Intra-train abnormality detection system and method therefor

Abstract

To detect abnormality in a train on the basis of a measurement result by measuring movement of passengers in the train without using information by a video image.SOLUTION: An intra-train abnormality detection system includes: a plurality of pieces of measurement means for measuring movement of passengers in each vehicle of a train obtained by connecting a plurality of vehicles; and abnormality detection means for comparing each measurement value of the plurality of pieces of measurement means with an abnormality determination value, and detecting that abnormality occurs in the train when a measurement value of at least one measurement means among the plurality of pieces of measurement means exceeds a range of the abnormality determination value.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a train abnormality detection system and method for detecting an abnormality inside a train. [Background technology]

[0002] To ensure the reliability of railways, it is important to not only ensure quality under normal circumstances, but also to respond quickly to abnormalities that occur during operation.Initial responses to abnormalities that occur on trains are generally carried out by crew members, who are often stationed in the crew cabins at the front and rear of the train in order to monitor the areas in front and behind the train while it is moving.

[0003] Therefore, in a train consisting of multiple carriages, it is difficult for the crew to visually detect an abnormality in an intermediate carriage.

[0004] Therefore, systems have been introduced that notify the occurrence of abnormalities according to the type of abnormality. For example, regarding abnormalities in vehicle equipment, there is a system that detects wheel derailment using the internal pressure value of air springs (Patent Document 1), and a system that detects vehicle overload using the internal pressure value of air springs (Patent Document 2). Regarding abnormalities in passengers, there is a system that can monitor passengers and grasp abnormalities by delivering images from surveillance cameras installed in passenger compartments and analysis results of suspicious people to the operation manager (Patent Document 3). [Prior art documents] [Patent documents]

[0005] [Patent Document 1] JP 2013-241134 A [Patent Document 2] JP 2006-218933 A [Patent Document 3] JP 2021-141389 A Summary of the Invention [Problem to be solved by the invention]

[0006] However, in order to monitor passengers based on image information, including surveillance camera footage, on existing trains that do not have surveillance cameras installed, new equipment such as surveillance cameras must be installed. Moreover, sending and receiving image information, including surveillance camera footage, requires a network with the communication capacity to transmit image information with a large amount of data. For this reason, the cost of installing the equipment increases when using surveillance cameras to detect abnormalities inside trains.

[0007] An object of the present invention is to measure the movement of passengers inside a train without using video information and to detect abnormalities inside the train based on the measurement results. [Means for solving the problem]

[0008] In order to solve the above problems, the present invention is characterized in comprising a plurality of measurement means for measuring passenger movement in each car of a train having a plurality of coupled cars, and an abnormality detection means for comparing the measurement values ​​of each of the plurality of measurement means with an abnormality determination value, and detecting that an abnormality has occurred in the train when the measurement value of at least one of the plurality of measurement means exceeds the range of the abnormality determination value. Effect of the Invention

[0009] According to the present invention, the movement of passengers inside a train can be measured without using video information, and abnormalities inside the train can be detected based on the measurement results.

[0010] Problems, configurations and effects other than those described above will become apparent from the following description of the preferred embodiment of the invention. [Brief description of the drawings]

[0011] [Figure 1] 1 is a configuration diagram showing a configuration example of an in-train anomaly detection system according to a first embodiment of the present invention. [Diagram 2] 5 is a flowchart showing an example of a process of an anomaly detection logic according to the first embodiment of the present invention. [Diagram 3] FIG. 2 is a characteristic diagram for explaining the air spring internal pressure and the time difference of the air spring internal pressure value of each car in the in-train abnormality detection system according to the first embodiment of the present invention. [Figure 4] FIG. 2 is a characteristic diagram of a normal distribution applied to the in-train anomaly detection system according to the first embodiment of the present invention. [Diagram 5] FIG. 11 is a configuration diagram showing a configuration example of an in-train anomaly detection system according to a second embodiment of the present invention. [Figure 6] 11 is a flowchart illustrating an example of a process of an anomaly detection logic according to the second embodiment of the present invention. [Figure 7] FIG. 11 is a characteristic diagram for explaining the air spring internal pressure and the time difference of the air spring internal pressure value of each car and a train car in the in-train abnormality detection system according to the second embodiment of the present invention. [Figure 8] FIG. 11 is a configuration diagram showing a configuration example of an in-train anomaly detection system according to a third embodiment of the present invention. [Figure 9] 13 is a flowchart illustrating an example of a process of an anomaly detection logic according to the third embodiment of the present invention. [Figure 10] FIG. 11 is a characteristic diagram for explaining the internal pressure of the air spring and the time difference of change in the internal pressure of the air spring of each car in the in-train abnormality detection system according to the third embodiment of the present invention. [Figure 11] FIG. 11 is a configuration diagram showing a configuration example of an in-train anomaly detection system according to a fourth embodiment of the present invention. [Figure 12] 13 is a flowchart illustrating an example of a process of an anomaly detection logic according to the fourth embodiment of the present invention. [Figure 13] FIG. 11 is a characteristic diagram for explaining the air spring internal pressure and the air spring internal pressure decrease time difference of each car and each train car in the in-train anomaly detection system according to the fourth embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0012] Hereinafter, embodiments 1 to 4 will be described with reference to the drawings. The symbols used in embodiment 1 are synonymous with the symbols used in the description of the other embodiments. EXAMPLES

[0013] The first embodiment is an anomaly detection method for dealing with an abnormality occurring when a passenger is trapped inside a train while the train is stopped. The evacuation behavior of passengers moving between cars is detected using the variation in the air spring internal pressure value between cars. In this embodiment, an example in which an air spring internal pressure value measuring device that measures the air spring internal pressure is used as a representative measurement means for measuring passenger movement will be described. As the measurement means, other means that can grasp the number of passengers and the occupancy rate can also be used, such as a PCS (Passenger Count System) including an infrared sensor or a thermal sensor, a tire pressure measuring device, a car suspension device, a concentration measuring device that measures oxygen concentration or carbon dioxide concentration, a mat using a piezoelectric element, an optical sensor, etc. In that case, it is desirable to configure it so that the direction of passenger movement can be detected.

[0014] Fig. 1 is a configuration diagram showing an example of the configuration of an in-train abnormality detection system according to a first embodiment of the present invention. In Fig. 1, the in-train abnormality detection system includes a plurality of cars 101 and 102, an abnormality alarm device 103, a car information control device 104, abnormality detection logic 105, air springs 106 and 107, and air spring internal pressure value measuring devices 108 and 109. Car 101 is equipped with the abnormality alarm device 103, car information control device 104, abnormality detection logic 105, air spring 106, and air spring internal pressure value measuring device 108, and car 102 is equipped with the air spring 107 and air spring internal pressure value measuring device 109.

[0015] Air springs 106, 107 are composed of, for example, a diaphragm and laminated rubber, and are installed between the bogie and car body (neither shown) of vehicles 101, 102, respectively. Air spring internal pressure value measuring device 108 measures the internal pressure value acting on air spring 106, and outputs the measurement result as air spring internal pressure value 111 to vehicle information control device 104. Air spring internal pressure value measuring device 109 measures the internal pressure value acting on air spring 107, and outputs the measurement result as air spring internal pressure value 112 to vehicle information control device 104.

[0016] Vehicle information control device 10 4During operation, the vehicle information control device 10 constantly receives the measurement results of the air spring internal pressure measuring devices 108, 109 and monitors the air spring internal pressure values ​​111, 112 of the air springs 106, 107 mounted on the vehicles 101, 102. 4 uses anomaly detection logic 105 to monitor changes in air spring internal pressure values ​​111, 112 of air springs 106, 107, and when it detects an abnormal movement in air spring internal pressure values ​​111, 112 due to, for example, passengers' evacuation behavior, it determines that an abnormality has occurred in one of the vehicles 101, 102 and outputs an abnormality detection result 110 to the abnormality notification device 103. The abnormality notification device 103 notifies the abnormality on the condition that it receives the abnormality detection result 110. When passenger movement is measured by the number of passengers, the vehicle information control device 10 4 can calculate the number of passengers in each vehicle based on the measured values ​​(load acting on the vehicle) of the air spring internal pressure measuring devices 108, 109 and the average weight of the passengers.

[0017] The vehicle information control device 104 is configured, for example, by a computer (not shown) including a processor, a main storage device, an auxiliary storage device, an input device, an output device, and a communication device.

[0018] The processor is configured using, for example, a CPU (Central Processing Unit) and an MPU (Micro Processing Unit).

[0019] The main memory device is a device that stores computer programs and data, and is, for example, a Read Only Memory (ROM), a Random Access Memory (RAM), or a non-volatile semiconductor memory.

[0020] The auxiliary storage device is, for example, a hard disk drive, a solid state drive (SSD), an optical storage medium (i.e., a compact disc (CD) and a digital versatile disc (DVD)), a storage system, an integrated circuit card (IC card), a secure digital (SD) memory card, and other storage medium read / write devices, as well as a storage area of ​​a cloud server, etc. Computer programs and data stored in the auxiliary storage device are loaded into the main storage device as needed.

[0021] An example of a computer program stored in the auxiliary storage device is an anomaly detection program that serves as an anomaly detection logic. The CPU reads out the anomaly detection program from the auxiliary storage device and executes it to realize the function of the anomaly detection logic or the anomaly detection unit.

[0022] Examples of the input device include a keyboard, a mouse, a touch panel, a card reader, an audio input device, etc. The output device (display device) is a user interface that provides the user with various information such as the progress and results of processing. Examples of the output device include a screen display device (i.e., a liquid crystal monitor, LCD (Liquid Crystal Display), or a graphics card), an audio output device (i.e., a speaker, etc.), or a printing device, etc.

