shows an example of the current waveform β, which is the monitored current waveform. The current waveform comparison processing unit 44 acquires the current waveform β shown in Figure 9 from the current waveform acquisition unit 41.

[0058] Referring to Figures 10 and 11, the waveform comparison performed by the current waveform comparison processing unit 44 and the identification of the conversion operation process by the conversion operation process identification processing unit 45 will be explained. In Figure 10, in order to clearly illustrate the relationship between corresponding points in the two waveforms, the values ​​on the vertical axes of current waveform α and current waveform β are shown as being different.

[0059] The current waveform comparison processing unit 44 calculates D-DTW for current waveforms α and β to determine corresponding points and scores in the two waveforms. In Figure 10, only the points necessary for identifying the conversion process are shown, but the current waveform comparison processing unit 44 detects the correspondence between all points at a predetermined sampling rate. Furthermore, in the two waveforms in Figure 10, the D-DTW score is below a predetermined value, indicating a sufficient similarity to determine the phase.

[0060] The current waveform comparison processing unit 44 supplies the detection result of the correspondence between the points of the two waveforms, along with the monitored current waveform and the correct answer data, to the conversion operation process identification processing unit 45.

[0061] The conversion operation process identification processing unit 45 determines points p', q', r', and s' in the current waveform β of the observed current, corresponding to the start point p of the unlocking phase P1, the transition point q from the unlocking phase P1 to the conversion phase P2, the transition point r from the conversion phase P2 to the locking phase P3, and the end point s of the locking phase P3, respectively, in the current waveform α. As shown in Figure 10, assuming that point p in the current waveform α and point p' in the current waveform β are at approximately the same time, points q', r', and s' in the current waveform β occur at an earlier time than points q, r, and s in the current waveform α. In other words, the lengths of the time series of the two waveforms are different.

[0062] Then, as shown in Figure 11, the conversion operation process identification processing unit 45 determines the ranges of the unlocking phase P1, conversion phase P2, and locking phase P3 in the observed current waveform β based on points p', q', r', and s' of the observed current waveform β, and supplies the determination result and the measured current waveform to the state determination processing unit 46.

[0063] Thus, as shown in Figure 10, even if the magnitude and shape of the current waveform α of the electric switch machine's motor differ from the current waveform β, which is the monitored current waveform, by correlating the characteristic points of the two, the switching operation process of the observed current waveform can be identified.

[0064] [Electric Point Machine Monitoring Process] Next, the electric switch machine monitoring process performed by the switch machine status monitoring device 31 will be described with reference to the flowcharts in Figures 12 and 13.

[0065] In step S1, the current waveform acquisition unit 41 acquires the monitored current waveform as explained with reference to Figure 9 and supplies it to the current waveform comparison processing unit 44.

[0066] In step S2, the teacher information acquisition unit 43 acquires the correct answer data, as explained using Figure 8, from the teacher information stored in the teacher information storage unit 42, and supplies it to the current waveform comparison processing unit 44.

[0067] In step S3, the current waveform comparison processing unit 44 compares the monitored current waveform acquired by the current waveform acquisition unit 41 with the correct data supplied by the training information acquisition unit 43, detects corresponding points in the two waveforms as explained with reference to Figure 10, and calculates the similarity.

[0068] In step S4, the current waveform comparison processing unit 44 determines whether the similarity is equal to or greater than a predetermined value. If it is determined in step S4 that the similarity is not equal to or greater than a predetermined value, the process proceeds to step S7, which will be described later.

[0069] In step S4, if it is determined that the similarity is equal to or greater than a predetermined value, in step S5, the current waveform comparison processing unit 44 supplies the detection result of the correspondence between each point of the two waveforms, along with the monitored current waveform and the correct data, to the conversion operation process identification processing unit 45. Based on the correspondence between the measurement points, the conversion operation process is identified as explained with reference to Figure 11, and this process is supplied to the state determination processing unit 46.

[0070] In step S6, the state determination processing unit 46 determines the state in each conversion operation process based on the information supplied by the conversion operation process identification processing unit 45, and supplies the determination result to the output processing unit 47. After the completion of the processing in step S6, the process proceeds to step S12.

