JP7849227B2 - Rail fracture detection device - Google Patents

Description

This invention relates to a rail fracture detection device that detects the fracture state of rails used for train operation and the presence of trains or other vehicles on the tracks.
More specifically, this relates to a rail fracture detection device that determines the rail fracture state and train presence status in a section of a railway track consisting of pairs of rails, based on the energization status of that section.
More specifically, this relates to a rail break detection device that can detect not only the state of rail breakage but also the presence or absence of a train by sending an inspection signal different from the return current (train current, electric vehicle current) during train operation to a track circuit that eliminates the need for impedance bonding by allowing the left and right rails of a rail pair to be energized as appropriate.

Regarding information indicating the rail fracture state, typical examples include the "unbalance ratio α obtained by dividing the difference between two measured values for the left and right rails (seaside rail/mountainside rail)" and the "equilibrium degree expressed in decibels as the voltage ratio." However, any physical quantity or numerical value that represents the degree of equilibrium or unbalance of the track circuit, such as the difference between the two measured values, may also be used. Therefore, in this application, these are collectively referred to as "equilibrium state value" or "equilibrium state value k0."

Conventional rail break detection devices using track circuits (see, for example, Non-Patent Document 1) have the advantage of being able to detect rail breaks even at night when no return current is flowing or during overhead line power outages. However, they have the limitation that the left and right rails cannot be simply short-circuited.
Therefore, when introducing a train control system that detects trains without using track circuits, such as by using wireless communication, a rail break detection device that can be used even for uninsulated track circuits that do not have or do not require impedance bonds is desirable.

Therefore, there was a need to realize a rail break detection device that could detect rail breakage by passing an inspection signal different from the return current through a track circuit that eliminates the need for impedance bonds by appropriately short-circuiting the left and right rails relative to the return current.
However, even in train control systems that eliminate the need for impedance bonds through the use of wireless technology, it is desirable to prevent the misplacement of maintenance vehicles, which tend to be outside the scope of control. Therefore, it is desirable to be able to easily and appropriately detect not only rail fracture conditions but also the presence or absence of trains and maintenance vehicles on the tracks.

To address such needs, rail break detection devices have been developed that can easily detect not only the presence or absence of rail breaks but also the presence or absence of trains (see, for example, Patent Documents 1 and 2).
Since the present invention relates to further improvements to such rail break detection devices, this section will first describe a known single-track rail break detection device 20 that has been improved upon the previous version (see, for example, Example 1 of Patent Documents 1 and 2), and then describe a known double-track rail break detection device 70 that has also been improved upon the previous version (see, for example, Example 5 of Patent Document 1 and Example 4 of Patent Document 2).

[Known Improvement Example 1 for Single-Track Systems]
The rail break detection device 20, which is a known improvement example 1 for single-track rails, will be described in detail with reference to the drawings.

Figure 6(a) shows an example of the arrangement of four rails 11-14 in a double track 10, a typical example of a railway track. Figure 6(b) is a schematic block diagram of a rail break detection device 20 installed in the downhill single track section (11 & 12), and Figure 6(c) is a detailed block diagram of its measurement unit 50 and determination unit 60. Furthermore, Figure 7 is a schematic diagram showing the equilibrium state value k0 obtained based on the measured values k1 and k2 of the diversion signals i1 and i2, expressed as the track circuit unbalance rate α (= |k1 - k2| / (k1 + k2) × 100%).

The railway track where the rail break detection device 20 is installed is typically a double track 10 with a down line and an up line running parallel to each other with a small gap between them (see Figure 6(a)). The down line consists of rail pair 11 & 12, where rail 11 on the mountain side (left side from the perspective of the oncoming train) and rail 12 on the sea side (right side from the perspective of the oncoming train) run parallel. The up line consists of another rail pair 13 & 14, where rail 13 on the mountain side (right side from the perspective of the oncoming train) and rail 14 on the sea side (left side from the perspective of the oncoming train) run parallel. In single-track lines, only one rail pair is provided and used for both up and down lines. In quadruple-track lines, more rail pairs are provided and used accordingly.

The rail break detection device 20 (see Figure 6(b)) is a basic configuration applied to rail pair 11 & 12 of a double track 10, and comprises multiple section end short-circuit lines 21, 22 attached at multiple locations on rail pair 11 & 12 at appropriate distances, for example, 1 km apart, an unbalancing means 25 attached to one of the section end short-circuit lines 21, a circulating circuit forming member 30 with one end connected to section end short-circuit line 21 and the other end connected to section end short-circuit line 22, a transmitting unit 40 that sends an inspection signal i0 to the circulating circuit forming member 30, a measuring unit 50 that measures a pair of current division signals i1, i2 related to the inspection signal i0 for each of the sub-lines 22a, 22b that are divided on both sides at the connection point 22c of the circulating circuit forming member 30 among the section end short-circuit lines 22 to which the circulating circuit forming member 30 is connected, and a determination unit 60 that makes a determination regarding rail breakage based on an equilibrium state value k0 calculated from the measured values k1, k2.

The section end short-circuit line 21 (see Figure 6(b)) is made of a good conductor such as an electric wire or copper plate. One end is connected to the rail 11 at section end 11a, and the other end is connected to the rail 12 at section end 12a, thus short-circuiting the left and right rails 11 and 12 at section ends 11a and 12a. One end of the circulating circuit forming member 30 is connected to this section end short-circuit line 21 at connection point 21c, and the section end short-circuit line 21 is divided into a section line 21a near section end 11a of rail 11 and a section line 21b near section end 12a of rail 12.

The section end short-circuit line 22 (see Figure 6(b)) is also made of good conductor material. One end is connected to the rail 11 at section end 11b, and the other end is connected to the rail 12 at section end 12b, thus short-circuiting the left and right rails 11 and 12 at section ends 11b and 12b. The other end of the circulating circuit forming member 30 is connected to this section end short-circuit line 22 at connection point 22c, and the section end short-circuit line 22 is divided into a section line 22a near section end 11b of rail 11 and a section line 22b near section end 12b of rail 12.

The circuit-forming member 30 (see Figure 6(b)) consists of an insulated wire, and the output section of the transmitting unit 40 is connected to its intermediate portion. If we divide the circuit-forming member 30 into two circuit-forming lines 31 and 32 at this connection point, and rephrase the connection state of both ends of the circuit-forming member 30, one circuit-forming line 31 is connected to the section end short-circuit line 21 at the aforementioned connection point 21c, and the other circuit-forming line 32 is connected to the section end short-circuit line 22 at the aforementioned connection point 22c.

Then, one side 31 of the circulating circuit forming member 30, the partial line 21a of the section end short-circuit line 21, the rail 11, the partial line 22a of the section end short-circuit line 22, and the other side 32 of the circulating circuit forming member 30 form a circulating circuit through which the diversion signal i1 of the inspection signal i0 is transmitted, and the portion of the rail 11 located between the section ends 11a and 11b becomes the inspection section 11c.
Furthermore, a circular circuit is formed by one side 31 of the circulating circuit forming member 30, the partial line 21b of the section end short-circuit line 21, the rail 12, the partial line 22b of the section end short-circuit line 22, and the other side 32 of the circulating circuit forming member 30, which carries the diversion signal i2 of the inspection signal i0, and the portion of the rail 12 located between the section ends 12a and 12b becomes the inspection section 12c.

The transmitting unit 40 (see Figure 6(b)) sends an inspection signal i0, which is different from the return current, to the circuit forming member 30. However, the inspection signal i0 splits into two branch signals i1 and i2 beyond the circuit forming member 30, so branch signal i1 is sent to one circuit on the rail 11 side (31, 21a, 11c, 22a, 32) and branch signal i2 is sent to one circuit on the rail 12 side (31, 21b, 12c, 22b, 32).
Since overhead lines are either DC electrified or AC electrified (commercial frequency), the return current uses DC or AC at commercial frequency (50Hz, 60Hz). Therefore, an AC signal with a higher frequency that is easily frequency-discriminated is suitable for the test signal i0. However, the test signal i0 is not limited to these; it can be any signal that is distinctly different from the return current.