[0023] The communication device is a wired or wireless communication interface that realizes communication with other devices via a communication means such as a LAN, the Internet, etc. The communication device is, for example, a network interface card (NIC), a wireless communication module, a universal serial bus (USB) module, or a serial communication module.

[0024] Fig. 2 is a flowchart showing an example of processing of the abnormality detection logic according to the first embodiment of the present invention. In Fig. 2, this processing is started by the processor activating the abnormality detection logic 105. The abnormality detection logic 105 acquires air spring internal pressure values ​​111, 112 measured by the air spring internal pressure measuring devices 108, 109 as the current air spring internal pressure values ​​of each vehicle (S201).

[0025] Next, the abnormality detection logic 105 calculates the difference and time difference of the air spring internal pressure values ​​of each vehicle based on the air spring internal pressure values ​​111, 112 measured by the air spring internal pressure value measuring devices 108, 109 (S202). At this time, the abnormality detection logic 105 calculates the time change (change per unit time) of the air spring internal pressure values ​​111, 112 acquired in step S201 as the difference of the air spring internal pressure values ​​of each vehicle, compares the air spring internal pressure values ​​111, 112 acquired in step S201 with the air spring internal pressure values ​​111, 112 acquired in step S201 in the previous processing loop (the previous measurement cycle), and calculates the comparison result indicating the difference between the two as the time difference of the air spring internal pressure values ​​of each vehicle. That is, the abnormality detection logic 105 acquires the air spring internal pressure values ​​111, 112 for each measurement cycle, and calculates the difference between the acquired air spring internal pressure values ​​111, 112 for each measurement cycle as the time difference of the air spring internal pressure values ​​of each vehicle. Note that in the first calculation, set values ​​are used as the air spring internal pressure values ​​111, 112 acquired in the previous processing loop. Also, one processing loop means the processing of steps S201 to S203.

[0026] Next, based on the calculation results of the difference in the air spring internal pressure values ​​and the time difference of each vehicle, the abnormality detection logic 105 determines whether or not the difference in the air spring internal pressure values ​​and the time difference of each vehicle simultaneously exceed the increase threshold or the decrease threshold, which are thresholds for determining the occurrence of an abnormality (S203).

[0027] If the abnormality detection logic 105 obtains a negative (NO) determination result in step S203, it returns to step S201 and repeats the processes of steps S201 to S203.

[0028] On the other hand, when the abnormality detection logic 105 obtains a positive (YES) determination result in step S203, it outputs the abnormality detection result 110 to the abnormality alarm device 103. For example, when an abnormality occurs in the vehicle 101, due to the evacuation behavior of the passengers (evacuation behavior in which the passengers in the vehicle 101 move to the vehicle 102), the air spring internal pressure value 111 decreases in the vehicle 101, and the air spring internal pressure value 112 increases in the vehicle 102. Therefore, in the vehicle 101, the difference and time difference of the air spring internal pressure value 111 exceed the threshold value for the decrease (exceed the range of the abnormality determination value), and in the vehicle 102, the difference and time difference of the air spring internal pressure value exceed the threshold value for the increase (exceed the range of the abnormality determination value). Specifically, in the vehicle 101, the difference and time difference of the air spring internal pressure value 111 fall below the threshold value for the decrease, and in the vehicle 102, the difference and time difference of the air spring internal pressure value exceed the threshold value for the increase. In this case, the abnormality detection logic 105 outputs an abnormality detection result 110 indicating that an abnormality has occurred in the vehicle 101 to the abnormality notification device 103 .

[0029] Thereafter, the abnormality notification device 103 notifies the driver that an abnormality has occurred in the vehicle in which the air spring internal pressure value has decreased. For example, when an abnormality has occurred in the vehicle 101, the abnormality notification device 103 notifies the driver that an abnormality has occurred in the vehicle in which the air spring internal pressure value has decreased, i.e., in the vehicle 101.

[0030] FIG. 3 is a characteristic diagram for explaining the air spring internal pressure and the air spring internal pressure value time difference of each vehicle in the train anomaly detection system according to the first embodiment of the present invention. FIG. 3(a) shows the behavior of the air spring internal pressure of the vehicle 101, where the horizontal axis indicates the elapsed time and the vertical axis indicates the air spring internal pressure measured by the air spring internal pressure value measuring device 108. In FIG. 3(a), the air spring internal pressure (difference) of the vehicle 101 (vehicle 1) normally changes at a substantially constant value over time as the air spring internal pressure value 301. Here, if an abnormality occurs in the vehicle 101 at time 303 and the passengers of the vehicle 101 evacuate to the vehicle 102 (vehicle 2), the air spring internal pressure value 301 of the vehicle 101 gradually decreases, and thereafter, the decrease 304 falls below the threshold value 306 indicating the air spring internal pressure threshold value of the decrease. In other words, the decrease 304 falls outside the range of the threshold value 306 indicating the abnormality determination value.

[0031] 3(b) shows the behavior of the air spring internal pressure of the vehicle 102, where the horizontal axis indicates elapsed time and the vertical axis indicates the air spring internal pressure measured by the air spring internal pressure value measuring device 109. In FIG. 3(b), the air spring internal pressure (difference) of the vehicle 102 (vehicle 2) normally changes as an air spring internal pressure value 302 at a substantially constant value over time. Here, at time 303, when an abnormality occurs in the vehicle 101 and the passengers of the vehicle 101 evacuate to the vehicle 102 (vehicle 2), the air spring internal pressure value 301 of the vehicle 101 gradually decreases while the air spring internal pressure value 302 of the vehicle 102 gradually increases, and thereafter, the increment 305 exceeds the threshold value 307 indicating the air spring internal pressure threshold value of the increment. In other words, the increment 305 falls outside the range of the threshold value 307 indicating the abnormality determination value.

[0032] FIG. 3(c) shows the behavior of the air spring internal pressure value time difference of the vehicle 101, where the horizontal axis indicates elapsed time and the vertical axis indicates the air spring internal pressure value time difference measured by the air spring internal pressure value measuring device 108. In FIG. 3(c), the air spring internal pressure value time difference of the vehicle 101 (vehicle 1) normally changes at a substantially constant value as time passes as the air spring internal pressure value time difference 308. Here, at time 303, when an abnormality occurs in the vehicle 101 and the passengers of the vehicle 101 evacuate to the vehicle 102 (vehicle 2), the air spring internal pressure value time difference 308 of the vehicle 101 decreases rapidly and falls below the threshold value 309 indicating the air spring internal pressure value time difference threshold. Thereafter, the air spring internal pressure value time difference 308 continues to be below the threshold value 309, and returns to the original value when the evacuation of the passengers of the vehicle 101 is completed. In this case, if the state in which air spring internal pressure value time difference 308 is below threshold value 309 continues for more than the set time from time 303 and exceeds duration threshold value 310 indicating the air spring internal pressure value elapsed time threshold, in other words, if air spring internal pressure value time difference 308 is continuously below threshold value 309 for more than duration threshold value 310, air spring internal pressure value time difference 308 of vehicle 101 is determined to be abnormal.

[0033] FIG. 3(d) shows the behavior of the air spring internal pressure value time difference of the vehicle 102, where the horizontal axis indicates elapsed time and the vertical axis indicates the air spring internal pressure value time difference measured by the air spring internal pressure value measuring device 109. In FIG. 3(d), the air spring internal pressure value time difference of the vehicle 102 (vehicle 2) normally changes as an air spring internal pressure value time difference 311 with time elapse at a substantially constant value. Here, at time 303, when an abnormality occurs in the vehicle 101 and the passengers of the vehicle 101 evacuate to the vehicle 102 (vehicle 2), the air spring internal pressure value time difference 311 of the vehicle 102 increases rapidly as the air spring internal pressure value time difference 308 of the vehicle 101 decreases rapidly, and exceeds the threshold value 312 indicating the air spring internal pressure value time difference threshold. Thereafter, the air spring internal pressure value time difference 311 continues to exceed the threshold value 312, and returns to the original value when the evacuation of the passengers of the vehicle 101 is completed. In this case, if the air spring internal pressure value time difference 311 continues to exceed the threshold value 312 for more than the set time and exceeds the duration threshold value 313 indicating the air spring internal pressure value elapsed time threshold, in other words, if the air spring internal pressure value time difference 311 continuously exceeds the threshold value 312 for more than the duration threshold value 313, the air spring internal pressure value time difference 311 of the vehicle 102 is determined to be abnormal.

[0034] At time 303, if an abnormality occurs in vehicle 101, passengers in vehicle 101 evacuate to vehicle 102, air spring internal pressure value 301 of vehicle 101 decreases, and air spring internal pressure value 302 of vehicle 102 increases, and all of the following conditions (1) to (4) are satisfied, an abnormality is detected in vehicle 101 (the vehicle in which air spring internal pressure value 301 has decreased), and information that an abnormality has occurred in vehicle 101 is notified to the crew.

[0035] (1) The decrease 304 in the air spring internal pressure value 301 falls below the threshold value 306 . (2) The increase 305 in the air spring internal pressure value 302 exceeds the threshold value 307. (3) The air spring internal pressure value time difference 308 falls below the threshold value 309 continuously for a duration threshold value 310 or more. (4) The air spring internal pressure value time difference 311 exceeds the threshold value 312 continuously for a duration threshold value 313 or more.

[0036] As a result of the above, it becomes possible to detect evacuation behavior in which passengers move from vehicle 101 where the abnormality occurred to adjacent vehicle 102 when an abnormality occurs in the vehicle, and by notifying the crew of vehicle 101 where the abnormality occurred, it becomes possible to respond to the abnormality.