[0071] If, in step S4, it is determined that the similarity is not equal to or greater than a predetermined value, then in step S7, the current waveform comparison processing unit 44 controls the teacher information acquisition unit 43 to acquire a characteristic current waveform at the time of failure from the teacher information stored in the teacher information storage unit 42.

[0072] In step S8, the current waveform comparison processing unit 44 calculates the similarity between the acquired characteristic current waveform at the time of the fault and the monitored current waveform, and supplies the calculation result to the conversion operation process identification processing unit 45. The conversion operation process identification processing unit 45 supplies the similarity calculation result to the state determination processing unit 46.

[0073] In step S9, the state determination processing unit 46 determines, based on the information supplied by the conversion operation process identification processing unit 45, whether or not there is fault state data whose similarity is equal to or greater than a predetermined value.

[0074] In step S9, if it is determined that there is fault state data with a similarity of a predetermined value or higher, in step S10, the state determination processing unit 46 provides the corresponding fault state as the state determination result to the output processing unit 47.

[0075] If, in step S9, it is determined that there is no fault state data with a similarity of a predetermined value or higher, then in step S11, the state determination processing unit 46 determines that the state is an abnormal state that does not correspond to the fault state information stored in the training information storage unit 42, and supplies this to the output processing unit 47.

[0076] After the completion of the processing in step S6, step S10, or step S11, in step S12, the output processing unit 47 performs processing to output the status determination result to a predetermined external output device, and the processing is completed.

[0077] Through this process, the motor's current waveform can be used to detect the switching operation process, and the state of the motor at each stage of the switching operation can be determined.

[0078] [Example of a different configuration] In the embodiment described above, the teacher information storage unit 42 is assumed to store one correct answer data, but it may store multiple correct answer data. In that case, the current waveform comparison processing unit 44 supplies the detection result of the correspondence between the correct answer data with the highest similarity and each point of the monitored waveform to the conversion operation process identification processing unit 45, along with the monitored current waveform and the correct answer data with the highest similarity.

[0079] This allows for accurate detection of the switching operation process, even when, for example, the waveform of the motor drive current during the switching operation of different electric points differs between individual electric points due to significant differences in their installation conditions.

[0080] Furthermore, the teacher information storage unit 42 can store one or more characteristic current waveforms during a malfunction.

[0081] This allows for the identification of specific malfunctions if the conversion is not normal. [Supplementary explanation of the embodiment] The embodiments described above are all preferred examples of the present invention. The numerical values, components, arrangement positions of components, and connection configurations shown in the embodiments above are examples only and are not intended to limit the present invention. Furthermore, the figures are not necessarily strictly illustrative.

[0082] The series of processes described above can be executed by hardware or by software. When the series of processes are executed by software, the programs that make up the software are installed from a program storage medium on a computer that is built into dedicated hardware, or on a general-purpose computer that can perform various functions by installing various programs.

[0083] The programs executed by the computer may be programs that are processed chronologically in the order described herein, or they may be programs that are processed in parallel or at necessary times, such as when a call is made.

[0084] [Note] The contents described in some of the embodiments above can be understood, for example, as follows:

[0085] (1) Identify the conversion process by comparing waveforms and determine the state for each conversion process. The switch machine status monitoring device 31 includes a teacher information acquisition unit 43 that acquires the first current waveform of the motor of the first electric switch machine when the reference first electric switch machine is driven, and the switching operation process of the first electric switch machine corresponding to the first current waveform (correct data in Figure 8) (step S2), A current waveform acquisition unit 41 acquires a monitoring current waveform, which is the current waveform of the motor 11 when the electric point machine 10, which is the target of monitoring, is driven (step S1), A current waveform comparison processing unit 44 compares the first current waveform with the monitored current waveform and detects corresponding points in the first current waveform and the monitored current waveform (step S3), A current waveform comparison processing unit 44 detects a first current waveform and a monitoring current waveform, and based on corresponding points in each of these, a conversion operation process identification processing unit 45 identifies the conversion operation process in the monitoring current waveform (step S5). For each switching operation process identified by the switching operation process identification processing unit 45, the state determination processing unit 46 determines the state of the electric point machine 10 (step S6) It holds.

[0086] This allows the motor's current waveform to be used to detect the switching process and determine the state at each stage of the switching process.

