Wind speed correction device, wind speed correction system and operation management system, wind speed correction method and wind speed correction program
The wind speed correction device addresses the issue of uniform wind speed corrections across different vehicle types by incorporating vehicle-specific data to adjust wind speed values, improving operational efficiency and reducing excessive restrictions.
Patent Information
- Application Number
- JP2023080347
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2023-05-15
- Publication Date
- 2024-11-27
Smart Images

Figure 2024164684000001_ABST
Abstract
Description
[Technical field]
[0001] The present disclosure relates to a wind speed correction device, a wind speed correction system, a traffic control system, a wind speed correction method, and a wind speed correction program that correct the wind speed received by a vehicle from an anemometer installed along a railway track. [Background technology]
[0002] When trains are operated in strong winds, there is a possibility that the trains may overturn or derail depending on the wind direction. For this reason, wind speed monitoring systems are adopted for trains, in which wind vane and anemometers are installed along the tracks and train operation is managed according to the observation values of the anemometers. Patent Document 1 discloses an operation restriction system that calculates a conversion factor for each wind direction that quantifies the effect of wind on train operation based on the critical wind speed for overturning for each wind direction angle of the train, and restricts train operation based on an estimated wind speed calculated by multiplying the wind speed observed by the anemometer by the conversion factor corresponding to the wind angle observed by the anemometer. [Prior art documents] [Patent documents]
[0003] [Patent Document 1] JP 2013-159259 A Summary of the Invention [Problem to be solved by the invention]
[0004] In the technology described in Patent Document 1, when there are multiple types of vehicles running on the track, the vehicle's critical wind speed for overturning is determined by the minimum value of the critical wind speed distribution for each wind angle obtained for each type of vehicle. In other words, for each wind angle, the vehicle's critical wind speed for overturning is determined to be the smallest critical wind speed for the multiple vehicle types. Then, using the critical wind speed for overturning at each wind angle, a conversion factor for each wind angle common to all vehicles passing the installation position of the anemometer is determined. The evaluated wind speed is calculated by multiplying the observed wind speed by the conversion factor corresponding to the wind angle observed by the anemometer. The evaluated wind speed calculated in this way is the same regardless of the type of vehicle passing the installation position of the anemometer. Therefore, even if the calculated evaluated wind speed is a wind speed that imposes operation restrictions on one type of vehicle, it may not be a wind speed that imposes operation restrictions on another type of vehicle. In this case, there is a problem that excessive operational restrictions are imposed on the other types of vehicles, which in turn reduces the efficiency of train operation management.
[0005] The present disclosure has been made in consideration of the above, and aims to obtain a wind speed correction device that can correct the wind speed value according to the wind direction observed by an anemometer to a wind speed value that takes into account the type of vehicle that will be experienced by vehicles passing the installation position of the anemometer. [Means for solving the problem]
[0006] In order to solve the above-mentioned problems and achieve the object, the wind speed correction device according to the present disclosure includes a wind direction and speed observation data receiver, a passing train information receiver, and a wind speed correction unit. The wind direction and speed observation data receiver receives wind direction and speed observation data including wind direction observation values and wind speed observation values, which are wind direction and wind speed values observed by an anemometer installed along a railway track. The passing train information receiver receives passing train information including vehicle type information, which is the type of vehicle passing through the track. The wind speed correction unit calculates a wind speed correction value by correcting the wind speed observation value based on correction amount information corresponding to the vehicle type information and the wind direction observation value among a plurality of correction amount information in which the wind direction and the correction amount for correcting the wind speed observation value are associated with each vehicle type. Effect of the Invention
[0007] According to the present disclosure, it is possible to correct a wind speed value according to the wind direction observed by an anemometer to a wind speed value that takes into account the type of vehicle that will be subjected to wind passing the installation location of the anemometer. [Brief description of the drawings]