The measurement unit 50 (see Figures 6(b) and (c)) is attached to the partial line 22a of the section end short-circuit line 22 to which the circulating circuit forming member 30 is connected, on the side leading from the connection point 22c to the rail 11. It measures the current diversion signal i1 flowing through the partial line 22a and, consequently, the circulating circuit (31, 21a, 11c, 22a, 32) on the rail 11 side, and the section end short-circuit line 22 from the connection point 22c The system comprises a current probe 52 (current sensor, current transformer, etc.) attached to the section line 22b leading to rail 12, which measures the shunt signal i2 flowing through section line 22b and, consequently, through the complete circuit (31, 21b, 12c, 22b, 32) on the rail 12 side; and calculation units 53-57 that determine the equilibrium state value k0 (calculated value) from the measured value k1 (measured signal) of current probe 51 and the measured value k2 (measured signal) of current probe 52.

Furthermore, calculation units 53 to 57 (see Figure 6(c)) comprise an addition unit 53 that generates the sum of measured values k1 and k2 by processing the measurement signal in its original state, i.e., the analog signal before AC is converted to DC by detection, etc.; a subtraction unit 54 that generates the difference current of measured values k1 and k2 in its original state; a receiving unit 55 that obtains the sum (k1 + k2), which is the sum of measured values k1 and k2, by performing detection, etc., from the sum of measured values k1 and k2; a receiving unit 56 that obtains the difference |k1 - k2|, which is the absolute value of the subtraction of measured values k1 and k2, by performing detection, etc., from the difference current of the subtraction unit 54; and an equilibrium state value calculation unit 57 that calculates the unbalance rate α, which is a typical example of the equilibrium state value k0 (equilibrium of the orbital circuit), by performing the calculation of dividing the above difference |k1 - k2| by the above sum (k1 + k2).

The determination unit 60 (see Figure 6(c)) includes a state determination unit 61 that determines the state of unbalance related to the diversion signals i1 and i2 in order to make a determination regarding rail breakage and train presence using the equilibrium state value k0, specifically the unbalance rate α, obtained from the measured values k1 and k2, and a determination unit 62 that makes a determination regarding "presence or absence of rail breakage" and "presence or absence of train presence" according to the current state of the unbalance state, including the equilibrium state, and the state transition from the previous state to the present. The implementation of these determination units 62, state determination unit 61, and the aforementioned equilibrium state value calculation unit 57 may be analog circuits, digital circuits, or microprocessor-embedded circuits.

The unbalancing means 25 (see Figure 6(b)) is designed to ensure that the balanced/unbalanced state of the paired shunt signals i1 and i2 is in a predetermined unbalanced state when there are no rail breaks or trains present. This can be achieved using known methods such as making the connection state of the section end short-circuit line 21 uneven or inserting an impedance member into the section end short-circuit line 21 (see, for example, Figure 3 of Patent Document 1 or Figure 2 of Patent Document 2). Although not shown here, inserting an inductance member such as a coil is easily implemented. The degree of unbalancing achieved by this means is greater than the intrinsic unbalanced state caused by the difference in impedance between the inspection section 11c of rail 11 and the inspection section 12c of rail 12, which is due to factors such as the length and material of the rails 11 and 12.

The determination method of the determination unit 60 in the rail fracture detection device 20 equipped with such unbalancing means 25 will be described in detail. Figure 7 shows the equilibrium state value k0 obtained based on the measured values k1 and k2 of the diversion signals i1 and i2, more specifically the track circuit unbalance rate α = {|k1 - k2| / (k1 + k2)}, expressed as a percentage (%). Furthermore, each of the multiple unbalance states shown in Figure 7 has a width, and a certain gap is maintained between adjacent unbalance states. The latter gap is intended to introduce hysteresis characteristics to the recognition and determination of state transitions by not recognizing a state transition until it crosses that gap and enters another unbalance state, thereby suppressing the over-detection of state transitions.

To explain this concretely with numerical examples, in a state where there are no rail fractures and no trains are present, the current diversion signal i1 becomes smaller than the current diversion signal i2 by the amount corresponding to the setting of the unbalancing means 25. Similarly, the measured value k1 becomes smaller than the measured value k2 by the amount corresponding to that unbalancing setting, resulting in a balance state value k0 of approximately 12% (see the left and right sides of the thick solid line in Figure 7 that correspond to the period when no trains are present). Therefore, if the unbalance rate α (balance state value k0) falls within the range of 10.6% to 14.4%, including this range, it is determined that the system has reached the "set unbalance state corresponding to the unbalancing setting of the unbalancing means 25."

Subsequently, if the unbalance ratio α (equilibrium state value k0) exceeds 15.1%, transitioning to a "major unbalance state" (greater than the set unbalance state), it is determined that a rail has fractured. If the unbalance ratio α (equilibrium state value k0) falls below 9.9%, transitioning to a "minor unbalance state" (less than the set unbalance state), it is determined that a train is present.

In that state, for example, if a rail fracture occurs in the inspection section 12c of rail 12, resulting in a situation where there is a rail fracture and no train present, the current diversion signal i1 will significantly exceed the current diversion signal i2, exceeding the aforementioned unbalance setting adjustment amount. Similarly, the measured value k1 will significantly exceed the measured value k2, causing the unbalance rate α (balance state value k0) to exceed approximately 40% (see the left and right sides of the dashed line in Figure 7 that correspond to the period when no train is present). In this case, it is determined that the system has entered the "highly unbalanced state" described above.

Furthermore, although not shown in the diagram, when the inspection section 11c of rail 11 breaks, the diversion signal i2 becomes larger than the diversion signal i1, exceeding the amount corresponding to the unbalance setting described above, and the value of the unbalance rate α (balance state value k0) becomes approximately 30% (see the left and right sides of the dashed line in Figure 7 that correspond to when no train is present), and thus it is determined that the "highly unbalanced state" described above has occurred.
Furthermore, when both rails 11 and 12 break, the inspection signal i0, which is the sum of the current diversion signals i1 and i2, does not reach a predetermined value such as a constant value. As a result, the sum of the measured values k1 and k2, k1+k2, also falls outside the predetermined value, and this can be detected separately to determine the cause.

The above describes a method for determining the condition of rail fractures when there are no trains present. However, when train 15 is in inspection sections 11c and 12c, the determination is made as follows.
First, when there is no rail breakage, a bypass current id flows through the axle of the train 15 in a direction that mitigates the effect of the unbalancing means 25, reducing the difference between the shunt signals i1 and i2, and bringing the measured values k1 and k2 closer together, so that the equilibrium state value k0 becomes approximately 5% (see the central part of the thick solid line in Figure 7 that corresponds to when a train is present). When the unbalance rate α (equilibrium state value k0) falls within the range of 0% to 9.9%, including this value, it is determined that the system has entered the "slightly unbalanced state" described above.

Furthermore, even if there is a rail break, if train 15 is located between the break point and the measurement side (22), the diversion signal i2 from the break side flows to the other rail 11 via the axle of train 15 and merges with the other diversion signal i1. In this way, the diversion signals i1 and i2 flow by bypassing the unbalancing means 25, so there is no significant difference between the diversion signals i1 and i2 and, consequently, the measured values k1 and k2, resulting in the aforementioned "slightly unbalanced state".
In contrast, when train 15 is located between the fracture point and the non-measurement side (21), the flow of the diversion signal i2 on the fracture side is obstructed, resulting in the aforementioned "highly unbalanced state."