[0087] (2) Comparison of current waveforms using an algorithm that can detect corresponding points on two waveforms and determine waveform similarity even if the length and period of the time series are different. In the switch machine condition monitoring device 31, the current waveform comparison processing unit 44 detects corresponding points in the first current waveform and the monitored current waveform using an algorithm that can detect corresponding points on two waveforms even if the length and period of the time series are different, and calculate the similarity of the waveforms.

[0088] This allows for accurate determination of the state at each stage of the conversion process, even if the length and period of the time series of each waveform differ.

[0089] (3) Comparison of current waveforms using D-DTW In the switch machine status monitoring device 31, the current waveform comparison processing unit 44 detects corresponding points in the first current waveform and the monitored current waveform based on calculations using D-DTW.

[0090] This allows for accurate determination of the state at each stage of the conversion operation, even if the monitored current waveform is time-series data that shows localized increases and decreases over time. [Explanation of Symbols]

[0091] 10. Electric Point Machine 11. Motor 31. Switch machine condition monitoring device 41...Current waveform acquisition section 42...Teacher information storage unit 43...Teacher information acquisition department 44. Current waveform comparison processing unit 45. Conversion operation process identification processing unit. 46. ​​State determination processing unit 47. Output Processing Unit

Claims

1. A first acquisition unit that acquires the first current waveform of the motor of the first electric point machine when the first electric point machine, which serves as a reference, is driven, and the switching operation process of the first electric point machine corresponding to the first current waveform, A second acquisition unit acquires a second current waveform, which is the motor current waveform when the second electric switch machine, which is the target of monitoring, is driven, A current waveform comparison processing unit compares the first current waveform and the second current waveform and detects corresponding points in the first current waveform and the second current waveform, A specification unit identifies the conversion operation process in the second current waveform based on corresponding points in the first current waveform and the second current waveform detected by the current waveform comparison processing unit, A determination unit determines the state of the second electric point machine for each switching operation process identified by the specified unit. A switch machine condition monitoring device characterized by having the following features.

2. In the switch machine status monitoring device according to claim 1, The current waveform comparison processing unit uses an algorithm that can detect corresponding points on two waveforms even if the length and period of the time series are different, and calculate the similarity of the waveforms, to detect corresponding points in the first current waveform and the second current waveform, respectively. A switch machine condition monitoring device characterized by the following features.

3. In the switch machine status monitoring device according to claim 2, The current waveform comparison processing unit detects corresponding points in the first current waveform and the second current waveform based on calculations performed by D-DTW. A switch machine condition monitoring device characterized by the following features.

4. A method for monitoring the status of an electric point machine, A first acquisition step of acquiring a first current waveform of the motor of the first electric point machine when the first electric point machine, which serves as a reference, is driven, and the switching operation process of the first electric point machine corresponding to the first current waveform, A second acquisition step involves acquiring a second current waveform, which is the motor current waveform when the second electric point machine, which is the target of monitoring, is driven, A current waveform comparison step involves comparing the first current waveform with the second current waveform and detecting corresponding points in the first current waveform and the second current waveform, A selection step to identify the conversion operation process in the second current waveform based on corresponding points in the first current waveform and the second current waveform detected by the processing of the current waveform comparison step, A determination step is taken to determine the state of the second electric point machine for each switching operation process identified by the processing of the specified step. A method for monitoring the condition of a railway switch, characterized by including the following:

5. The computer that performs the process of monitoring the status of the electric point machine, A first acquisition step of acquiring a first current waveform of the motor of the first electric point machine when the first electric point machine, which serves as a reference, is driven, and the switching operation process of the first electric point machine corresponding to the first current waveform, A second acquisition step involves acquiring a second current waveform, which is the motor current waveform when the second electric point machine, which is the target of monitoring, is driven, A current waveform comparison step involves comparing the first current waveform with the second current waveform and detecting corresponding points in the first current waveform and the second current waveform, A selection step to identify the conversion operation process in the second current waveform based on corresponding points in the first current waveform and the second current waveform detected by the processing of the current waveform comparison step, A determination step is taken to determine the state of the second electric point machine for each switching operation process identified by the processing of the specified step. A program that performs a process characterized by including the following.

Images