[0008] [Figure 1] FIG. 1 is a schematic diagram illustrating an example of the configuration of a train control management system including a wind speed correction device according to a first embodiment. [Diagram 2] FIG. 13 is a diagram showing an example of a correction amount table; [Diagram 3] FIG. 1 is a diagram for explaining a method for calculating combined wind direction and speed information; [Figure 4] 1 is a flowchart showing an example of a procedure of a correction amount table generating method according to the first embodiment. [Diagram 5] A flowchart showing an example of a procedure of a wind speed correction method according to the first embodiment. [Figure 6] FIG. 1 is a diagram for explaining a wind speed correction method in a train control management system according to the first embodiment. [Figure 7] FIG. 1 is a diagram for explaining a wind speed correction method in a train control management system according to the first embodiment. [Figure 8] FIG. 1 is a diagram for explaining a wind speed correction method in a train control management system according to the first embodiment. [Figure 9] FIG. 1 is a diagram for explaining a wind speed correction method in the prior art. [Figure 10] FIG. 1 is a schematic diagram illustrating an example of a configuration of a train control management system according to a second embodiment. [Figure 11] FIG. 2 is a diagram showing an example of the configuration of a learning unit of a wind direction and speed data accumulation device; [Figure 12] A flowchart showing an example of a processing procedure of a learning method of the learning unit. [Figure 13] FIG. 2 is a diagram showing an example of the configuration of an abnormality determination unit of a wind direction and speed data accumulation device; [Figure 14]A flowchart showing an example of a processing procedure of an abnormality determination method of the abnormality determination unit. [Figure 15] FIG. 1 is a diagram showing an example of the configuration of a learning unit of a wind direction and speed data accumulation device. [Figure 16] FIG. 2 is a diagram showing an example of a neural network used by the model generation unit; [Figure 17] A flowchart showing an example of a processing procedure of a learning method performed by a learning unit. [Figure 18] FIG. 2 is a diagram showing an example of the configuration of an abnormality determination unit of a wind direction and speed data accumulation device; [Figure 19] A flowchart showing an example of a processing procedure of an abnormality determination method of the abnormality determination unit. [Figure 20] FIG. 1 is a block diagram showing an example of a configuration of a computer system that realizes a wind speed correction device according to a first embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0009] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A wind speed correction device, a wind speed correction system, a traffic control system, a wind speed correction method, and a wind speed correction program according to embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings.
[0010] Embodiment 1 1 is a diagram illustrating an example of the configuration of a train control management system including a wind speed correction device according to embodiment 1. The train control management system 1 includes a surrounding area information providing device 10, a correction amount table generating device 20, an anemometer 30, a traffic management device 40, a wind speed correction device 50, and an information transmission device 70. The correction amount table generating device 20, the anemometer 30, the traffic management device 40, the wind speed correction device 50, and the information transmission device 70 are connected to each other via a communication line such as a network.
[0011] The surrounding area information providing device 10 is a device that provides surrounding area information, which is information about the topography and structures of the surrounding area around the installation position of the anemometer 30. The surrounding area information includes map information of the surrounding area and structure information, which is information including the size of the structures. The map information is information having the layout state and altitude of the structures in the surrounding area. The map information may be map data including the layout state and altitude of the structures in the surrounding area, or may be altitude data, which is data indicating an aerial photograph and altitude. The structure information includes size information, including the height, width, and depth, of the structures included in the surrounding area shown in the map information. The structure information may further include shape information, which is information about the shape of the structures.
[0012] The correction amount table generating device 20 is a device that generates a correction amount table for each type of vehicle 110, in which a correction amount for correcting the value of the wind speed observed by the anemometer 30 to the wind speed value received by the vehicle 110 passing the installation position of the anemometer 30 is calculated for each observed wind direction value. Hereinafter, the value of the wind speed observed by the anemometer 30 is referred to as a wind speed observation value, and the value of the wind direction observed by the anemometer 30 is referred to as a wind direction observation value. The correction amount table corresponds to the correction amount information, and the correction amount table generating device 20 corresponds to a correction amount information generating device. The correction amount table generating device 20 includes a surrounding area information receiving unit 21, a surrounding area information storage unit 22, a vehicle detailed information receiving unit 23, a vehicle detailed information storage unit 24, a difference determining unit 25, a correction amount table generating unit 26, and a correction amount table transmitting unit 27.
[0013] The surrounding area information receiving unit 21 receives surrounding area information from the surrounding area information providing device 10. The surrounding area information receiving unit 21 receives surrounding area information about the installation positions of the anemometers 30 for which the correction amount table generating device 20 generates a correction amount table.