[0037] It should be noted that even if the conditions (1) to (2) among the above (1) to (4) are satisfied, it is possible to detect that an abnormality has occurred in the vehicle 101. However, if an abnormality has occurred in the vehicle 101 is detected only when the conditions (1) to (2) are satisfied, the decrease 304 and the increase 305 may not be measured accurately if the air spring internal pressure values ​​301 and 302 contain errors such as sudden noise or outliers, and it may not be possible to accurately detect that an abnormality has occurred in the vehicle 101. Also, even if the conditions (3) to (4) among the above (1) to (4) are satisfied, it is possible to detect that an abnormality has occurred in any of the vehicles. However, it is difficult to identify the vehicle in which an abnormality has occurred only when the conditions (3) to (4) are satisfied. For this reason, by detecting an abnormality in the vehicle when all of the above conditions (1) to (4) are satisfied, it is possible to accurately detect that an abnormality has occurred in the vehicle 101.

[0038] In addition, the values ​​of each threshold may be calculated from empirical values, statistical information, etc., and the calculated values ​​may be recorded in advance in a recording means such as a memory, or may be transmitted wirelessly from a ground station to the vehicle, or may be calculated based on data collected on board the vehicle.

[0039] As a statistical method for distinguishing between normal values ​​and abnormal values, for example, a method based on normal distribution can be adopted. Note that, since it is only necessary to determine a threshold value when distinguishing between normal values ​​and abnormal values, any of various statistical methods such as t distribution, F distribution, Fisher's linear discriminant, quadratic discriminant analysis, support vector machine, Bayesian statistics (MCMC method, conjugate distribution method), in addition to normal distribution, may be used.

[0040] One of the simplest ways to use a normal distribution is to use the central limit theorem to take the average over a finite time period and collect the average values.

[0041] 4 is a characteristic diagram of a normal distribution applied to the in-train anomaly detection system according to the first embodiment of the present invention. The horizontal axis represents a random variable, and the vertical axis represents a probability.

[0042] In FIG. 4, for example, when the one-minute average value X of the air spring internal pressure value at a certain point in time is 85, the average value μ of the air spring internal pressure value while the train is stopped is 100, and the standard deviation σ is 3, calculation can be performed using the following formula (1) of Mathematical Expression 1.

number

[0043] In a normal distribution, the probability of exceeding ±5 is approximately 0.000057%, so under normal circumstances, such a phenomenon can only occur approximately once every 29,240 hours.

[0044] If we apply the above to an actual train, it is equivalent to 15 people moving from a train with 100 passengers to an adjacent train every minute. If the train has doors on both sides, this is equivalent to one person moving to an adjacent train every 8 seconds per door, which is unlikely to happen under normal circumstances, but is considered a possible event if we limit it to when an abnormality occurs.

[0045] The average value X can be other values ​​such as 30 seconds instead of 1 minute, and the average value μ does not necessarily have to be the average, but can be a value at a point when the value is relatively stable (such as 1 minute after the doors close). The standard deviation can be calculated using values ​​from the past when X and μ were taken at the same time, or values ​​after the doors close or while the train is stopped. When the number of samples is small, t-distribution can be used instead of normal distribution.

[0046] In addition, although "5" was used as the threshold value in the above, other values ​​may be used depending on the differences between each railway operator and each line.

[0047] According to this embodiment, the movement of passengers in a train is measured without using video information, and an abnormality in the train can be detected based on the measurement result. Also, according to this embodiment, it is possible to detect an evacuation behavior in which an abnormality occurs in a train and passengers move from the train 101 in which the abnormality occurred to the adjacent train 102, and it is possible to respond to the abnormality by notifying the crew of the train 101 in which the abnormality occurred. Furthermore, according to this embodiment, even in an existing train in which a monitoring camera is not installed, if an air spring internal pressure measurement device and a train information control device are installed, it is possible to detect an abnormality in the train by modifying the software (abnormality detection logic) installed in the train information control device, and it is possible to significantly reduce the cost of introducing equipment for an existing train in which a monitoring camera is not installed. EXAMPLES

[0048] The second embodiment is an anomaly detection method for dealing with a state in which passengers are trapped in a car while the train is stopped and an abnormality occurs, and passengers escape through windows or the like. In this state, the amount of passenger evacuation to cars where no abnormality occurs decreases. To deal with this, the second embodiment employs an anomaly detection method using the total value of the air spring internal pressure of the entire car. That is, if some passengers escape through a window or the like when an abnormality occurs, the amount of movement between cars decreases, and therefore the method of the first embodiment cannot detect the abnormality. On the other hand, when the total value of the air spring internal pressure of the entire train decreases while the train is stopped at a station or even when the door opening operation is not performed, it is highly likely that some abnormality has occurred in the car. Therefore, a method for detecting an abnormality in a car based on the total value of the air spring internal pressure of the entire train and its time difference is adopted.

[0049] In this embodiment, as in the first embodiment, an example will be described in which an air spring internal pressure measuring device is used as a measuring means for measuring the movement of a passenger. do may use other means.

[0050] 5 is a configuration diagram showing a configuration example of an in-train anomaly detection system according to a second embodiment of the present invention. In this embodiment, an anomaly detection logic 405 is arranged in the train information control device 104 instead of the anomaly detection logic 105, and the other configuration is the same as in the first embodiment. Hereinafter, the processing of the anomaly detection logic 405 will be described, focusing mainly on the processing different from the first embodiment.

[0051] 6 is a flowchart for explaining an example of the processing of the anomaly detection logic according to the second embodiment of the present invention. This processing is started by the processor activating the anomaly detection logic 405. In addition, in steps S201 to S204, the same processing as in steps S201 to S204 in Fig. 2 is performed, and therefore the description of these processing is omitted. However, if the result in step S204 is negative, the processing proceeds to step S601 instead of returning to step S201.

[0052] When a negative (NO) determination result is obtained in step S203, the abnormality detection logic 405 sums up the air spring internal pressure values ​​of all the cars in the formation (S601). For example, when cars 101 and 102 are cars in the formation, the abnormality detection logic 405 acquires and sums up the air spring internal pressure values ​​111, 112 measured by air spring internal pressure measuring devices 108, 109 as the air spring internal pressure values ​​of all the cars in the formation.

[0053] Next, the abnormality detection logic 405 calculates the difference and the time difference between the total value of the air spring internal pressure values ​​and the total value of the air spring internal pressure values ​​in the previous processing loop based on the air spring internal pressure values ​​111, 112 measured by the air spring internal pressure value measuring devices 108, 109 (S602). At this time, the abnormality detection logic 405 compares the total value of the air spring internal pressure values ​​111, 112 acquired in step S601 with the total value of the air spring internal pressure values ​​111, 112 acquired in the previous processing loop, and calculates the difference and the time difference of the total value of the air spring internal pressure values.

[0054] Next, the abnormality detection logic 405 determines whether or not the difference in the total value of the air spring internal pressure value and the time difference exceed a threshold value for the decrease, based on the calculation results of the difference in the total value of the air spring internal pressure value and the time difference (S603).

[0055] If the abnormality detection logic 405 obtains a negative (NO) determination result in step S603, it returns to step S201 and repeats the processes of steps S201 to S203.

[0056] On the other hand, if the abnormality detection logic 405 obtains an affirmative (YES) determination result in step S603, it outputs the abnormality detection result 110 to the abnormality alarm device 103. For example, when an abnormality occurs in the vehicle 101, some of the passengers in the vehicle 101 escape through a window or the like and are trapped in the vehicle 101, and the remaining passengers move from the vehicle 101 to the vehicle 102, the number of people who move due to the evacuation behavior of the passengers is reduced compared to the first embodiment, so that the abnormality detection logic 405 outputs the abnormality detection result 110 indicating that an abnormality has occurred in the train vehicle to the abnormality alarm device 103 on the condition that the total value of the air spring internal pressure values ​​of the entire vehicle exceeds the threshold value for the reduction.

[0057] Thereafter, the abnormality alarm device 103 notifies a crew member that an abnormality has occurred in the vehicle in which the air spring internal pressure value has decreased (S204). For example, when an abnormality has occurred in a train vehicle, the abnormality alarm device 103 notifies a crew member that an abnormality has occurred in the vehicle in which the air spring internal pressure value has decreased.

[0058] FIG. 7 is a characteristic diagram for explaining the air spring internal pressure and the time difference of the air spring internal pressure value of each vehicle and a train car in the in-train abnormality detection system according to the second embodiment of the present invention. FIG. 7(a) shows the behavior of the air spring internal pressure of the vehicle 101, where the horizontal axis indicates the elapsed time and the vertical axis indicates the air spring internal pressure measured by the air spring internal pressure value measuring device 108. In FIG. 7(a), the air spring internal pressure (difference) of the vehicle 101 (vehicle 1) normally changes at a substantially constant value over time as the air spring internal pressure value 301. Here, if an abnormality occurs in the vehicle 101 at time 303, and some of the passengers of the vehicle 101 escape to the outside of the vehicle 101 through a window or the like, and the remaining passengers evacuate to the vehicle 102 (vehicle 2), the air spring internal pressure value 301 of the vehicle 101 gradually decreases, and then the decrease 304 falls below a threshold value 306 indicating the air spring internal pressure threshold value of the decrease. That is, the decrease 304 falls outside the range of the threshold 306 indicating the abnormality determination value.

[0059] FIG. 7(b) shows the behavior of the air spring internal pressure of the vehicle 102, where the horizontal axis indicates elapsed time and the vertical axis indicates the air spring internal pressure measured by the air spring internal pressure value measuring device 109. In FIG. 7(b), the air spring internal pressure (difference) of the vehicle 102 (vehicle 2) normally changes at a substantially constant value as time passes as the air spring internal pressure value 602. Here, at time 303, if an abnormality occurs in the vehicle 101, some of the passengers of the vehicle 101 escape to the outside of the vehicle 101 through a window or the like, and the remaining passengers take refuge in the vehicle 102 (vehicle 2), the air spring internal pressure value 602 of the vehicle 102 gradually increases. However, the increment 605 of the air spring internal pressure value 602 of the vehicle 102 is thereafter maintained at a value smaller than the threshold value 307 indicating the air spring internal pressure threshold value of the increment. That is, the increment 605 falls outside the range of the threshold value 307 indicating the abnormality determination value.