Based on these case distinctions, the determination unit 60 determines that if the unbalance rate α (balance state value k0) obtained from the values of the diversion signals i1 and i2 is in a "set unbalance state," then the inspection sections 11c and 12c of rail pair 11 and 12 are in a state where "there are no rail breaks and no trains are present." If the values of the diversion signals i1 and i2 are determined to be in a "highly unbalanced state," then the inspection sections 11c and 12c of rail pair 11 and 12 are in a state where "there are rail breaks and the presence of trains is unknown." If the values of the diversion signals i1 and i2 are determined to be in a "slightly unbalanced state," then the inspection sections 11c and 12c of rail pair 11 and 12 are in a state where "a train is present."

The rail break detection device 20 of the known improved example 1 for single-track rails, which has the configuration described above, will now be explained in terms of its usage and operation.

When a train is not present in the inspection sections 11c and 12c of rail pair 11 and 12, and in a normal state without rail breakage, the imbalance caused by the unbalancing means 25 is clearly greater than the inherent imbalance due to the impedance variation that originally existed between the left and right rails. Therefore, the unbalance rate α (equilibrium state value k0) obtained from the shunt signals i1 and i2 becomes approximately 12%, which is the "set unbalanced state" (see the left and right sides of the thick solid lines in Figure 7 that correspond to the state when no train is present). Thus, an accurate determination is made that "there is no rail breakage and no train is present."

When a train is not present in the inspection sections 11c and 12c of rail pair 11 and 12, if a rail is fractured, only a small current flows through rail 12 where the fracture occurred. This results in a large imbalance exceeding the imbalance created by the unbalancing means 25, causing the unbalance rate α (equilibrium state value k0) obtained from the shunt signals i1 and i2 to exceed 30% or 40%, resulting in a "large unbalance state" (see the left and right sides of the dashed-dotted and double-dotted lines in Figure 7, which correspond to the period when no train is present). Therefore, a determination of "rail fracture confirmed" is made.

When a train is located in inspection sections 11c and 12c of rail pair 11 and 12, under normal conditions without rail breakage, the values of the current diversion signals i1 and i2 approach each other due to the short circuit between the left and right rails caused by the axle of train 15. The resulting unbalance rate α (equilibrium state value k0) becomes approximately 5%, indicating a "slightly unbalanced state" (see the central portion of the thick solid line in Figure 7 that corresponds to the train being present). Therefore, a determination is made confirming the presence of a train, indicating that "a train is present."

When a train is located in inspection sections 11c and 12c of rail pair 11 and 12, and there is a rail fracture, a detailed explanation that would be repeated will be omitted, but in short, depending on the relationship between the train's position and the rail fracture location, it will result in either a "minor imbalance" or a "major imbalance" (see the central part of the dashed-dotted or dashed-dotted lines in Figure 7 that corresponds to when a train is located). In a "minor imbalance," the determination is made that "a train is located," and in a "major imbalance," the determination is made that "there is a rail fracture."
If a train is detected or a rail fracture is detected, safety is confirmed by checking for any forgotten maintenance vehicles and searching for the location of the rail fracture.

[Known Improvement Example for Double Tracks 2]
The rail break detection device 70, which is a known improvement example 2 for double tracks, will be described in detail with reference to Figure 8.

The rail break detection device 70 differs from the rail break detection device 20 of the known improved example 1 described above in that inspection sections 11e and 12e have been added to the inspection sections 11c and 12c for rail pair 11 and 12, and inspection sections 13c, 14c and 13e and 14e have been added for another rail pair 13 and 14. These are referred to as the first modifications. Furthermore, the second modification that differentiates the rail break detection device 70 from the rail break detection device 20 is that, in addition to the inspection section 13c for one rail 13 of another rail pair 13 and 14, the inspection section 14c for the other rail 14 is also incorporated into the circulating circuit forming member 30, and that the same measurement unit 50a and determination unit 60a as in the measurement unit 50 and determination unit 60a of the other rail pair 13 and 14 are installed.

To elaborate on the first modification, for rail 11, section end 11d and inspection section 11e are set on the opposite side of section end 11b and inspection section 11c, with the transmission point 11aa, which is located at the center of the section, in between section end 11a. For rail 12, section end 12d and inspection section 12e are set on the opposite side of section end 12b and inspection section 12c, with the transmission point 12aa, which is located at the center of the section, in between section end 12a. For rail 13, section end 13d and inspection section 13e are set on the opposite side of section end 13b and inspection section 13c, with the transmission point 13aa, which is located at the center of the section, in between section end 13aa. For rail 14, section end 14d and inspection section 14e are set on the opposite side of section end 14b and inspection section 14c, with the transmission point 14aa, which is at the center of the section, in between section end 14aa. Furthermore, in conjunction with these, the short-circuit line 21 connecting transmission point 11aa and transmission point 12aa becomes the section-internal short-circuit line 21aa, and the connecting line 23 connecting transmission point 13aa and transmission point 14aa becomes the section-internal short-circuit line 23aa.

Furthermore, section ends 11d and 12d are connected by section end short-circuit line 26, where the left and right rails 11 and 12 are short-circuited; section ends 13d and 14d are connected by section end short-circuit line 27, where the left and right rails 13 and 14 are short-circuited; and section ends 13b and 14b are connected by section end short-circuit line 24, where the left and right rails 13 and 14 are short-circuited. Also, one end of the circulating circuit forming line 33 is connected to section end short-circuit line 22 at connection point 22c, and the other end is connected to section end short-circuit line 24 at connection point 24c; and one end of the circulating circuit forming line 34 is connected to section end short-circuit line 26 at connection point 26c, and the other end is connected to section end short-circuit line 27 at connection point 27c.

Furthermore, the same unbalancing means as the unbalancing means 25 described above is also attached to the short-circuit lines 26 and 27 at the end of those sections (the same reference numeral "25" is used in the figure).
Furthermore, the unbalancing means 25 that was attached to the section end short-circuit line 21 has been moved to the section end short-circuit line 22.
Furthermore, the connection destination of the current probes 51 and 52 has been changed from the section end short-circuit line 22 to the section in-section short-circuit line 21aa, which has a changed position from the section end short-circuit line 21.

To elaborate on the second modification, the transmission point 13aa of rail 13 and the transmission point 14aa of rail 14 are short-circuited by the section-internal short-circuit line 23aa, and the connection point of the circulating circuit forming line 32 is the intermediate connection point 23c of the section-internal short-circuit line 23aa.
Furthermore, a section end short-circuit line 24, similar to the section end short-circuit line 22, is connected to another rail pair 13 & 14 near the section end short-circuit line 22, so that the section end 13b of rail 13 and the section end 14b of rail 14 are short-circuited by the section end short-circuit line 24, and the connection point of the circulating circuit forming line 33 is the intermediate connection point 24c of the section end short-circuit line 24.

As a result, the inspection section 14c of rail 14 is incorporated into the circulating circuit forming member 30 in parallel with the inspection section 13c of rail 13, and the return current is also short-circuited for the left and right rails of the other rail pair 13 & 14.
Furthermore, the same unbalancing means as the unbalancing means 25 described above is also attached to the section end short-circuit line 24 described above (the same reference numeral "25" is used in the illustration).

Furthermore, a current probe 58 (current sensor, current transformer, etc.) is attached to the portion of the short-circuit line 23aa within the section that is closer to the transmission point 13aa of the rail 13 than the connection point 23c of the circulating circuit formation line 32. This probe measures the shunt signal i3 flowing through the inspection section 13c of the rail 13. Similarly, a current probe 59 (current sensor, current transformer, etc.) is attached to the portion of the short-circuit line 23 at the end of the section that is closer to the end of the section 14a of the rail 14 than the connection point 23c. This probe measures the shunt signal i4 flowing through the inspection section 14c of the rail 14. Based on the measured values k3 of the shunt signal i3 and k4 of the shunt signal i4, the measurement unit 50a and the determination unit 60a perform a determination regarding the fracture of the separate rail pair 13 & 14.