[0014] The surrounding area information storage unit 22 stores the surrounding area information received from the surrounding area information providing device 10. As described later, in this example, if there is a difference between the received surrounding area information and the surrounding area information used to generate the latest correction amount table, the correction amount table is regenerated, so it is sufficient for the surrounding area information storage unit 22 to store at least the surrounding area information used to generate the latest correction amount table. The surrounding area information storage unit 22 stores the surrounding area information for each installation position of the anemometer 30.
[0015] The vehicle detail information receiver 23 receives vehicle detail information, which is information about the vehicle 110 passing the installation position of the anemometer 30 for which the correction amount table is calculated, transmitted from the traffic management device 40. The vehicle detail information includes vehicle type information indicating the type of the vehicle 110, the weight of the vehicle 110 indicated in the vehicle type information, the shape of the vehicle 110 indicated in the vehicle type information, and the like. The type of the vehicle 110 is classified according to the model of the vehicle 110, and is also classified according to the number of cars in the train even for the same model of the vehicle 110. In other words, the type of the vehicle 110 is determined by the combination of the model and the number of cars in the train.
[0016] The vehicle detail information storage unit 24 stores the vehicle detail information received from the traffic management device 40. As with the surrounding area information storage unit 22, the vehicle detail information storage unit 24 is required to store at least the vehicle detail information used to generate the latest correction amount table. The vehicle detail information storage unit 24 stores the vehicle detail information for each installation position of the anemometer 30.
[0017] The difference determination unit 25 determines whether or not a correction amount table needs to be generated based on the surrounding area information stored in the surrounding area information storage unit 22 and the vehicle detail information stored in the vehicle detail information storage unit 24, the surrounding area information received from the surrounding area information providing device 10, and the vehicle detail information received from the traffic management device 40. Specifically, the difference determination unit 25 instructs the correction amount table generation unit 26 to generate a correction amount table when the surrounding area information received from the surrounding area information providing device 10 and stored therein differs from the surrounding area information stored in the surrounding area information storage unit 22, or when the vehicle detail information received from the traffic management device 40 and stored therein differs from the vehicle detail information stored in the vehicle detail information storage unit 24. When the correction amount table is generated for the first time, the surrounding area information is not stored in the surrounding area information storage unit 22, and the vehicle detail information is not stored in the vehicle detail information storage unit 24. In this case, the correction amount table is compared with the state where the information is not stored. Therefore, when surrounding area information is received for the first time from the surrounding area information providing device 10 or when vehicle detail information is received for the first time from the traffic management device 40, the correction amount table generation unit 26 will be instructed to generate a correction amount table.
[0018] When the correction amount table generating unit 26 receives an instruction to generate a correction amount table from the difference determining unit 25, the correction amount table generating unit 26 performs an analysis using a simulation based on the surrounding area information received from the surrounding area information providing device 10 and the vehicle detailed information received from the traffic management device 40, and generates a correction amount table. In one example, the correction amount table generating unit 26 simulates a correction amount that is a ratio of the wind speed received by the vehicle 110 to the wind speed at the installation position of the anemometer 30 in each wind direction, based on the weight and shape of the vehicle 110 and the map information and structure information of the surrounding area. In one example, the vehicle 110 is a vehicle 110 that passes through the track 100 near the installation position of the anemometer 30. The correction amount table generating unit 26 performs this simulation process for each type of vehicle 110, thereby generating a correction amount table for each type of vehicle 110. In addition, when there are multiple anemometers 30, the correction amount table generating unit 26 generates a correction amount table for each of the multiple anemometers 30. In this case, a correction amount table is associated with each anemometer 30. In this manner, the correction amount table is information that associates the wind direction with a correction amount for correcting the observed wind speed value to the wind speed of the wind experienced by the vehicle 110 traveling on the track 100 for each type of vehicle 110.
[0019] The correction amount table may be generated for each type of vehicle 110. For this reason, the correction amount table generating unit 26 simulates the correction amount based on the surrounding area information and the vehicle detailed information, but may simulate the correction amount based on at least the vehicle detailed information. However, as in the above example, by performing a simulation using the surrounding area information in addition to the vehicle detailed information, the calculation accuracy of the wind speed value received by the vehicle 110 passing the installation position of the anemometer 30 is improved. In addition, by performing a simulation using not only size information of the structure but also shape information as the structure information, the calculation accuracy of the wind speed value received by the vehicle 110 passing the installation position of the anemometer 30 is improved. The correction amount table generating unit 26 corresponds to the correction amount information generating unit.