[0060] Fig. 7(c) shows the behavior of the total air spring internal pressure of the entire train including cars 101 and 102 (total train), where the horizontal axis indicates elapsed time and the vertical axis indicates the total air spring internal pressure measured by air spring internal pressure value measuring devices 108 and 109. In Fig. 7(c), the total air spring internal pressure of the entire train normally remains at a substantially constant value over time as total air spring internal pressure value 606 of the entire train. If an abnormality occurs in car 101 at time 303, and some of the passengers in car 101 escape to the outside of car 101 through a window or the like, and the remaining passengers evacuate to car 102 (car 2), then total air spring internal pressure value 606 of the entire train (entire train) gradually decreases and falls below threshold value 608 indicating the total air spring internal pressure threshold for the decrease in the entire train.

[0061] Fig. 7(d) shows the behavior of the air spring internal pressure total time difference of the entire train, where the horizontal axis indicates elapsed time and the vertical axis indicates the air spring internal pressure total time difference measured by air spring internal pressure value measuring devices 108, 109. In Fig. 7(d), the air spring internal pressure total time difference of the entire train normally changes over time at a substantially constant value as air spring internal pressure total value time difference 609 of the entire train. If an abnormality occurs in train 101 at time 303, and some of the passengers in train 101 escape to the outside of train 101 through a window or the like, and the remaining passengers evacuate to train 102 (train 2), then air spring internal pressure total value time difference 609 of the entire train will suddenly decrease and fall below threshold value 610 indicating the air spring internal pressure total value time difference threshold for the decrease in the entire train. Thereafter, the air spring internal pressure total value time difference 609 of the entire train continues to be below the threshold value 610, and returns to its original value when the evacuation of passengers from car 101 is completed. At this time, if the state in which the air spring internal pressure total value time difference 609 of the entire train continues to be below the threshold value 610 for more than a set time from time 303 and exceeds a duration threshold value 611 indicating an elapsed time threshold for the air spring internal pressure total value time difference of the entire train, in other words, if the air spring internal pressure total value time difference 609 of the entire train falls below the threshold value 610 for more than the duration threshold value 611 continuously, the air spring internal pressure total value time difference 609 of the entire train is determined to be abnormal.

[0062] At time 303, if an abnormality occurs in vehicle 101, some of the passengers in vehicle 101 escape to the outside of vehicle 101 through a window or the like, the remaining passengers evacuate to vehicle 102 (vehicle 2), the air spring internal pressure value 301 of vehicle 101 decreases, and the air spring internal pressure value 602 of vehicle 102 increases, and all of the following conditions (1) to (2) are satisfied, an abnormality is detected in vehicle 101 and information that an abnormality has occurred in vehicle 101 is notified to the crew.

[0063] (1) The air spring internal pressure sum 606 falls below the air spring internal pressure sum threshold 608. (2) The air spring internal pressure total value time difference 609 falls below the threshold value 610 continuously for a duration threshold value 611 or more.

[0064] According to this embodiment, even in a case where passengers are trapped inside a vehicle while the train is stopped and an abnormality occurs, causing passengers to escape through a window or the like, it is possible to detect an abnormality inside the vehicle based on the total value of the air spring internal pressure of the entire train and its time difference. Furthermore, according to this embodiment, it is possible to detect an evacuation behavior in which an abnormality occurs inside a vehicle and passengers move from the vehicle 101 where the abnormality occurred to the adjacent vehicle 102, and it is possible to respond to the abnormality by notifying the crew of the vehicle 101 where the air spring internal pressure value has decreased the most out of the air spring internal pressure values ​​301 and 602 as the vehicle where the abnormality has occurred. The threshold value can be set in the same manner as in the first embodiment. EXAMPLES

[0065] The third embodiment detects an abnormality occurring inside the vehicle while the vehicle is traveling. When the vehicle is traveling, the air spring internal pressure value is subject to fluctuations due to the vehicle motion in addition to fluctuations due to the evacuation behavior of passengers. To deal with this, the third embodiment employs a method of comparing the measured air spring internal pressure value with a base air spring internal pressure value obtained by analyzing past travel data and vehicle motion, and detecting an abnormality inside the vehicle based on the comparison result.

[0066] In this embodiment, an example in which an air spring internal pressure measuring device is used as a measuring means for measuring the movement of a passenger will be described as a representative example, similar to the embodiment 1. Also, in this embodiment, similar to the embodiment 1, other means can be used as the measuring means.

[0067] 8 is a configuration diagram showing a configuration example of an in-train abnormality detection system according to a third embodiment of the present invention. In this embodiment, abnormality detection logic 705 is arranged in the vehicle information control device 104 instead of the abnormality detection logic 105, an air spring internal pressure base value recording unit 710 that records a base value of an air spring internal pressure is mounted on the vehicle 101 as an air spring internal pressure base value recording device, and an air spring internal pressure base value 711 recorded in the air spring internal pressure base value recording unit 710 is output to the vehicle information control device 104. The other configurations are the same as those of the first embodiment. Hereinafter, the processing of the abnormality detection logic 705 will be described, focusing mainly on the processing different from the first embodiment.

[0068] 9 is a flowchart for explaining an example of a process of the anomaly detection logic according to the third embodiment of the present invention. This process is started by the processor activating the anomaly detection logic 705. Note that the process of step S901 is executed in parallel with step S201, and the process of step S902 is executed as a process between step S201 and step S203. Also, the processes of steps S201, S203, and S204 are similar to the processes of steps S201, S203, and S204 in FIG. 2.

[0069] In parallel with the process of acquiring the current air spring internal pressure value of each vehicle (S201), the abnormality detection logic 705 acquires the air spring internal pressure base value 711 recorded in the air spring internal pressure base value recording unit 710 from the air spring internal pressure base value recording unit 710 (S901).

[0070] Next, the abnormality detection logic 705 calculates the difference and time difference between the air spring internal pressure value and the base value of each vehicle based on the current air spring internal pressure value of each vehicle obtained in step S201 and the air spring internal pressure base value 711 obtained in step S901 (S902).

[0071] In other words, if an abnormality occurs inside the train while it is running, the air spring internal pressure value will fluctuate due to the movement of the passengers as well as the running motion of the train. For this reason, a process is executed to compare the air spring internal pressure value of each car with a base value.

[0072] Here, when no abnormality occurs, the fluctuation of the air spring internal pressure value due to the running motion of the train can be inferred using past data. Specifically, one method is to refer to past data on the air spring internal pressure value for the same vehicle, same running section, and similar occupancy rate, and calculate the base value of the air spring internal pressure value by averaging the obtained data over time for each running section. Another method is to analytically solve the vehicle vibration and the force acting on the air spring from the vehicle speed, line alignment, occupancy rate, vehicle specifications, etc., to calculate the base value of the air spring internal pressure.

[0073] For this reason, a reference value for the fluctuation in the air spring internal pressure value due to vehicle running is obtained as the air spring internal pressure base value 711, and the difference and time difference between the actual air spring internal pressure value 901 and the air spring internal pressure base value 711 are obtained, as well as the difference value and time difference between the actual air spring internal pressure value 902 and the air spring internal pressure base value 711.

[0074] Thereafter, based on the calculation results of step S902, the abnormality detection logic 705 determines whether the difference in the air spring internal pressure values ​​and the time difference in each vehicle simultaneously exceed the increase threshold or the decrease threshold, which are thresholds for determining an abnormality (S203).

[0075] If the abnormality detection logic 705 obtains a negative (NO) determination result in step S203, it returns to step S201 and repeats the processes of steps S201, S901, S902, and S203.

[0076] On the other hand, when the abnormality detection logic 705 obtains a positive (YES) determination result in step S203, it outputs the abnormality detection result 110 to the abnormality alarm device 103. For example, when an abnormality occurs in the vehicle 101 while the train is traveling, the air spring internal pressure value 111 in the vehicle 101 decreases below the air spring internal pressure base value 711 due to the evacuation behavior of the passengers (the evacuation behavior in which the passengers in the vehicle 101 move to the vehicle 102), and the air spring internal pressure value 112 in the vehicle 102 increases above the air spring internal pressure base value 711. At this time, the air spring internal pressure values ​​111 and 112 fluctuate due to the evacuation behavior of the passengers, but also fluctuate due to the movement of the vehicles 101 and 102. That is, the air spring internal pressure values ​​111 and 112 are affected by the fluctuation due to the evacuation behavior of the passengers as well as the fluctuation due to the movement of the vehicles 101 and 102.

[0077] Therefore, in the vehicle 101, the difference and the time difference between the air spring internal pressure value 111 and the air spring internal pressure base value 711 fall below the respective threshold values ​​for decrease, and in the vehicle 102, the difference and the time difference between the air spring internal pressure value 112 and the air spring internal pressure base value 711 exceed the respective threshold values ​​for increase. In this case, the abnormality detection logic 705 outputs the abnormality detection result 110 indicating that an abnormality has occurred in the vehicle 101 to the abnormality alarm device 103.

[0078] Thereafter, the abnormality alarm device 103 notifies a crew member that an abnormality has occurred in the car in which the air spring internal pressure value has decreased. For example, when an abnormality occurs in a car while the train is running, the abnormality alarm device 103 notifies a crew member that an abnormality has occurred in the car in which the air spring internal pressure value has decreased, i.e., car 101.