In this case, in addition to detecting the presence or absence of rail fractures and trains in the inspection section 11c (12c) of rail pair 11 & 12, it is also possible to detect the presence or absence of rail fractures and trains in other inspection sections 11e (12e) of the same rail pair 11 & 12, and inspection sections 13c (14c) and 13e (14e) of other rail pair 13 & 14.
As for the equipment required, the measurement unit and judgment unit will be installed in two sets: one set for rail pair 11 & 12 (50 & 60) and another set for rail pair 13 & 14 (50a & 60a). However, the transmission unit 40 will only require one unit with high transmission power.
Therefore, both equipment costs and installation costs can be saved.

The operation of this rail break detection device 70 will now be explained (see Figures 9 and 10). For clarity and simplification, only one side of the bidirectional transmission of the inspection signal i0 will be described. In this case, the inspection signal i0 sent from the transmitting unit 40 to the circulating circuit forming member 30 passes through the circulating circuit forming line 31 and the section short-circuit line 21aa, splitting into two sets: a pair of branch signals i1 for the inspection section 11c of rail 11 and i2 for the inspection section 12c of rail 12, and a pair of branch signals i5 for the inspection section 11e of rail 11 and i6 for the inspection section 12e of rail 12.

Then, the pair of branch signals i1 and i2 merge at the circulating circuit formation line 33 and then split again into branch signal i3 for the inspection section 13c of rail 13 and branch signal i4 for the inspection section 14c of rail 14. The pair of branch signals i5 and i6 merge at the circulating circuit formation line 34 and then split again into branch signal i7 for the inspection section 13e of rail 13 and branch signal i8 for the inspection section 14e of rail 14.
Then, i3, i4, i7, and i8 combine at the short-circuit line 23aa within the section, and at the circuit formation line 32 they become the test signal i0 and return to the transmission unit 40.

In such a signal transmission state, if there are no rail breaks and no trains present (see Figure 9(a)), the corresponding balance state value (unbalance rate α) of the diversion signals i1 and i2 will reach the set unbalance state (see both ends of the solid line graph in Figure 7), the corresponding balance state value (unbalance rate α) of the diversion signals i3 and i4 will reach the set unbalance state, the corresponding balance state value (unbalance rate α) of the diversion signals i5 and i6 will reach the set unbalance state, and the corresponding balance state value (unbalance rate α) of the diversion signals i7 and i8 will reach the set unbalance state.

Then, the sum of the currents from the shunt signals i1 and i5 (i1 + i5) and the sum of the currents from the shunt signals i2 and i6 (i2 + i6) will reach the corresponding equilibrium state value (unbalance rate α) which is the set unbalance state. Similarly, the sum of the currents from the shunt signals i3 and i7 (i3 + i7) and the sum of the shunt signals i4 and i8 (i4 + i8) will reach the corresponding equilibrium state value (unbalance rate α) which is the set unbalance state.

Consequently, the equilibrium value (unbalance rate α) related to the measured values k1 of the sum current (i1 + i5) and k2 of the sum current (i2 + i6) becomes the set unbalance state. Therefore, the determination unit 60 determines that the inspection sections 11c and 11e of rail 11 and 12c and 12e of rail 12 are in the "set unbalance state," resulting in the determination that "there are no rail fractures and no trains are present."

Furthermore, since the equilibrium state value (unbalance rate α) related to the measured values k3 of the sum current (i3 + i7) and k4 of the sum current (i4 + i8) falls into the set unbalance state, the determination unit 60a determines that the inspection sections 13c and 13e of rail 13 and 14c and 14e of rail 14 also fall into the "set unbalance state," resulting in the determination that "there are no rail fractures and no trains are present."

In contrast, if a rail fracture 12x occurs in any of the inspection sections 11c, 11e, 12c, or 12e relating to rail pair 11 & 12, for example in inspection section 12e (see Figure 9(b)), although a lengthy and repetitive explanation will be omitted, the equilibrium state value (unequilibrium rate α) becomes highly unequilibrium (see both ends of the dashed-dotted line graph corresponding to the fracture of the sea-side rail 12 in Figure 5), and a determination is made that "a rail fracture exists".
Furthermore, if a train enters any of the inspection sections 11c, 11e, 12c, or 12e relating to rail pair 11 and 12, for example, inspection section 11e or 12e, then, to put it simply, the equilibrium state value (unequilibrium rate α) will become slightly unequilibrium (see the midpoint between the train entry position and the transmission point in the solid line graph of Figure 7), and a judgment such as "a train is present" will be issued.

Thus, the rail fracture detection device 70 (known improved example 2 for double tracks) can detect the presence or absence of rail fractures and the presence or absence of trains in the inspection sections 11c, 11e, 12c, 12e, 13c, 13e, 14c, and 14e of the double track 10.
Furthermore, the rail break detection device 70 (known improved example 2 for double tracks) is designed so that the circulating circuit formation lines 31 and 32 and the section short-circuit lines 21aa and 23aa, which are directly involved in the transmission and reception of the inspection signal i0, are located on the inside (11aa, 12aa, etc.) of the inspection section rather than on the outside (11b, 11d, etc.), making it easy to install by connecting them along the length of the rail.

Japanese Patent Publication No. 2021-046163 Japanese Patent Publication No. 2021-066353

"An Introduction to Railway Signals for Railway Engineers: Track Circuits," published by the Japan Railway Electrical Engineering Association, revised edition, 2nd printing, May 20, 2005, pp. 3-5.

As a known improvement example 2 for double tracks, the rail break detection device 70 described above can be summarized as follows: "Multiple section end short-circuit lines 22, 26, 24, 27 and section in-section short-circuit lines 21aa, 23aa are attached to multiple locations on rail pairs 11 & 12, 13 & 14 that form the railway track, and at each attachment location, the return current is short-circuited for the left and right rails of the rail pair; a circulating circuit forming member 30 (31, 32, 33, 34) is connected to the section end short-circuit lines and the section in-section short-circuit lines and forms a circuit that circulates together with the rail pair; and the section in-section short-circuit line 21 via the circulating circuit forming member 30 This can be described as a "rail break detection device" comprising: a transmitting unit 40 that sends an inspection signal different from the return current to aa and 23aa; measuring units 50 and 50a that measure a pair of current division signals k1 & k2, k3 & k4 related to the inspection signal on both sides of the connection point of the circulating circuit forming member 30 for the short-circuit lines 21aa and 23aa within the section; an unbalancing means 25 that moves the balance state of the pair of current division signals away from equilibrium or into an unbalanced state when there is no rail break; and determination units 60 and 60a that make a determination regarding rail break and train presence based on the measured values of the pair of current division signals.

Furthermore, the determination unit determines whether the pair of current diversion signals are in a balanced or unbalanced state. If it determines that the system is in a pre-set unbalanced state corresponding to the unbalanced state created by the unbalanced means, it determines that there are no rail breaks and no trains present. If it determines that the system is in a highly unbalanced state, which is greater than the pre-set unbalanced state, it determines that there are rail breaks. If it determines that the system is in a slightly unbalanced state, which is less than the pre-set unbalanced state, it determines that a train is present. Thus, a determination regarding rail breaks and other issues is made.

A prototype rail fracture detection device 70 was built and installed on a double track 10. Simulation tests and experiments were conducted by running trains and maintenance vehicles on the track. The results showed that, regardless of whether the train was traveling at high speeds comparable to Shinkansen or limited express trains, or at lower speeds corresponding to maintenance vehicles, the device could distinguish between a rail fracture state, a train running state, or a normal state (see Figure 7) depending on whether the unbalance ratio α (equilibrium state value k0) was in a highly unbalanced state, a slightly unbalanced state, or an intermediate set unbalanced state.

However, after repeating experiments while changing various setting conditions, it was found that the unequilibrium rate α temporarily shifted from a small unequilibrium state to the set unequilibrium state for a short period of time while the train was running.
However, even if the transition to the unbalanced state is temporary, it can lead to unwanted false detections. To further understand the situation, we tried reducing the train's speed, which increased the duration of the unbalanced state. Furthermore, we found that stopping the train at the relevant location caused the undesirable transition state (unbalanced state) to persist.