[0020] In addition, in this example, by providing the difference determination unit 25, the correction amount table is automatically generated or regenerated every time the vehicle details information or the surrounding area information is updated, but the correction amount table may be generated or regenerated in accordance with instructions from the user.
[0021] There may be a discrepancy between the wind speed observation value of the anemometer 30 and the wind speed value experienced by the vehicle 110 passing the installation position of the anemometer 30. For this reason, it is desirable to consider driving restrictions on the vehicle 110 based on a wind speed correction value obtained by correcting the wind speed observation value to the wind speed value experienced by the vehicle 110. Therefore, in the first embodiment, the correction amount table generating unit 26 calculates a correction amount for correcting the wind speed observation value to the wind speed value experienced by the vehicle 110 for each wind direction and creates a table.
[0022] FIG. 2 is a diagram showing an example of a correction amount table. FIG. 2 shows a case where the vehicle type information passing the installation position of the anemometer 30 is three types, namely, vehicle type information A, vehicle type information B, and vehicle type information C. As shown in FIG. 2, the correction amount table 260 is generated for each type of vehicle 110, that is, for each vehicle type information. That is, the correction amount table 260 has a correction amount table 261A for the vehicle type information A, a correction amount table 261B for the vehicle type information B, and a correction amount table 261C for the vehicle type information C. Each correction amount table 261A, 261B, and 261C has items of wind direction and a correction amount corresponding to the wind direction. That is, a correction amount for each wind direction is associated. The wind direction is a relative wind direction taking into account the speed of the vehicle 110 as described later. The correction amount is a coefficient for correcting the wind speed observation value at the anemometer 30 to the wind speed value received by the vehicle 110 passing through the track 100. In FIG. 2, correction amount table 260 is a broad correction amount table generated corresponding to an anemometer 30, and correction amount tables 261A, 261B, and 261C are narrow correction amount tables generated for each vehicle type information of the position of the anemometer 30.
[0023] Returning to Fig. 1, when the correction amount table is generated by the correction amount table generation unit 26, the correction amount table transmission unit 27 transmits the correction amount table to the wind speed correction device 50. When there are multiple anemometers 30, the correction amount table is associated with each anemometer 30. The correction amount table transmission unit 27 corresponds to the correction amount information transmission unit.
[0024] The anemometer 30 is installed along the railway track 100 and is an instrument that observes the wind direction and wind speed at a set period. In one example, the anemometer 30 observes the wind direction and wind speed every second, and transmits the observed wind direction and wind speed values as the observation results to the wind speed correction device 50. Hereinafter, the collected wind direction and wind speed observation values may be referred to as wind direction and wind speed observation data.
[0025] Any anemometer capable of measuring wind direction and speed may be used as the anemometer 30. As an example, a propeller-type anemometer that measures wind speed by the number of rotations of a propeller and measures wind direction by the orientation of a wind speed sensor may be used.
[0026] The traffic management device 40 is a device that manages the operation of vehicles 110 on each railway line. Here, the traffic management device 40 transmits to the correction amount table generating device 20 vehicle detailed information on all vehicle type information that travels through the installation position of the anemometer 30. The traffic management device 40 transmits to the wind speed correction device 50 passing train information including information for identifying the anemometer 30, and vehicle type information and traveling speed information of the vehicle 110 that will soon pass through the installation position of the anemometer 30.
[0027] The traffic management device 40 manages the traffic of the vehicle 110 using the wind speed corrected by the wind speed correction device 50 and the wind direction observed by the anemometer 30. Specifically, when the traffic management device 40 receives wind direction and speed correction data including the wind direction and speed of the wind received by the vehicle 110 of the vehicle type information that will soon pass the installation position of the anemometer 30 from the wind speed correction device 50, the traffic management device 40 judges whether the wind direction and speed correction data satisfies the conditions of the driving restriction according to the vehicle type information of the vehicle 110 that will soon pass. If the wind direction and speed correction data satisfies the conditions of the driving restriction, the traffic management device 40 transmits a driving restriction instruction such as slow driving or stopping driving according to the conditions to the vehicle 110 that will soon pass. If the wind direction and speed correction data does not satisfy the conditions of the driving restriction, the traffic management device 40 does not do anything in particular to the vehicle 110 that will soon pass. In other words, the traffic management device 40 manages the traffic of the vehicle 110 based on the wind direction and speed correction data. In addition, when multiple anemometers 30 are installed, the wind direction and wind speed correction data includes information for identifying the anemometer 30 on which the wind direction and wind speed correction data is based. This enables the traffic management device 40 to identify which anemometer 30 the wind direction and wind speed correction data is for.