[0079] Fig. 10 is a characteristic diagram for explaining the air spring internal pressure and the time difference of the change in the air spring internal pressure of each vehicle in the in-train abnormality detection system according to the third embodiment of the present invention. Fig. 10(a) shows the behavior of the air spring internal pressure of vehicle 101, where the horizontal axis indicates the elapsed time and the vertical axis indicates the air spring internal pressure measured by air spring internal pressure value measuring device 108. In Fig. 10(a), the air spring internal pressure (difference) of vehicle 101 (vehicle 1) normally changes over time as air spring internal pressure value 901 in accordance with the oscillation of vehicle 101 (vehicle 1). Here, if an abnormality occurs in car 101 at time 303 while the train is running and the passengers of car 101 evacuate to car 102 (car 2), the air spring internal pressure value 901 of car 101 gradually decreases, and thereafter, a decrease (a decrease including high frequency components due to passenger movement) 904 indicating the difference between the air spring internal pressure value 901 and the air spring internal pressure base value 711 falls below a threshold value 906 indicating the air spring internal pressure threshold value. In other words, the decrease 904 goes outside the range of threshold value 906 indicating the abnormality determination value.

[0080] Fig. 10(b) shows the behavior of the air spring internal pressure of vehicle 102, where the horizontal axis indicates elapsed time and the vertical axis indicates the air spring internal pressure measured by air spring internal pressure value measuring device 109. In Fig. 10(b), the air spring internal pressure (difference) of vehicle 102 (vehicle 2) normally changes over time as air spring internal pressure value 902 in accordance with the vibration of vehicle 102 (vehicle 2). Here, if an abnormality occurs in vehicle 101 at time 303 while the train is running and the passengers of vehicle 101 evacuate to vehicle 102 (vehicle 2), the air spring internal pressure value 902 of vehicle 102 gradually increases, and thereafter, an increment 905 indicating the difference between the air spring internal pressure value 902 and air spring internal pressure base value 711 (an increment including high frequency components due to passenger movement) exceeds a threshold value 907 indicating the air spring internal pressure threshold value. That is, the increment 904 falls outside the range of the threshold 907 indicating the abnormality determination value.

[0081] FIG. 10(c) shows the behavior of the air spring internal pressure decrease time difference of the vehicle 101, where the horizontal axis indicates elapsed time and the vertical axis indicates the air spring internal pressure decrease time difference measured by the air spring internal pressure value measuring device 108. In FIG. 10(c), the air spring internal pressure decrease time difference of the vehicle 101 (vehicle 1) changes over time at a substantially constant value as an air spring internal pressure value decrease time difference 908. Here, when an abnormality occurs in the vehicle 101 at time 303 while the train is running and the passengers of the vehicle 101 evacuate to the vehicle 102 (vehicle 2), the air spring internal pressure value decrease time difference 908 of the vehicle 101 decreases rapidly and falls below a threshold value 909 indicating an air spring internal pressure value decrease time difference threshold. Thereafter, the air spring internal pressure value decrease time difference 908 continues to be below the threshold value 909, and returns to substantially the original value when the evacuation of the passengers of the vehicle 101 is completed. At this time, if the state in which the air spring internal pressure value decrease time difference 908 is below the threshold value 909 continues for more than the set time from time 303 and exceeds a duration threshold value 910 indicating the air spring internal pressure value decrease elapsed time threshold, in other words, if the air spring internal pressure value decrease time difference 908 is continuously below the threshold value 909 for more than the duration threshold value 910, the air spring internal pressure decrease time difference 908 of the vehicle 101 is determined to be abnormal.

[0082] Fig. 10(d) shows the behavior of the air spring internal pressure increase time difference of vehicle 102, where the horizontal axis indicates elapsed time and the vertical axis indicates the air spring internal pressure increase time difference measured by air spring internal pressure value measuring device 109. In Fig. 3(d), the air spring internal pressure value time difference of vehicle 102 (vehicle 2) changes over time at a substantially constant value as air spring internal pressure value increase time difference 911. Here, if an abnormality occurs in vehicle 101 at time 303 while the train is running and the passengers of vehicle 101 evacuate to vehicle 102 (vehicle 2), as air spring internal pressure value decrease time difference 908 of vehicle 101 decreases rapidly, air spring internal pressure value increase time difference 911 of vehicle 102 increases rapidly and exceeds threshold value 912 indicating the air spring internal pressure value increase time difference threshold. Thereafter, the air spring internal pressure value increase time difference 911 continues to exceed the threshold value 912, and returns to approximately the original value when the evacuation of the passengers of the vehicle 101 is completed. At this time, if the state in which the air spring internal pressure value increase time difference 911 exceeds the threshold value 912 continues for a set time or more from time 303 and exceeds a duration threshold value 913 indicating an air spring internal pressure value elapsed time threshold, in other words, if the air spring internal pressure value increase time difference 911 continuously exceeds the threshold value 912 for the duration threshold value 913 or more, the air spring internal pressure increase time difference 911 of the vehicle 102 is determined to be abnormal.

[0083] While the train is running, at time 303, an abnormality occurs in car 101, the passengers in car 101 evacuate to car 102, the air spring internal pressure value 901 of car 101 decreases, and the air spring internal pressure value 902 of car 102 increases. If all of the following conditions (1) to (4) are satisfied, the occurrence of an abnormality in car 101 is detected and information to the effect that an abnormality has occurred in car 101 is notified to the crew.

[0084] (1) The air spring internal pressure value 901 of the vehicle 101 falls below the air spring internal pressure threshold value 906 of the vehicle 101. (2) The air spring internal pressure value 902 of the vehicle 102 exceeds the air spring internal pressure threshold value 907 of the vehicle 102. (3) The air spring internal pressure decrease time difference 908 of the vehicle 101 falls below the threshold value 909 continuously for a duration threshold value 910 or more. (4) The air spring internal pressure increase time difference 911 of the vehicle 102 exceeds the threshold value 912 continuously for a duration threshold value 913 or more.

[0085] According to this embodiment, when an abnormality occurs in a vehicle while the train is running, it is possible to detect evacuation behavior in which passengers move from the vehicle 101 in which the abnormality occurred to the adjacent vehicle 102, and it is possible to respond to the abnormality by notifying the crew of the vehicle 101 in which the abnormality occurred. The threshold value can be set in the same manner as in the first embodiment. Even if the conditions (1) and (2) out of the above conditions (1) to (4) are satisfied, it is possible to detect that an abnormality has occurred in the vehicle 101, as long as the air spring internal pressure values ​​901 and 902 do not contain errors such as sudden noise or outliers. EXAMPLES

[0086] Example 4 is a method for detecting an abnormality when the train is stopped at a station and the door is opened. When the train is stopped at a station and the door is opened, a large number of passengers are expected to evacuate from the train. Since the door is opened, the method using the total air spring internal pressure value of the entire train as described in Example 2 cannot determine whether an abnormality has occurred. In order to address this, Example 4 shows a method for detecting an abnormality using an air spring internal pressure base value obtained by processing past travel data.

[0087] In this embodiment, an example in which an air spring internal pressure measuring device is used as a measuring means for measuring the movement of a passenger will be described as a representative example, similar to the embodiment 1. Also, in this embodiment, similar to the embodiment 1, other means can be used as the measuring means.

[0088] 11 is a configuration diagram showing a configuration example of an in-train abnormality detection system according to a fourth embodiment of the present invention. In this embodiment, abnormality detection logic 1005 is arranged in the vehicle information control device 104 instead of abnormality detection logic 105, an air spring internal pressure base value recording unit 1010 that records a base value of the air spring internal pressure is mounted on the vehicle 101, and the air spring internal pressure base value 1011 recorded in the air spring internal pressure base value recording unit 1010 is output to the vehicle information control device 104. The other configurations are the same as those of the first embodiment. Hereinafter, the processing of the abnormality detection logic 1005 will be described, focusing mainly on the processing different from the first embodiment.

[0089] 12 is a flowchart for explaining a processing example of the anomaly detection logic according to the fourth embodiment of the present invention. This processing is started by the processor activating the anomaly detection logic 1005. Note that, as a processing different from that in FIG. 2, the processing of step S1201 is executed in parallel with step S201, the processing of step S1202 is executed as a processing between step S201 and step S203, and the processing of steps S1203 to S1205 is executed as a processing between step S203 and step S204. Also, the processing of steps S201, S203, and S204 are similar to the processing of steps S201, S203, and S204 in FIG. 2.

[0090] In parallel with the process of acquiring the current air spring internal pressure value of each vehicle (step S201), the abnormality detection logic 1005 acquires the air spring internal pressure base value 1011 recorded in the air spring internal pressure base value recording unit 1010 from the air spring internal pressure base value recording unit 1010 (S1201).

[0091] Next, the abnormality detection logic 1005 calculates the difference and time difference between the air spring internal pressure value and the base value of each vehicle based on the current air spring internal pressure value of each vehicle obtained in step S201 and the air spring internal pressure base value 1011 obtained in step S1201 (S1202).

[0092] That is, if an abnormality occurs in the car when the door is opened, it is considered that passengers will escape from the car through the door all at once to evacuate. On the other hand, the number of passengers boarding the car in which the abnormality occurs from the station will decrease compared to normal times. At this time, the air spring internal pressure value will decrease more than the fluctuation caused by passengers getting on and off when the door is normally opened. At this time, as in the third embodiment, if no abnormality occurs, the fluctuation in the air spring internal pressure value caused by passengers getting on and off when the car door is opened can be inferred using past data. For this reason, based on the inference, the fluctuation in the air spring internal pressure value caused by the running of the car is obtained as the air spring internal pressure base value, and the difference and time difference between the actual air spring internal pressure value and the air spring internal pressure base value are calculated.

[0093] At this time, a reference value for the fluctuation in the air spring internal pressure value due to vehicle running is obtained as an air spring internal pressure base value 711, and the difference and time difference between the actual air spring internal pressure value 901 and the air spring internal pressure base value 711 are obtained, and the difference and time difference between the actual air spring internal pressure value 902 and the air spring internal pressure base value 711 are also obtained.