Therefore, when we examined a graph with the distance from the transmission point on the horizontal axis and the unbalance rate α on the vertical axis (see Figure 10), we found that there is a "train detection impossible section" in the vicinity of the transmission point where the unbalance rate α transitions from the original small unbalance state to an undesirable set unbalance state when a train is running.
Furthermore, if maintenance vehicles or other vehicles remain parked in sections near transmission points where train detection is impossible, this situation is not normally conceivable, but it cannot be completely ruled out. Therefore, considering this possibility, it could lead to the oversight of undesirable incidents such as vehicles or other equipment being left behind on the tracks.

Therefore, a fundamental technical challenge is to improve the system so that it can distinguish between a train in motion and a normal state, including when the train is stopped, even in the aforementioned sections where train detection is impossible.
Furthermore, in order to avoid or suppress cost increases, it will be an even greater technical challenge to realize improvements by changing or expanding the functions of the judgment unit while avoiding the expansion of the measurement unit, etc.
Furthermore, one phenomenon that we focused on as a criterion that could be useful in solving such technical challenges is the difference in state transitions between a minor unbalanced state and a set unbalanced state between the train entry/exit positions at the end of the section and the transmission points within the section.

A concrete example of such judgment criteria is shown in the graph in Figure 10, which represents the difference in the slope of the non-horizontal portion related to the train entry and exit points and the section where train detection is impossible in the solid white line graph during train operation.
However, the slope is the rate of change β of the unbalance rate α per unit displacement (difference in distance) (hereinafter referred to as the rate of change β per unit displacement), and this rate of change β per unit displacement is obtained by dividing the difference in the unbalance rate in the vertical axis direction Δα in the non-horizontal portion between the train entry/exit point and the section where train detection is impossible by the difference in distance Δm in the horizontal axis direction.

Therefore, for a rail fracture detection device that does not possess precise positional or speed information of the moving body, it is difficult to calculate the rate of change β per unit displacement.
Therefore, a specific technical challenge is to improve the rail fracture detection device so that it can determine whether a train or other vehicle is remaining in the section near the transmission point where train detection is impossible, or has left the inspection section, by using other physical quantities that can be measured or calculated, rather than such a physical quantity (β).

The rail break detection device of the present invention (Solution 1) was devised to solve the above problems,
A rail break detection device comprising: multiple section end short-circuit lines and section in-section short-circuit lines attached to multiple locations on a pair of rails forming a railway track, each of which short-circuits the return current for the left and right rails of the rail pair at the respective attachment locations; a circulating circuit forming member connected to the section end short-circuit lines and the section in-section short-circuit lines and forming an electrical circuit that circulates together with the rail pair; a transmitting unit that sends an inspection signal different from the return current to the section in-section short-circuit lines via the circulating circuit forming member; a measuring unit that measures a pair of shunt signals related to the inspection signal on both sides of the connection point of the circulating circuit forming member for the section in-section short-circuit lines; an unbalancing means that moves the balance state of the pair of shunt signals away from a balanced state or into an unbalanced state when there is no rail break; and a determination unit that makes a determination regarding rail break and train presence based on the measured values of the pair of shunt signals,
The system comprises means for calculating an unbalance ratio, which indicates the degree of the unbalance state as a ratio, from the measured value of the current distribution signal, and means for calculating the rate of change per unit time based on the change in the unbalance ratio over time, wherein the determination unit selects one of the following states: a track-occupied state, an unoccupied state, or a rail-broken state, according to the changes in the unbalance ratio and the rate of change per unit time.
The determination unit is characterized in that, when determining a state transition in at least the occupied state or the unoccupied state, it uses the classification of the unequilibrium rate into large, medium, and small, and the classification of the rate of change per unit of time into large and small, as factors for selecting the destination state .

Furthermore, the rail break detection device of the present invention (Solution 2) is a rail break detection device of Solution 1 described above, characterized in that, at the start of operation, the determination unit selects one of the occupying state, the non-occupying state, and the rail break state according to the classification of large, medium, and small related to the unbalance rate and adopts it as the initial state.

Furthermore, the rail fracture detection device of the present invention (Solution 3) is a rail fracture detection device of Solution 2 described above, characterized in that the determination unit , in addition to classifying the unequilibrium rate into large, medium, and small, also uses the classification of the rate of change per unit of time as a factor in selecting the transition destination state.

Furthermore, the rail break detection device of the present invention (Solution 4) is a rail break detection device of Solution 3 described above, characterized in that the determination unit maintains the rail-occupied state as long as the classification of the unbalance rate is small when the rail-occupied state is in the present state, maintains the rail-occupied state even if the classification of the unbalance rate becomes medium as long as the rate of change per unit time is small, transitions to the non-occupied state when the classification of the unbalance rate becomes medium and the rate of change per unit time becomes large, and transitions to the rail break state when the classification of the unbalance rate becomes large.

Furthermore, the rail break detection device of the present invention (Solution 5) is a rail break detection device of Solution 3 described above, characterized in that the determination unit maintains the non-occupied state as long as the classification of the unbalance rate is medium when the non-occupied state is present, maintains the non-occupied state even if the classification of the unbalance rate becomes small as long as the rate of change per unit time is small, transitions to the occupied state when the classification of the unbalance rate becomes small and the rate of change per unit time becomes large, and transitions to the rail break state when the classification of the unbalance rate becomes large.

Furthermore, the rail break detection device of the present invention (Solution 6) is the rail break detection device of Solution 3, wherein the determination unit is
When the rail is fractured, the rail remains fractured as long as the unbalance ratio is classified as high; when the unbalance ratio is classified as medium, the state transitions to the non-occupied state; and when the unbalance ratio is classified as low, the state transitions to the occupied state.
When the rail is occupied, the rail is maintained as long as the classification of the unbalance rate is small; even if the classification of the unbalance rate becomes medium, the rail is maintained as long as the rate of change per unit time is small; when the classification of the unbalance rate becomes medium and the rate of change per unit time becomes large, the state transitions to the non-occupied state; and when the classification of the unbalance rate becomes large, the state transitions to the rail fractured state.
The present invention is characterized in that, when the rail is not present, the rail is not present as long as the classification of the unbalance rate is medium, the rail is not present even if the classification of the unbalance rate becomes small as long as the rate of change per unit time is small, the rail is transitioned to the present state when the classification of the unbalance rate becomes small and the rate of change per unit time becomes large, and the rail is transitioned to the broken state when the classification of the unbalance rate becomes large.

In the rail fracture detection device of the present invention (Solution 1), by making the state determination using the rate of change per unit time (γ) that can be calculated from the originally available unbalance rate (α), the rail fracture detection device can determine whether a train or the like is remaining in the section where train detection is impossible near the transmission point or has left the inspection section, without having to use the rate of change per unit displacement (β), which is difficult to calculate.
Therefore, this invention can solve the specific technical problems described above.

Furthermore, in the rail break detection device of the present invention (Solution 2), for a brief moment at the start of operation, the rate of change per unit time (γ) based on the time-dependent change of the unbalance rate (α) cannot be obtained. Therefore, the initial state is tentatively determined based solely on the unbalance rate (α). However, since the rate of change per unit time (γ) is introduced for the purpose of distinguishing between the track-occupied state (small unbalance state) and the unoccupied state (set unbalance state), and does not affect the selection of the most important rail break state (large unbalance state), the means for solving the problem can be easily implemented.

Furthermore, in the rail fracture detection device of the present invention (Solution 3), the classification based on the rate of change per unit time (γ) utilizes the difference in physical properties between the section where train detection is impossible, including the transmission point, and the end of the section far from the transmission point. Therefore, a simple classification of large or small is sufficient for selecting the transition state, allowing for a simple implementation of the problem-solving means.