[0028] The wind speed correction device 50 includes a passing train information receiving unit 51, a wind direction and speed observation data receiving unit 52, a correction amount table receiving unit 53, a correction amount table memory unit 54, a vehicle direction information memory unit 55, a correction amount table selection unit 56, a wind direction and speed synthesis unit 57, a wind speed correction unit 58, and a wind direction and speed data transmission unit 59.
[0029] The passing train information receiver 51 receives passing train information from the traffic management device 40. The passing train information includes information for identifying the anemometer 30 and vehicle type information that is the type of vehicle 110 that will soon pass over the track 100 near the installation position of the anemometer 30. In this example, the passing train information also includes running speed information that is the running speed of the vehicle 110 that will soon pass.
[0030] The wind direction and speed observation data receiving unit 52 receives wind direction and speed observation data observed by the anemometer 30. The wind direction and speed observation data includes a wind direction observation value and a wind speed observation value.
[0031] The correction amount table receiving unit 53 receives the correction amount table from the correction amount table generating device 20 .
[0032] The correction amount table storage unit 54 stores the correction amount table received by the correction amount table receiving unit 53. The correction amount table is stored for each vehicle type information as described above. Since the wind speed correction device 50 normally calculates wind direction and wind speed correction data for the installation positions of multiple anemometers 30, the correction amount table is stored in the correction amount table storage unit 54 in association with information identifying the anemometer 30. However, in the following, for the sake of simplicity of explanation, a case will be taken as an example in which a correction amount table for each vehicle type information for one anemometer 30 is stored in the correction amount table storage unit 54.
[0033] The vehicle direction information storage unit 55 stores the vehicle direction, which is the extension direction of the track 100 at the installation position of the anemometer 30. The vehicle direction can be determined arbitrarily. In one example, the angle of the vehicle direction can be expressed by taking the north direction as 0° and the clockwise direction as the positive direction.
[0034] The correction amount table selection unit 56 selects a correction amount table from the correction amount table storage unit 54, which corresponds to a combination of vehicle type information included in the passing train information received by the passing train information receiving unit 51. The correction amount table selection unit 56 selects a correction amount table when the selected correction amount table is updated or when the passing train information is updated.
[0035] The wind direction and wind speed synthesis unit 57 calculates a wind direction synthesis value and a wind speed synthesis value, which are the wind direction and the wind speed based on the traveling speed information of the vehicle 110, from the wind direction observation value and the wind speed observation value received from the wind direction and wind speed observation data receiving unit 52. Specifically, the wind direction and wind speed synthesis unit 57 calculates wind direction and wind speed synthesis information indicating the wind direction and wind speed value based on the traveling speed information of the vehicle 110 from the wind direction observation value and the wind speed observation value, and calculates the wind direction synthesis value, which is the wind direction value of the wind acting on the vehicle 110, and the wind speed synthesis value, which is the wind speed value acting on the vehicle 110, from the wind direction and wind speed synthesis information. FIG. 3 is a diagram for explaining a method of calculating the wind direction and wind speed synthesis information. The wind obtained from the anemometer 30 is represented by a vector w. Here, the wind direction observation value with the north direction being 0° is θ0, and the wind speed observation value is w. When the vehicle 110 is traveling in a windless state, it receives a wind of vector -V, which is the same magnitude as the traveling speed information V and in the opposite direction. For this reason, while the vehicle 110 is traveling through the wind, it is subjected to wind of vector W, which is the sum of vector w and vector -V. The wind direction of this vector W is θ1, and the wind speed is W. Here, the vector W, which is the sum of vector w and vector -V, becomes a vector that represents the combined wind direction and wind speed information. The magnitude W of the vector W that represents the combined wind direction and wind speed information becomes the combined wind speed value, and the direction θ1 becomes the combined wind direction value.