[0094] Thereafter, based on the calculation results of step S1202, the abnormality detection logic 1005 determines whether the difference in the air spring internal pressure values ​​and the time difference in each vehicle simultaneously exceed the increase threshold or the decrease threshold, which are thresholds for determining abnormality (S203).

[0095] If the abnormality detection logic 1005 obtains a positive (YES) determination result in step S203, it proceeds to processing of step S204, and if the abnormality detection logic 1005 obtains a negative (NO) determination result in step S203, it proceeds to processing of step S1203.

[0096] When the abnormality detection logic 1005 obtains a negative (NO) determination result in step S203, it sums up the air spring internal pressure values ​​and air spring internal pressure base values ​​of all the cars in the formation (S1203). For example, when cars 101 and 102 are a formation of cars, the abnormality detection logic 1005 sums up the air spring internal pressure values ​​111, 112 of each car acquired in step S201 as the air spring internal pressure value of all the cars in the formation, and acquires and sums up the air spring internal pressure base value 1011 from the air spring internal pressure base value recording unit 1010 as the air spring internal pressure base value of all the cars in the formation.

[0097] Next, the abnormality detection logic 1005 calculates the difference and time difference of the air spring internal pressure values ​​of all the cars in the formation based on the calculation result of step S1203 (S1204). At this time, the abnormality detection logic 1005 compares the total value of the air spring internal pressure base values ​​1011 acquired in step S1201 with the total values ​​of the air spring internal pressure values ​​111 and 112 acquired in the previous processing loop (the previous measurement cycle), and calculates the comparison result indicating the difference between the two as the time difference of the total value of the air spring internal pressure values.

[0098] Next, the abnormality detection logic 1005 determines whether or not the difference and time difference of the air spring internal pressure values ​​(total values) of all cars in the formation exceed threshold values ​​based on the calculation results in step S1204 (S1205).

[0099] If the abnormality detection logic 1005 obtains a negative (NO) determination result in step S1205, it returns to steps S201 and S1201, and repeats the processes of steps S201 to S203 and the process of S1201.

[0100] On the other hand, if an abnormality detection logic 1005 obtains an affirmative (YES) determination result in step S1203, it outputs abnormality detection result 110 to abnormality alarm device 103. For example, when an abnormality occurs in vehicle 101, some of the passengers in vehicle 101 escape through a window or the like and are trapped in vehicle 101, and the remaining passengers move from vehicle 101 to vehicle 102, the number of people who move due to the evacuation behavior of passengers is reduced compared to the first embodiment, so that abnormality detection logic 1005 outputs abnormality detection result 110 indicating that an abnormality has occurred in the train vehicle to abnormality alarm device 103 on condition that the total value of the air spring internal pressure values ​​of the entire vehicle exceeds the threshold value for the reduction.

[0101] Thereafter, the abnormality alarm device 103 notifies a crew member that an abnormality has occurred in the vehicle in which the air spring internal pressure value has decreased (S204). For example, when an abnormality has occurred in a train vehicle, the abnormality alarm device 103 notifies a crew member that an abnormality has occurred in the vehicle in which the air spring internal pressure value has decreased.

[0102] FIG. 13 is a characteristic diagram for explaining the air spring internal pressure and the air spring internal pressure decrease time difference of each car and a train car in the in-train abnormality detection system according to the fourth embodiment of the present invention. FIG. 13(a) shows the behavior of the air spring internal pressure of the car 101, where the horizontal axis indicates the elapsed time and the vertical axis indicates the air spring internal pressure measured by the air spring internal pressure value measuring device 108. In FIG. 13(a), the air spring internal pressure (difference) of the car 101 (car 1) normally changes at a substantially constant value over time as the air spring internal pressure value 1201. Here, at time 303, when the door is opened, an abnormality occurs in the car 101, and the passengers of the car 101 simultaneously escape from the door to the outside of the car, and some of the passengers of the car 102 (car 2) escape from the door to the outside of the car, and the air spring internal pressure value 1201 of the car 101 gradually decreases. The air spring internal pressure value 1201 of the vehicle 101 becomes a substantially constant value after a decrease 1202 from the air spring internal pressure base value 1011 gradually increases.

[0103] FIG. 13(b) shows the behavior of the air spring internal pressure of the vehicle 102, where the horizontal axis indicates elapsed time and the vertical axis indicates the air spring internal pressure measured by the air spring internal pressure value measuring device 109. In FIG. 13(b), the air spring internal pressure (difference) of the vehicle 102 (vehicle 2) normally changes as an approximately constant value over time as an air spring internal pressure value 1204. Here, at time 303, when the door is opened, an abnormality occurs in the vehicle 101, and the passengers of the vehicle 101 simultaneously escape from the door to the outside of the vehicle, and some of the passengers of the vehicle 102 (vehicle 2) escape from the door to the outside of the vehicle, so that the air spring internal pressure value 1204 of the vehicle 102 gradually decreases. However, the air spring internal pressure value 1204 of the vehicle 102 is gradually increased in a decrease 1205 from the air spring internal pressure base value 1011, but is smaller than the decrease 1202, and then becomes an approximately constant value.

[0104] Fig. 13(c) shows the behavior of the total air spring internal pressure of the entire train including cars 101 and 102 (total train), where the horizontal axis indicates elapsed time and the vertical axis indicates the air spring internal pressure of the entire train measured by air spring internal pressure value measuring devices 108 and 109. In Fig. 13(c), the air spring internal pressure of the entire train normally changes over time as an approximately constant value 1206 of the air spring internal pressure of the entire train. Here, at time 303, when an abnormality occurs in car 101 when the doors are opened, and the passengers of car 101 simultaneously escape from the doors to the outside of the train, and some of the passengers of car 102 (car 2) escape from the doors to the outside of the train, the air spring internal pressure total value 1206 of the entire train gradually decreases, and a decrease 1207 from air spring internal pressure base value 1011 falls below threshold value 1208 indicating the air spring internal pressure total threshold value of the entire train, and then becomes an approximately constant value.

[0105] Fig. 13(d) shows the behavior of the air spring internal pressure decrease time difference of the entire train, where the horizontal axis indicates elapsed time and the vertical axis indicates the air spring internal pressure decrease time difference measured by air spring internal pressure value measuring devices 108, 109. In Fig. 13(d), the air spring internal pressure decrease time difference of the entire train normally remains at a substantially constant value over time as air spring internal pressure decrease time difference 1209 of the entire train. Here, at time 303, when the doors are opened, if an abnormality occurs in car 101 and the passengers of car 101 all escape from the doors at once to the outside of the train, and some of the passengers of car 102 (car 2) escape from the doors to the outside of the train, then air spring internal pressure decrease time difference 1209 of the entire train will suddenly decrease and fall below threshold value 1210 indicating the air spring internal pressure decrease time difference threshold value of the entire train. Thereafter, the air spring internal pressure value decrease time difference 1209 of the entire train continues to be below the threshold value 1210, and returns to its original value when the evacuation of passengers from car 101 is completed. At this time, if the state in which the air spring internal pressure value decrease time difference 1209 of the entire train continues to be below the threshold value 1210 for more than a set time from time 303 and exceeds a duration threshold value 1211 indicating an elapsed time threshold for the air spring internal pressure value decrease time difference of the entire train, in other words, if the air spring internal pressure value decrease time difference 1209 of the entire train falls below the threshold value 1210 for more than the duration threshold value 1211 continuously, the air spring internal pressure total time difference 1209 of the entire train is determined to be abnormal.

[0106] At time 303, when the door is opened, an abnormality occurs in vehicle 101, the passengers of vehicle 101 all escape from the door to the outside, some of the passengers of vehicle 102 (vehicle 2) escape from the door to the outside, the air spring internal pressure value 1201 of vehicle 101 decreases, the air spring internal pressure value 1204 of vehicle 102 decreases, and if all of the following conditions (1) to (2) are satisfied, the occurrence of an abnormality in vehicle 101 is detected and information that an abnormality has occurred in vehicle 101 is notified to the crew.

[0107] (1) The total air spring internal pressure value 1206 of the entire train falls below the total air spring internal pressure threshold value 1208. (2) The air spring internal pressure decrease time difference 1209 of the entire train is continuously below the threshold value 1210 for a duration threshold value 1211 or more.

[0108] According to this embodiment, when the door is opened, it becomes possible to detect an evacuation behavior in which passengers simultaneously escape from the vehicle 101 in which the abnormality occurred when an abnormality occurs in the vehicle, and it becomes possible to respond to the abnormality by notifying the driver of the vehicle 101 in which the air spring internal pressure value has decreased the most among the air spring internal pressure values ​​1201 and 1204 as the vehicle in which the abnormality has occurred. Note that the threshold value can be set in the same manner as in the first embodiment.

[0109] In Examples 1 to 4, air spring internal pressure value measuring devices 108, 109 are used as multiple measuring means for measuring passenger movement in each vehicle, the vehicle information control device 104 is used as abnormality detection means for comparing each measurement value of the multiple measuring means with an abnormality determination value and detecting that an abnormality has occurred within the train when the measurement value of at least one of the multiple measuring means exceeds the range of the abnormality determination value, and the abnormality alarm device 103 is used as abnormality alarm means for notifying within the train that an abnormality has occurred within the train when the abnormality detection means detects that an abnormality has occurred within the train, but passenger movement can also be measured as a change in the number of passengers.