Furthermore, in the rail break detection device of the present invention (Solution 4), regarding the state transition when the train is present, the train remains present as before when the unbalance rate (α) is small, and transitions to the rail break state as before when the unbalance rate (α) becomes large. However, when the unbalance rate (α) becomes medium, instead of immediately transitioning to the non-present state as before, the rate of change per unit time (γ) is also checked, and when the rate of change per unit time (γ) is large, the train transitions to the non-present state, while when the rate of change per unit time (γ) is small, unlike before, it is considered that there is no state change and the train remains present.
This ensures that the train's presence is correctly maintained even when it is in a section where train detection is impossible, resulting in a low rate of change per unit time (γ). Therefore, this rail break detection device correctly determines the presence of a train even if it remains in a section where train detection is impossible near the transmission point.

Furthermore, in the rail break detection device of the present invention (Solution 5), regarding the state transition when the train is not present, the train remains in the non-presence state as before when the unbalance ratio (α) is medium, and transitions to the rail break state as before when the unbalance ratio (α) becomes large. However, when the unbalance ratio (α) becomes small, instead of immediately transitioning to the present state as before, the rate of change per unit time (γ) is also checked. When the rate of change per unit time (γ) is large, the train transitions to the present state, while when the rate of change per unit time (γ) is small, unlike before, it is considered that there is no state change and the train remains in the non-presence state. As a result, the non-presence state is maintained if there are no trains in the inspection section, including sections where train detection is impossible. Therefore, this rail break detection device can correctly determine that trains, etc., are not remaining in the inspection section.

Furthermore, in the rail fracture detection device of the present invention (Solution 6), in addition to accurately determining the presence and absence of rails based on the unbalance rate (α) and the rate of change per unit time (γ) as described above, the rail fracture state can also be accurately determined based on the unbalance rate (α).

Embodiment 1 of the present invention shows the structure of a rail break detection device, where (a) is an outline block diagram relating to a rail break detection device installed on a double-track section, which is a typical example of a railway track, and (b) is a state transition diagram showing the function of the determination unit of the rail break detection device. (a) and (b) are both characteristic graphs related to high-speed running in clear weather. (a) is a graph showing the change in the unbalance rate α over time, with the unbalance rate α at the time of passing high-speed train plotted on the vertical axis and the elapsed time t on the horizontal axis. (b) is a graph showing the transition state of the rate of change γ per unit time relative to the running position, with the "rate of change γ per unit time of unbalance rate α per unit time" at the vertical axis and the distance from the transmission point on the horizontal axis. Similarly, (a) shows the change in the unequilibrium rate α over time, and (b) shows the change in the rate of change per unit time γ for each driving position, but both are characteristic graphs related to high-speed driving in rainy weather. Similarly, (a) shows the change in the unequilibrium rate α over time, and (b) shows the change in the rate of change per unit time γ for each driving position, but both are characteristic graphs related to low-speed driving in clear weather. Similarly, (a) shows the change in the unequilibrium rate α over time, and (b) shows the change in the rate of change per unit time γ for each driving position, but both are characteristic graphs related to low-speed driving in rainy weather. The present invention relates to a known improvement example 1 of a conventional rail break detection device for single-track use, illustrating the background technology. (a) shows an example of rail arrangement in a double-track railway, which is a typical example of a railway track. (b) is an overview block diagram of a rail break detection device installed in a single-track section. (c) is a detailed block diagram of the measurement unit and the determination unit. This is a schematic diagram showing the equilibrium state of an orbital circuit as represented by the unequilibrium rate. This diagram shows the structure of a rail fracture detection device, relating to a known improved example 2 for double-track rail fracture detection devices, which is part of the background technology. The above known improvement example 2 shows the operating state of the rail break detection device, where (a) is the normal state (set unbalanced state), (b) is the rail break state (large unbalanced state), and (c) is the train detection state (small unbalanced state). This graph relates to the above known improvement example 2, with the distance from the transmission point on the horizontal axis and the unbalance ratio on the vertical axis.

A specific embodiment of the rail fracture detection device of the present invention will be described below by Example 1.
Embodiment 1, shown in Figures 1 to 5, embodies all of the solutions 1 to 6 described above (claims 1 to 6 in the original application).
In these illustrations, detailed and complex circuit diagrams were omitted, and block diagrams were frequently used for simplification, focusing on those necessary or related to the explanation of the invention.

The specific configuration of Embodiment 1 of the rail break detection device of the present invention will be described with reference to Figure 1.
Figure 1(a) shows the structure of the rail break detection device 80 and is an overview block diagram relating to the rail break detection device 80 installed in the double-track section 11-14, which is a typical example of a railway track.
Figure 1(b) is a state transition diagram showing the function of the determination unit 82 of the device 80.

Figures 2(a) to 5(a) are graphs showing the change in the unequilibrium rate α over time when a train passes, with the unequilibrium rate α on the vertical axis and the elapsed time t on the horizontal axis.
Figures 2(b) to 5(b) are graphs showing the transition state of the rate of change γ per unit time relative to the travel position, with the rate of change γ per unit time plotted on the vertical axis and the distance from the transmission point on the horizontal axis.
Of these, Figure 2 shows the results of high-speed driving in sunny conditions, Figure 3 shows the results of high-speed driving in rainy conditions, Figure 4 shows the results of low-speed driving in sunny conditions, and Figure 5 shows the results of low-speed driving in rainy conditions.

The rail break detection device 80 (see Figure 1(a)) is a modified version of the rail break detection device 70, which is the previously described known improved example 2 for double tracks. The differences from the rail break detection device 70 are that the previously described measuring units 50 and 51a have become measuring units 81 and 81a, respectively, and the previously described determination units 60 and 60a have become determination units 82 and 82a, respectively.
Furthermore, the differences between the measurement unit 81 and the determination unit 82 and the measurement unit 81a and the determination unit 82a are the same as those for the rail fracture detection device 70 described above, and the difference lies in whether the measurement target or the determination target is rail 11 & 12 or a different rail 13 & 14.
Therefore, we will omit the repetitive explanation and instead describe the measurement unit 81 and the determination unit 82 in detail.

The measurement unit 81 (see Figure 1(a)) calculates the unbalance ratio α, which indicates the degree of unbalance as a ratio, from the measured values k1 and k2 of the shunt signals i1 and i2 using the formula [|k1-k2|/(k1+k2)×100%], similar to the measurement unit 50 described above. In addition, it also calculates the rate of change per hour γ, which is the rate of change of the unbalance ratio α over time. Theoretically, this rate of change per hour γ is the time derivative of the unbalance ratio α, but practically, it is calculated as the time difference of the unbalance ratio α. Various noise countermeasures, such as local moving averages, are also implemented, but since the specific methods are well-known, a detailed explanation is omitted. Furthermore, as described above, while the rate of change per displacement β is the distance derivative of the unbalance ratio α, the rate of change per hour γ is the time derivative of the unbalance ratio α. Therefore, the rate of change per hour γ can be easily calculated from the hourly unbalance ratio α even without precise positional or speed information of trains, etc.

The determination unit 82 (see Figure 1(a)) distinguishes between three states: a non-occupied state where no train is present in the inspection sections 11c, 12c, 11e, and 12e of the down-line rail pair 11 and 12; an occupied state where a train is present somewhere in those inspection sections 11c to 12e; and a broken rail state where a break exists in any of the rails in those inspection sections 11c to 12e. Furthermore, state transitions are performed not only in response to changes in the unbalance rate α, but also in response to changes in the rate of change per unit time γ. Therefore, one of the states—occupied, non-occupied, or broken rail—is selected based on changes in the combined data of the unbalance rate α and the rate of change per unit time γ. While hysteresis characteristics are considered in the determination and recognition of state transitions, as previously described, these characteristics will not be mentioned here to avoid complicating the explanation.