[0036] In this way, the wind direction and wind speed synthesis unit 57 synthesizes the traveling speed information of the vehicle 110 and the wind speed to calculate wind direction and wind speed synthesis information, which is the synthesis speed of the wind acting on the vehicle 110, and calculates the wind direction value after synthesis acting on the vehicle 110 from the wind direction and wind speed synthesis information, i.e., the angle from the reference position, as the wind direction synthesis value. The wind direction and wind speed synthesis unit 57 also calculates the magnitude of the wind direction and wind speed synthesis information as the wind speed value after synthesis acting on the vehicle 110, i.e., the wind speed synthesis value. Data including the wind direction synthesis value and the wind speed synthesis value is sometimes referred to as wind direction and wind speed synthesis data. The wind direction and wind speed synthesis unit 57 passes the wind direction and wind speed synthesis data to the wind speed correction unit 58.
[0037] Returning to FIG. 1, the wind speed correction unit 58 calculates a wind speed correction value by correcting the wind speed observation value based on a correction amount table corresponding to the vehicle type information among a plurality of correction amount tables in which the wind direction and the correction amount for correcting the wind speed observation value are associated with each type of the vehicle 110, and the wind direction observation value. In this example, the wind direction composite value is used as the wind direction observation value, and the wind speed composite value is used as the wind speed observation value. That is, the wind speed correction unit 58 calculates a wind speed correction value by correcting the wind speed composite value based on the correction amount table corresponding to the vehicle type information and the wind direction composite value. Specifically, the wind speed correction unit 58 obtains a correction amount corresponding to the wind direction composite value calculated by the wind direction and wind speed synthesis unit 57 from the correction amount table selected by the correction amount table selection unit 56, and calculates the wind speed correction value by multiplying the obtained correction amount by the wind speed composite value calculated by the wind direction and wind speed synthesis unit 57. The wind speed correction value is a value of the wind speed based on the wind direction received by the vehicle 110 passing the installation position of the anemometer 30. In this way, the wind speed correction unit 58 calculates a corrected wind speed value for the vehicle 110 passing through the installation position of the anemometer 30. The wind speed correction unit 58 passes data including the wind direction composite value and the wind speed correction value to the wind direction and speed data transmission unit 59 as wind direction and speed correction data.
[0038] The wind direction and speed data transmission unit 59 transmits the wind direction and speed correction data to the traffic management device 40.
[0039] The processes in the wind direction and wind speed synthesis unit 57, the wind speed correction unit 58, and the wind direction and wind speed data transmission unit 59 are executed every time wind direction and wind speed data is received by the wind direction and wind speed observation data receiving unit 52. As an example, if the wind speed correction device 50 receives passing train information five minutes before the vehicle 110 passes the installation position of the anemometer 30, then every time wind direction and wind speed data is received from the anemometer 30, the wind direction synthesis value and wind speed correction value are calculated and the wind direction and wind speed correction data is transmitted to the traffic management device 40.
[0040] The information transmission device 70 is a device that transmits information transmitted from each device in the train control management system 1 to a destination device. The information transmission device 70 is connected to the correction amount table generating device 20, the anemometer 30, the traffic management device 40, and the wind speed correction device 50 via communication lines. The information transmission device 70 includes an information receiving unit 71 and an information transmitting unit 72. The information receiving unit 71 receives information from any one of the correction amount table generating device 20, the anemometer 30, the traffic management device 40, and the wind speed correction device 50. The information transmitting unit 72 transmits the information received by the information receiving unit 71 to any one of the destinations, that is, the correction amount table generating device 20, the anemometer 30, the traffic management device 40, and the wind speed correction device 50. The information transmission device 70 is provided as necessary.
[0041] Although the above example illustrates a case where there is one anemometer 30, the same processing is performed when there are multiple anemometers. In this case, the passing train information, wind direction and speed data, correction amount table, and vehicle direction information associated with the installation positions of the anemometer 30 are used, and processing is performed for each installation position of the anemometer 30.
[0042] Next, a description will be given of the operation of the train control management system 1 according to the embodiment 1. First, a method for generating the correction amount table will be described below, and then a method for correcting the wind speed will be described.