[0110] In this case, the multiple measurement means include multiple passenger number measurement devices that measure the change in the number of passengers in each car as the movement of passengers, and the abnormality detection means detects that an abnormality has occurred in the train when the measurement value of at least one of the multiple passenger number measurement devices falls outside the range of the abnormality determination value indicating the allowable range of the change in the number of passengers. For example, the number of passengers is calculated by dividing the load (load acting on the car when passengers board) obtained from the air spring internal pressure value 301 (load acting on the air spring) by the average weight of the passengers, and when the calculated decrease in the number of passengers (corresponding to the decrease 304) falls outside the threshold value of the number of passengers corresponding to the threshold value 306, it is determined that the decrease in the number of passengers falls outside the range of the abnormality determination value indicating the allowable range of the change in the number of passengers, and an abnormality has occurred in the train. In this way, it is possible to detect that an abnormality has occurred in the train when the change in the number of passengers in any car falls outside the range of the abnormality determination value indicating the allowable range of the change in the number of passengers.

[0111] The multiple passenger number measuring devices can be configured to include multiple air spring internal pressure value measuring devices 108, 109 that measure the air spring internal pressure values ​​acting on the air springs in each vehicle, and a vehicle information control device 104 as a passenger number calculation device that calculates the number of passengers in each vehicle (a numerical value obtained by dividing the load acting on the vehicle by the average weight of the passengers) based on each measurement value (load) of the multiple air spring internal pressure value measuring devices 108, 109 and the average weight of the passengers.

[0112] The plurality of passenger number measurement devices may be configured to measure a first absolute value indicating the fluctuation in the number of passengers in each car within a set time (the absolute values ​​of the increase and decrease in the fluctuation in the number of passengers within the set time) and a second absolute value indicating a time difference in the fluctuation in the number of passengers in each car within the set time (the absolute values ​​of the increase and decrease in the time difference in the fluctuation in the number of passengers within the set time).The abnormality detection means may be configured to detect the occurrence of an abnormality in the train when a first absolute value of the measurement values ​​of at least one of the plurality of passenger number measurement devices exceeds a first threshold value belonging to a range of abnormality determination values ​​and a second absolute value of the measurement values ​​of one passenger number measurement device exceeds a second threshold value belonging to a range of abnormality determination values.

[0113] The multiple passenger number measurement devices may be configured to include multiple tire internal pressure value measuring devices that measure the tire internal pressure values ​​acting on the tires in each vehicle, and a passenger number calculation device that calculates the number of passengers in each vehicle based on the measurement values ​​of each of the multiple tire internal pressure value measuring devices and the average weight of the passengers.

[0114] The multiple passenger number measuring devices can be configured to include multiple spring compression force measuring devices that measure the compression force of springs mounted on the body suspension system of each vehicle, and a passenger number calculation device that calculates the number of passengers in each vehicle based on the measurement values ​​of each of the multiple spring compression force measuring devices and a proportionality constant that specifies the number of passengers.

[0115] The multiple passenger number measurement devices may be configured to include multiple concentration measurement devices that measure the oxygen concentration or carbon dioxide concentration in each vehicle, and a passenger number calculation device that calculates the number of passengers in each vehicle based on the measurement values ​​of each of the multiple concentration measurement devices and a proportionality constant that specifies the number of passengers.

[0116] The multiple passenger number measuring devices can be configured to include multiple pressure measuring devices that measure the pressure acting on a pressure sensing mat (a mat incorporating a piezoelectric element) placed in the aisle of each car, and a passenger number calculation device that calculates the number of passengers in each car based on the fluctuations in the measurement values ​​of each of the multiple pressure measuring devices.

[0117] The multiple passenger number measuring devices can be configured to include a passenger number calculation device that irradiates light (infrared rays) toward a measurement target (passenger) passing through a side door or an inter-car door in each car, and calculates the number of passengers in each car based on the variation in the amount of incident light reflected from the measurement target.

[0118] It should be noted that the present invention is not limited to the above-described embodiments, and includes various modifications and equivalent configurations within the spirit and scope of the appended claims. For example, the above-described embodiments have been described in detail to clearly explain the present invention, and the present invention is not necessarily limited to those including all of the described configurations.

[0119] In addition, each of the configurations, functions, etc. described above may be realized in hardware, for example by designing some or all of them as an integrated circuit, or may be realized in software by a processor interpreting and executing a program that realizes each function.

[0120] Information such as programs, tables, files, etc. that realize each function can be stored in storage devices such as memory, hard disks, and SSDs (Solid State Drives), or recording media such as IC (Integrated Circuit) cards, SD (Secure Digital) cards, and DVDs (Digital Versatile Discs). [Explanation of symbols]

[0121] 101, 102 vehicle, 103 abnormality notification device, 104 vehicle information control device, 105 abnormality detection logic, 106, 107 air spring, 108, 109 air spring internal pressure value measuring device, 405 abnormality detection logic, 705 abnormality detection logic, 710 air spring internal pressure base value recording unit, 1005 abnormality detection logic, 1010 air spring internal pressure base value recording unit

Claims

1. a plurality of measuring means for measuring passenger movement in each car of a train having a plurality of coupled cars; an abnormality detection means for comparing a measurement value of each of the plurality of measurement means with an abnormality determination value, and detecting that an abnormality has occurred in the train when a measurement value of at least one of the plurality of measurement means exceeds a range of the abnormality determination value, The abnormality detection means includes: a vehicle information control device that acquires each measurement value obtained by measurement by each measurement means from each measurement means in each of the vehicles, and calculates a difference in air spring internal pressure values ​​indicating a time change of each measurement value based on each of the acquired measurement values, The vehicle information control device includes: The in-train anomaly detection system detects an abnormality in one of the plurality of cars when four conditions are satisfied, including a first condition that a decrease indicating the difference in the measurement values ​​in one of the plurality of cars falls below a first measurement threshold, and a second condition that an increase indicating the difference in the measurement values ​​in an adjacent car adjacent to the one of the cars exceeds a second measurement threshold.

2. In the train anomaly detection system according to claim 1, Each of the plurality of measuring means is an air spring internal pressure value measuring device for measuring an air spring internal pressure value acting on an air spring in each of the vehicles, The abnormality detection means includes: a vehicle information control device that acquires each air spring internal pressure value measured by each of the air spring internal pressure value measuring devices in each of the vehicles, and calculates a difference in air spring internal pressure values ​​indicating a time change of each of the air spring internal pressure values ​​based on each of the acquired air spring internal pressure values, The vehicle information control device includes: The in-train abnormality detection system detects an abnormality in one of the plurality of cars when two conditions are satisfied, including a first condition that a decrease indicating the difference in the air spring internal pressure values ​​in one of the plurality of cars falls below a first air spring internal pressure threshold, and a second condition that an increase indicating the difference in the air spring internal pressure values ​​in an adjacent car adjacent to the one of the cars exceeds a second air spring internal pressure threshold.

3. In the train anomaly detection system according to claim 2, The vehicle information control device further calculates an air spring internal pressure value time difference indicating a change in each of the air spring internal pressure values ​​for each measurement cycle, an abnormality detection system within a train, characterized in that an abnormality in one of the cars is detected when four conditions are satisfied, including the first condition, the second condition, a third condition that the air spring internal pressure value time difference in the one of the cars is continuously below a first air spring internal pressure value time difference threshold for a first duration threshold or more, and a fourth condition that the air spring internal pressure value time difference in the adjacent car is continuously above a second air spring internal pressure value time difference threshold for a second duration threshold or more.

4. In the train anomaly detection system according to claim 3, The vehicle information control device includes: if the four conditions are not satisfied, summing up the air spring internal pressure values ​​acquired from the air spring internal pressure value measuring devices of all cars belonging to the train, and calculating, based on the sum of the air spring internal pressure values, an air spring internal pressure sum difference indicating a change over time in the sum of the air spring internal pressure values ​​and an air spring internal pressure sum time difference indicating a change in the sum of the air spring internal pressure values ​​for each measurement cycle; The in-train abnormality detection system detects an abnormality in one of the cars when two conditions are satisfied: a fifth condition that the air spring internal pressure total value difference falls below a decreased air spring internal pressure total threshold value; and a sixth condition that the air spring internal pressure total value time difference falls continuously below the air spring internal pressure total value time difference threshold value for a period of time equal to or longer than a third duration threshold value.

5. In the train anomaly detection system according to claim 1, Each of the plurality of measuring means is an air spring internal pressure value measuring device for measuring an air spring internal pressure value acting on an air spring in each of the vehicles, The abnormality detection means includes: an air spring internal pressure base value recording device that records an air spring internal pressure base value that is a reference for the air spring internal pressure value; a vehicle information control device that acquires each air spring internal pressure value measured by each air spring internal pressure value measuring device in each of the vehicles and acquires the air spring internal pressure base value from the air spring internal pressure base value recording device, and calculates an air spring internal pressure difference indicating a time change of each of the air spring internal pressure values ​​and an air spring internal pressure value time difference indicating a change in each of the air spring internal pressure values ​​for each measurement cycle based on a difference between each of the acquired air spring internal pressure values ​​and the air spring internal pressure base value, The vehicle information control device includes: a first condition that a decrease indicating a difference between the air spring internal pressure value and the air spring internal pressure base value in any one of the plurality of vehicles falls below a first air spring internal pressure threshold; a second condition that an increase indicating a difference between the air spring internal pressure value and the air spring internal pressure base value in an adjacent vehicle adjacent to the one vehicle exceeds a second air spring internal pressure threshold; a third condition that an air spring internal pressure value decrease time difference, among the air spring internal pressure value time differences in the one vehicle, is continuously below the air spring internal pressure value decrease time difference threshold for a first duration threshold or more; and a fourth condition that an air spring internal pressure value increase time difference, among the air spring internal pressure value time differences in the adjacent vehicle, is continuously above the air spring internal pressure value increase time difference threshold for a second duration threshold or more.