Furthermore, the functions of the determination unit 82 can be clearly and concisely illustrated in a state transition diagram (see Figure 1(b)), so the functions of the determination unit 82 will be described in detail with reference to that diagram.
First, at the start of operation, the unbalance rate α can be obtained immediately, but the acquisition of the rate of change per unit time γ, which is calculated from the time difference of the unbalance rate α, is delayed by a moment, so the initial state is assigned according to the value of the unbalance rate α. Specifically (see Figure 7 and Figure 1(b)), when the unbalance rate α is, for example, less than 9.9% (a small unbalance state, hereinafter simply referred to as "small"), the state of inspection section 11c to 12e is set to "occupied state", when the unbalance rate α is, for example, 10.0% to 15.0% (a set unbalance state, hereinafter simply referred to as "medium"), the state of inspection section 11c to 12e is set to "not occupied state", and when the unbalance rate α is, for example, more than 15.1% (a large unbalance state, hereinafter simply referred to as "large"), the state of inspection section 11c to 12e is set to "rail fracture state".

Subsequently (see Figure 1(b)), when the unbalance ratio α becomes large, the condition of inspection section 11c to 12e is considered a rail fracture state, regardless of whether the rate of change per unit time γ is large or small. In this "rail fracture state," as long as the unbalance ratio α remains large, the condition of inspection section 11c to 12e continues to be considered a rail fracture state. When the unbalance ratio α becomes medium, the condition of inspection section 11c to 12e transitions to an unoccupied state, and when the unbalance ratio α becomes small, the condition of inspection section 11c to 12e transitions to an occupied state. Therefore, when a train approaches transmission points 11aa and 12aa from the rail fracture location, detection of the train's presence takes priority, and when the rail fracture state is resolved through track repair, the system can automatically respond to that rail condition.

Unlike the rail fracture state, in the case of the occupying and non-occupying states, in addition to the classification of the unequilibrium rate α as described above, the classification of the rate of change per unit time γ is also a factor in selecting the destination state (see Figure 1(b)).
Specifically, taking into account fluctuating factors such as expected environmental conditions like sunny and rainy weather and the speed of the train, a comparison is first made between the minimum value of the rate of change per hour γ at the train entry/exit position (see γ = 5.3 in Figure 3) and the maximum value of the rate of change per hour γ at the transmission point (see γ = 4.5 in Figure 2), and the midpoint value (for example, 4.9). Then, if the value of the rate of change per hour γ is greater than the midpoint value (= 4.9), it is classified as "large," and if the value of the rate of change per hour γ is less than the midpoint value (= 4.9), it is classified as "small."

To elaborate on the minimum value of the rate of change per unit time γ (γ = 5.3), this corresponds to the smallest value among γ = 6.4 and 6.5 in Figure 2 for high-speed driving in sunny weather, γ = 5.3 and 5.4 in Figure 3 for high-speed driving in rainy weather, γ = 6.4 and 6.5 in Figure 4 for low-speed driving in sunny weather, and γ = 5.3 and 5.4 in Figure 5 for low-speed driving in rainy weather, which is "5.3" in Figure 3.
Furthermore, the highest value of the rate of change per unit time γ (γ = 4.5) is the largest value among γ = 4.5 in Figure 2, γ = 4.2 in Figure 3, γ = 2.3 in Figure 4, and γ = 2.1 in Figure 5, which is "4.5" in Figure 2.

Furthermore, in the "non-occupied state," as long as the unbalance ratio α is moderate, the state of inspection section 11c to 12e remains non-occupied. However, when the unbalance ratio α becomes large, the state of inspection section 11c to 12e becomes rail fractured. Moreover, when the unbalance ratio α becomes small, the state does not immediately transition; instead, the magnitude of the rate of change per unit time γ is also examined. When the rate of change per unit time γ is small, it is considered that the rate of change per unit time γ corresponds to the value at the time of passing the transmission point, and therefore no state change occurs, maintaining the non-occupied state without a state transition. Then, when the unbalance ratio α becomes small and the rate of change per unit time γ becomes large, the state transitions to the occupied state.

In contrast, in the "occupied state," as long as the unbalance ratio α is small, the state of inspection section 11c to 12e remains in the occupied state. However, when the unbalance ratio α becomes large, the state of inspection section 11c to 12e becomes the rail fracture state. Furthermore, when the unbalance ratio α becomes medium, the state does not immediately transition based solely on this. The magnitude of the rate of change per unit time γ is also examined. If the rate of change per unit time γ is small, it can be considered that the rate of change per unit time γ corresponds to the value at the time of passing the transmission point, and therefore no state change occurs. Thus, the occupied state is maintained without a state transition. Finally, when the unbalance ratio α becomes medium and the rate of change per unit time γ becomes large, the state transitions to the occupied state.

The usage and operation of the rail break detection device 80 of this embodiment 1 will be explained with reference to the drawings.
As described above, Figures 2(a) to 5(a) are graphs showing the change over time of the unequilibrium rate α when a train passes, and Figures 3(b) to 5(b) are graphs showing the transition state of the rate of change γ per unit time relative to the running position. Furthermore, Figure 2 is for high-speed running in sunny conditions, Figure 3 is for high-speed running in rainy conditions, Figure 4 is for low-speed running in sunny conditions, and Figure 5 is for low-speed running in rainy conditions.

Furthermore, the high-speed operation is assumed to be over 260 km/h, similar to that of Shinkansen trains, while the low-speed operation is assumed to be about half of that; in both cases, these speeds are faster than those of maintenance vehicles. Therefore, even if maintenance vehicles are not covered by the Shinkansen monitoring system, the rail break detection device 80 can detect whether or not maintenance vehicles are present on the tracks. The following provides a detailed explanation of the different cases.

First, in the normal state where there are no rail fractures and no trains are running, the unbalance rate α remains medium and the rate of change per unit time γ remains small, so the determination unit 82 makes an accurate determination that the track is not occupied (see Figure 1(b)).
In contrast, if a rail fracture occurs when there is no train running and the track is not occupied, the unbalance ratio α changes from medium to large, so the determination unit 82 makes an accurate determination that the rail is fractured (see Figure 1(b)). When the rail fracture is repaired, the unbalance ratio α returns to medium, so the determination unit 82 determines that the track is occupied (see Figure 1(b)).

Furthermore, if a rail fracture occurs while the train is within the inspection section and is on the track, the unbalance ratio α changes from small to large, so the determination unit 82 makes an accurate determination that a rail fracture has occurred (see Figure 1(b)). More specifically, if the fracture occurs at a position before the train's position relative to the transmission point, the fracture is detected immediately. If the fracture occurs behind the train's position relative to the transmission point, it is not detected immediately, but when the train passes the transmission point, the unbalance ratio α changes from small to large, and the fracture is detected at that point.
As long as the rail fracture exists, the unbalance ratio α remains high, and the determination unit 82 continues to determine that the rail fracture is present. However, once the rail fracture is repaired, if the train is still on the track, the unbalance ratio α returns to low, and the determination unit 82 returns to the appropriate train-on-track condition. If the train has already been cleared away, the unbalance ratio α returns to medium, and the determination unit 82 returns to the appropriate non-train-on-track condition (see Figure 1(b)).

Furthermore, if a train enters the inspection section without any rail fractures, regardless of whether it is traveling at high speed or low speed, or whether it is sunny or rainy (see Figures 2 to 5), first, upon the train's entry into the inspection section, the unbalance ratio α changes from medium to small (see the left end of (a) in each figure), and the rate of change per unit time γ becomes greater than the aforementioned intermediate value (= 4.9) (see the left end of (b) in each figure), so the determination unit 82 determines that the train is appropriately present on the track (see Figure 1(b)).
Furthermore, as long as the train is between the train entry position and the section where train detection is impossible, the unbalance ratio α remains small, even if the train moves forward, pauses, or reverses, so the determination unit 82 maintains an appropriate train presence status (see Figure 1(b)).

Then, when the train enters the section where train detection is impossible, the unbalance ratio α becomes medium (see the central part of Figures 2(a) to 5(a)), but since the rate of change per unit time γ remains small, the determination unit 82 maintains an appropriate train presence (see Figure 1(b)). Then, when the train leaves the section where train detection is impossible, regardless of whether the direction of travel is forward or backward, the unbalance ratio α returns to small (see the intermediate part between the central and end sections in Figures 2(a) to 5(a)), and the determination unit 82 maintains an appropriate train presence (see Figure 1(b)).