[0043] Fig. 4 is a flowchart showing an example of a procedure of a correction amount table generating method according to the first embodiment. It is preferable that the process shown in Fig. 4 is performed before the wind speed correction method is performed. The correction amount table generating method shown in Fig. 4 is executed by the correction amount table generating device 20.
[0044] First, the difference determination unit 25 determines whether or not the surrounding area information is held (step S11). This is to determine whether or not the surrounding area information receiving unit 21 has received and is holding the surrounding area information from the surrounding area information providing device 10.
[0045] If the surrounding area information is held (Yes in step S11), the difference determination unit 25 determines whether there is a difference between the held surrounding area information and the surrounding area information used to generate the latest correction amount table (step S12). If there is no difference between the two (No in step S12) or if the surrounding area information is not held in step S11 (No in step S11), the difference determination unit 25 determines whether the vehicle detail information is held (step S13). This is to determine whether the vehicle detail information receiving unit 23 has received and is holding the vehicle detail information from the traffic management device 40. If the vehicle detail information is not held (No in step S13), the process returns to step S11.
[0046] If the vehicle detailed information is held (Yes in step S13), the difference determination unit 25 determines whether there is a difference between the held vehicle detailed information and the vehicle detailed information used to generate the latest correction amount table (step S14). If there is no difference (No in step S14), the process returns to step S11.
[0047] If there is a difference in step S12 or step S14 (Yes in step S12 or step S14), the difference determination unit 25 instructs the correction amount table generation unit 26 to generate a correction amount table. The correction amount table generation unit 26 performs analysis using a simulation based on the latest surrounding area information and the latest vehicle detail information, and generates a correction amount table (step S15). Thereafter, the process returns to step S11. In this way, when the surrounding area information or the vehicle detail information is updated, the correction amount table is generated.
[0048] 5 is a flowchart showing an example of the procedure of the wind speed correction method according to Embodiment 1. The wind speed correction method shown in FIG.
[0049] First, the wind direction and wind speed observation data receiving unit 52 receives wind direction and wind speed observation data, i.e., wind direction observation values and wind speed observation values, from the anemometer 30 (step S31). The wind direction and wind speed synthesis unit 57 determines whether it has traveling speed information of the vehicle 110 that will soon pass the installation position of the anemometer 30 (step S32). In other words, it determines whether the passing train information receiving unit 51 has received passing train information including traveling speed information of the vehicle 110 that will soon pass.
[0050] If the wind direction and speed data transmitter 59 does not have the traveling speed information of the vehicle 110 about to pass (No in step S32), the wind direction and speed data transmitter 59 transmits the wind direction observation value and the wind speed observation value received from the anemometer 30 to the traffic management device 40 (step S33). Then, the process returns to step S31.
[0051] If the traveling speed information of the vehicle 110 about to pass is held (Yes in step S32), the wind direction and wind speed synthesis unit 57 calculates a wind direction synthesis value and a wind speed synthesis value, which are the wind direction and wind speed synthesis data that the vehicle 110 about to pass will receive while traveling, from the traveling speed information, vehicle direction information, wind direction observation value and wind speed observation value of the vehicle 110 about to pass (step S34).
[0052] Next, the correction amount table selection unit 56 judges whether a correction amount table is stored in the correction amount table storage unit 54 (step S35). If a correction amount table is not stored (No in step S35), the wind direction and wind speed data transmission unit 59 transmits the wind direction composite value and the wind speed composite value to the traffic management device 40 (step S37). Then, the process returns to step S31.
[0053] If the correction amount table is stored (Yes in step S35), the wind speed correction unit 58 judges whether it has vehicle type information (step S36). That is, it judges whether the passing train information receiving unit 51 has received passing train information including vehicle type information of the vehicle 110 that is about to pass. If it does not have vehicle type information (No in step S36), the process proceeds to step S37.
[0054] If the vehicle type information is held (Yes in step S36), the correction amount table selection unit 56 determines whether there is a difference from the previous time in the correction amount table or the vehicle type information (step S38). If there is a difference from the previous time in the correction amount table or the vehicle type information (Yes in step S38), the correction amount table selection unit 56 selects a correction amount table corresponding to the vehicle type information from the correction amount table storage unit 54 (step S39).