6. The in-train anomaly detection system according to claim 5, The vehicle information control device includes: if the four conditions are not satisfied, add up the air spring internal pressure values ​​acquired from the air spring internal pressure value measuring devices and the air spring internal pressure base values ​​acquired from the air spring internal pressure base value recording devices of all cars belonging to the train, and calculate, based on the sum of the air spring internal pressure values, an air spring internal pressure sum difference indicating a time change in the sum of the air spring internal pressure values ​​and an air spring internal pressure sum time difference indicating a change in the sum of the air spring internal pressure values ​​for each measurement cycle; The in-train abnormality detection system detects an abnormality in one of the cars when two conditions are satisfied: a fifth condition that the air spring internal pressure total value difference falls below a decrease in the air spring internal pressure total threshold; and a sixth condition that the air spring internal pressure total value decrease time difference, among the air spring internal pressure total value time differences, falls continuously below the air spring internal pressure total value decrease time difference threshold for a period of time equal to or longer than a third duration threshold.

7. A train having a plurality of coupled cars, comprising: a plurality of measuring means for measuring the movement of passengers in each car; an abnormality detection means for comparing a measurement value of each of the plurality of measurement means with an abnormality determination value, and detecting that an abnormality has occurred in the train when a measurement value of at least one of the plurality of measurement means exceeds a range of the abnormality determination value, The abnormality detection means includes: a vehicle information control device that acquires each measurement value obtained by measurement by the measurement means from each measurement means in each of the vehicles, and calculates a measurement value time difference indicating a change in each measurement value for each measurement cycle based on each of the acquired measurement values, The vehicle information control device includes: a first condition that the measurement value time difference in any one of the plurality of cars is continuously below a first measurement value time difference threshold by a first duration threshold or more, and a second condition that the measurement value time difference in the adjacent car is continuously above a second measurement value time difference threshold by a second duration threshold or more.

8. The train anomaly detection system according to claim 7, Each of the plurality of measuring means is an air spring internal pressure value measuring device for measuring an air spring internal pressure value acting on an air spring in each of the vehicles, The abnormality detection means includes: a vehicle information control device that acquires each air spring internal pressure value measured by each of the air spring internal pressure value measuring devices in each of the vehicles, and calculates an air spring internal pressure value time difference that indicates a change in each of the air spring internal pressure values ​​for each measurement cycle based on each of the acquired air spring internal pressure values, The vehicle information control device includes: The in-train anomaly detection system detects an abnormality in one of the plurality of cars when two conditions are satisfied: a first condition that the air spring internal pressure value time difference in any one of the plurality of cars is continuously below a first air spring internal pressure value time difference threshold for a first duration threshold or more, and a second condition that the air spring internal pressure value time difference in the adjacent car is continuously above a second air spring internal pressure value time difference threshold for a second duration threshold or more.

9. A train anomaly detection method in a train anomaly detection system including a plurality of measuring means and an anomaly detection means, a measurement step in which each of the plurality of measurement means measures a movement of passengers in each car of a train having a plurality of coupled cars; an abnormality detection step in which the abnormality detection means compares the measurement values ​​of the plurality of measurement means with an abnormality determination value, and detects that an abnormality has occurred in the train when the measurement value of at least one of the plurality of measurement means exceeds a range of the abnormality determination value; The abnormality detection step calculates a difference in measurement values ​​indicating changes over time of each measurement value based on each measurement value measured by each of the multiple measurement means in the measurement step, and detects an abnormality in one of the multiple cars when two conditions are satisfied from the calculated result, including a first condition that a decrease indicating the difference in the air spring internal pressure value in any one of the multiple cars falls below a first air spring internal pressure threshold, and a second condition that an increase indicating the difference in the air spring internal pressure value in an adjacent car adjacent to the one of the cars exceeds a second air spring internal pressure threshold.

10. The train abnormality detection method according to claim 9, Each of the plurality of measuring means is In the measuring step, an internal pressure value of an air spring acting on an air spring in each of the vehicles is measured, The abnormality detection means includes: In the abnormality detection step, an air spring internal pressure difference indicating a time change of each of the air spring internal pressure values ​​measured by each of the plurality of measuring means in the measurement step and an air spring internal pressure value time difference indicating a change of each of the air spring internal pressure values ​​for each measurement cycle are calculated based on each of the air spring internal pressure values ​​measured by each of the plurality of measuring means in the measurement step, and from the calculated results, a first condition that a decrease indicating the difference of the air spring internal pressure values ​​in any one of the plurality of vehicles falls below a first air spring internal pressure threshold value and a time difference of the air spring internal pressure values ​​in an adjacent vehicle adjacent to the one vehicle are detected. a second condition that an increment indicating a difference in air spring internal pressure values ​​exceeds a second air spring internal pressure threshold; a third condition that the air spring internal pressure value time difference in the one car is continuously below a first air spring internal pressure value time difference threshold for a first duration threshold or more; and a fourth condition that the air spring internal pressure value time difference in the adjacent car is continuously above a second air spring internal pressure value time difference threshold for a second duration threshold or more.

11. The train abnormality detection method according to claim 10, The abnormality detection means includes: a fifth condition that the air spring internal pressure total difference falls below a decreased air spring internal pressure total threshold value and a sixth condition that the air spring internal pressure total time difference falls below an air spring internal pressure total time difference threshold value for a third duration threshold or more continuously in the abnormality detection step, the air spring internal pressure values ​​of all cars belonging to the train are summed, and an air spring internal pressure total difference indicating a time change of the sum of the air spring internal pressure values ​​and an air spring internal pressure total time difference indicating a change of the sum of the air spring internal pressure values ​​for each measurement cycle are calculated based on the sum of the air spring internal pressure values, and an abnormality in one of the cars is detected if two conditions are satisfied from the calculated results, including a fifth condition that the air spring internal pressure total difference falls below a decreased air spring internal pressure total threshold value and a sixth condition that the air spring internal pressure total time difference falls below an air spring internal pressure total time difference threshold value for a third duration threshold value or more.

12. The train abnormality detection method according to claim 9, The abnormality detection means further includes a recording step of recording an air spring internal pressure base value serving as a reference for the air spring internal pressure value in an air spring internal pressure base value recording device, Each of the plurality of measuring means is In the measuring step, an internal pressure value of an air spring acting on an air spring in each of the vehicles is measured, The abnormality detection means includes: In the abnormality detection step, each air spring internal pressure value measured by each of the plurality of measuring means in the measurement step and the air spring internal pressure base value recorded in the recording step are acquired, and an air spring internal pressure difference indicating a time change of each air spring internal pressure value and an air spring internal pressure value time difference indicating a change of each air spring internal pressure value for each measurement cycle are calculated based on a difference between each of the acquired air spring internal pressure values ​​and the air spring internal pressure base value, and from the calculated results, a first condition that a decrease indicating a difference between the air spring internal pressure value and the air spring internal pressure base value in any one of the plurality of vehicles falls below a first air spring internal pressure threshold value and a second condition that a decrease indicating a difference between the air spring internal pressure value and the air spring internal pressure base value in any one of the plurality of vehicles falls below a first air spring internal pressure threshold value and a second condition that an increment indicating a difference between the air spring internal pressure value and the air spring internal pressure base value in an adjacent car exceeds a second air spring internal pressure threshold; a third condition that an air spring internal pressure value decrease time difference, among the air spring internal pressure value time differences in the one car, is continuously below the air spring internal pressure value decrease time difference threshold for a first duration threshold or more; and a fourth condition that an air spring internal pressure value increase time difference, among the air spring internal pressure value time differences in the adjacent car, is continuously above the air spring internal pressure value increase time difference threshold for a second duration threshold or more.

13. The train abnormality detection method according to claim 12, The abnormality detection means includes: a fifth condition that the air spring internal pressure total difference is below an air spring internal pressure total threshold value for a decrease in the air spring internal pressure; and a sixth condition that an air spring internal pressure value decrease time difference, of the air spring internal pressure total value time difference, is continuously below an air spring internal pressure value decrease time difference threshold value for a period not less than a third duration threshold value.

14. The train abnormality detection method according to claim 9, Each of the plurality of measuring means is In the measuring step, an internal pressure value of an air spring acting on an air spring in each of the vehicles is measured, The abnormality detection means includes: The abnormality detection step calculates an air spring internal pressure value difference indicating a time change of each air spring internal pressure value and an air spring internal pressure value time difference indicating a change of each air spring internal pressure value for each measurement cycle based on each air spring internal pressure value measured by each of the plurality of measuring means in the measurement step, and detects an abnormality in one of the plurality of cars when two conditions are satisfied from the calculated results, including a first condition that a decrease indicating the difference in the air spring internal pressure values ​​in any one of the plurality of cars falls below a first air spring internal pressure threshold, and a second condition that an increase indicating the difference in the air spring internal pressure values ​​in an adjacent car adjacent to the one of the cars exceeds a second air spring internal pressure threshold.

15. A method for detecting an abnormality in a train in a train abnormality detection system having a plurality of measuring means and an abnormality detection means, comprising: a measurement step in which each of the plurality of measurement means measures a movement of passengers in each car of a train having a plurality of coupled cars; an abnormality detection step in which the abnormality detection means compares the measurement values ​​of the plurality of measurement means with an abnormality determination value, and detects that an abnormality has occurred in the train when the measurement value of at least one of the plurality of measurement means exceeds a range of the abnormality determination value; a measurement value time difference indicating a change in each measurement value for each measurement cycle based on each measurement value measured by each of the plurality of measuring means in the measurement step, and detecting an abnormality in one of the plurality of cars when two conditions are satisfied from the calculated result, including a first condition that the measurement value time difference in any one of the plurality of cars is continuously below a first measurement value time difference threshold for a first duration threshold or more, and a second condition that the measurement value time difference in the adjacent car is continuously above a second measurement value time difference threshold for a second duration threshold or more.

Images