Furthermore, as the train proceeds and reaches the end of section 11b, 12b or 11d, 12d, and then exits inspection section 11c, 12c or 11e, 12e, the unbalance rate α changes from small to medium (see both ends of Figures 2(a) to 5(a)), and the rate of change per unit time γ becomes greater than the aforementioned intermediate value (=4.9) (see both ends of (b) in each figure), so the determination unit 82 determines that the train is not present (see Figure 1(b)).

Thus, the rail fracture detection device 80 can detect not only the rail fracture state in the inspection section 11c to 12e of the rail pair 11 and 12 on the down line, but also the presence of trains and even the remaining maintenance vehicles.
Furthermore, although I will omit the repetitive and complicated explanation, it is also possible to detect not only the rail fracture status in the inspection section 13c to 14e of the separate rails 13 and 14 on the up line, but also the presence of trains and even the remaining maintenance vehicles.

[others]
In the determination unit 82 described above (see Figure 1(b)), if a train or the like happens to be in a section where train detection is impossible at the start, the non-occupied state may be selected as the initial state. In such cases, it is desirable to equip the rail break detection device 80 with an extended function that issues an alarm to check the train presence status in the section where train detection is impossible, and, if necessary, transitions the state from non-occupied state to occupied state based on the operation input of the confirmation result.
However, even if such extended functions are not implemented, if maintenance vehicles or similar vehicles are first driven around under the supervision of personnel to check the track condition and safety at the start of the system, the state of the determination unit 82 will be automatically reset to an appropriate state.

In the above known improved example and embodiment, the measurement unit 50 performed addition and subtraction of the measured values k1 and k2 using a dedicated circuit before receiving the measurement signals (see Figure 1(c)). However, it is also possible to receive these measurement signals first and then process them using an appropriate dedicated calculation circuit or general-purpose processor. Furthermore, although the measurement unit 50 and the determination unit 60 were in separate blocks, the hardware does not need to be separate as long as both units can perform their functions. For example, the determination unit 60 and the balance state value calculation unit 57 may be implemented in a single processor, and in addition to 60 and 57, the addition unit 53, subtraction unit 54, and even the receiving units 55 and 56 may also be implemented in a single processor.

10. Double track (railway)
11 & 12 Rail pair (left and right rails, track)
11, 12 Rails; 12x Rail break; 13 & 14 Separate rail pair (left and right rails, track)
13, 14 Rails (separate rails)
11a, 12a, 13a, 14a Section ends 11aa, 12aa, 13aa, 14aa Transmission points (within the section)
11b, 12b, 13b, 14b Section ends 11c, 12c, 13c, 14c Inspection sections 11d, 12d, 13d, 14d Section ends 11e, 12e, 13e, 14e Inspection sections 15 Train 20 Rail break detection device 21, 22, 24, 26, 27 Section end short-circuit lines (connecting lines)
21aa (21), 23aa (23) Short-circuited lines (transmission lines) within the section
21a, 21b, 22a, 22b: Partial lines; 21c, 22c, 23c, 24c, 26c, 27c: Connection points (diversion points and confluence points)
25, 25a, 25b, 25c Unbalancing means 30 Circuit forming member 31, 32, 33, 34 Circuit forming line (connection line)
40 Transmitting unit 50, 50a Measurement unit 51, 52 Current probe 53 Adding unit 54 Subtracting unit 55, 56 Receiving unit 57 Equilibrium state value calculation unit 58, 59 Current probe 60, 60a Judgment unit 61 State discrimination unit 62 Determination unit 70 Rail break detection device 80 Rail break detection device 81, 81a Measurement unit 82, 82a Judgment unit i0 Inspection signal i1, i2, i3, i4, i5, i6, i7, i8 Diversion signal k1, k2, k3, k4 Measured value (measurement signal)
k0 Equilibrium state value (calculated value)
α Unequilibrium rate (calculated value)
β Rate of change per unit displacement (calculated value)
γ Rate of change per hour (calculated value)

Claims (6)

A rail break detection device comprising: multiple section end short-circuit lines and section in-section short-circuit lines attached to multiple locations on a pair of rails forming a railway track, each of which short-circuits the return current for the left and right rails of the rail pair at the respective attachment locations; a circulating circuit forming member connected to the section end short-circuit lines and the section in-section short-circuit lines and forming an electrical circuit that circulates together with the rail pair; a transmitting unit that sends an inspection signal different from the return current to the section in-section short-circuit lines via the circulating circuit forming member; a measuring unit that measures a pair of shunt signals related to the inspection signal on both sides of the connection point of the circulating circuit forming member for the section in-section short-circuit lines; an unbalancing means that moves the balance state of the pair of shunt signals away from a balanced state or into an unbalanced state when there is no rail break; and a determination unit that makes a determination regarding rail break and train presence based on the measured values of the pair of shunt signals,
The system comprises means for calculating an unbalance ratio, which indicates the degree of the unbalance state as a ratio, from the measured value of the current distribution signal, and means for calculating the rate of change per unit time based on the change in the unbalance ratio over time, wherein the determination unit selects one of the following states: a track-occupied state, an unoccupied state, or a rail-broken state, according to the changes in the unbalance ratio and the rate of change per unit time.
A rail break detection device characterized in that, when the determination unit determines the state transition in at least the occupied state or the unoccupied state, it uses the classification of the unbalance rate into large, medium, and small, and the classification of the rate of change per unit of time into large and small, as factors for selecting the destination state . The rail break detection device according to claim 1, characterized in that, when the determination unit starts operation, it selects one of the occupying state, the unoccupied state, and the rail break state according to the classification of large, medium, or small related to the unbalance rate and adopts it as the initial state. The rail break detection device according to claim 2, characterized in that the determination unit , in addition to classifying the unequilibrium rate into large, medium, and small, also uses the classification of the rate of change per unit of time as a factor in selecting the transition destination state. The rail break detection device according to claim 3, characterized in that the determination unit maintains the rail-occupied state as long as the classification of the unbalance rate is small when the rail-occupied state is in the rail-occupied state, maintains the rail-occupied state even if the classification of the unbalance rate becomes medium as long as the rate of change per unit of time is small, transitions to the non-occupied state when the classification of the unbalance rate becomes medium and the rate of change per unit of time becomes large, and transitions to the rail break state when the classification of the unbalance rate becomes large. The rail break detection device according to claim 3, characterized in that the determination unit maintains the non-occupied state as long as the classification of the unbalance rate is medium when the non-occupied state is present, maintains the non-occupied state even if the classification of the unbalance rate becomes small as long as the rate of change per unit time is small, transitions to the occupied state when the classification of the unbalance rate becomes small and the rate of change per unit time becomes large, and transitions to the rail break state when the classification of the unbalance rate becomes large. When the determination unit determines that the rail is broken, it maintains the broken rail state as long as the unbalance ratio classification is high, transitions to the non-occupied state when the unbalance ratio classification becomes medium, and transitions to the occupied state when the unbalance ratio classification becomes low.
When the rail is occupied, the rail is maintained as long as the classification of the unbalance rate is small; even if the classification of the unbalance rate becomes medium, the rail is maintained as long as the rate of change per unit time is small; when the classification of the unbalance rate becomes medium and the rate of change per unit time becomes large, the state transitions to the non-occupied state; and when the classification of the unbalance rate becomes large, the state transitions to the rail fractured state.
The rail break detection device according to claim 3, characterized in that, when the rail is not present, the rail is not present as long as the classification of the unbalance rate is medium, the rail is not present even if the classification of the unbalance rate becomes small as long as the rate of change per unit time is small, the rail is transitioned to the present state when the classification of the unbalance rate becomes small and the rate of change per unit time becomes large, and the rail is transitioned to the broken state when the classification of the unbalance rate becomes large.