[0055] Thereafter, or if there is no difference from the previous time in the correction amount table or vehicle type information (No in step S38), the wind speed correction unit 58 acquires the correction amount in the selected correction amount table when the wind direction composite value is the wind direction (step S40), and calculates a wind speed correction value by correcting the wind speed composite value using the acquired correction amount (step S41). After that, the wind direction and wind speed data transmission unit 59 transmits wind direction and wind speed correction data having the wind speed composite value and the wind speed correction value to the traffic management device 40 (step S42). Then, the process returns to step S31.
[0056] 6 to 8 are diagrams for explaining the wind speed correction method in the train control management system according to the first embodiment. As shown in Fig. 6, the anemometer 30 at point X transmits wind direction and speed observation data in which the wind speed observation value is 15.0 m / s and the wind direction observation value is 10° to the wind speed correction device 50 (step S51). In addition, the traffic management device 40, which is the command center, transmits passing train information in which the vehicle type information of a vehicle 110 that will soon pass point X where the anemometer 30 is installed is A and the running speed information is V to the wind speed correction device 50 (step S52).
[0057] The wind speed correction device 50 calculates the combined wind direction and speed data of the wind that the upcoming vehicle 110 will encounter from the traveling speed information V of the upcoming vehicle 110 and the observed wind speed and wind direction values (step S53). That is, the combined wind direction and speed information about the wind that the traveling vehicle 110 will encounter is calculated, and the combined wind direction and wind speed values are obtained. Here, the combined wind speed value of the combined wind direction and speed data is assumed to be 10.0 m / s, and the combined wind direction value is assumed to be 20°.
[0058] 7, the wind speed correction device 50 holds the correction amount tables 261A, 261B, and 261C for the vehicle type information A, B, and C. The wind speed correction device 50 then selects the correction amount table 261A corresponding to the vehicle type information A of the passing train information (step S54). The wind speed correction device 50 also acquires a correction amount 1.07 from the correction amount table 261A of the selected vehicle type information A, which makes the wind direction 20° of the wind direction composite value, and calculates 10.7 m / s by multiplying the acquired correction amount 1.07 by the wind speed composite value 10.0 m / s (step S55).
[0059] Then, as shown in FIG. 8, the wind speed correction device 50 transmits the wind speed correction value of 10.7 m / s and the wind direction composite value of 20° to the traffic management device 40 as wind direction and wind speed correction data (step S56). When the traffic management device 40 receives the wind direction and wind speed correction data, it determines whether the wind direction and wind speed correction data of the vehicle type information A corresponds to a driving restriction (step S57). In this example, a case is shown in which the wind direction and wind speed correction data of the vehicle type information A does not correspond to the condition. If the condition does not correspond, nothing is done in particular. On the other hand, if the wind direction and wind speed correction data of the vehicle type information A corresponds to the condition, the traffic management device 40 transmits a driving restriction that corresponds to the vehicle 110 of the vehicle type information A traveling near the X point. In the example of FIG. 8, a case is shown in which there are two types of driving restriction conditions, slow driving and driving stop. If the wind direction and wind speed correction data meets both conditions, the driving restriction of driving stop is prioritized.
[0060] Next, a wind speed correction method in the conventional technology will be described. FIG. 9 is a diagram for explaining the wind speed correction method in the conventional technology. Here, too, it is assumed that there are three types of vehicle type information, A, B, and C. As shown in FIG. 9, in the conventional technology, the correction amount in the wind direction of each vehicle type information is calculated, and correction amount tables 262A, 262B, and 262C are generated. In Patent Document 1, the correction amount is a conversion coefficient. Then, for each wind direction, the largest correction amount value is selected from the correction amounts of the three types of vehicle type information A, B, and C. This is to prevent the vehicle 110 from overturning or derailing due to wind in all of the vehicle type information A, B, and C. In the example of FIG. 9, when the wind direction is 10°, 0.42 of the vehicle type information C with the largest correction amount is selected, and when the wind direction is 20°, 0.45 of the vehicle type information A with the largest correction amount is selected. In this manner, the overall correction amount table 263 for the anemometer 30 at point X is generated using the correction amounts selected for all wind directions (step S71).
[0061] In Patent Document 1, when two data, a wind direction observation value observed by the anemometer 30 and a wind speed correction value, are transmitted, the amount of data increases, which increases the burden of data quality control, so an evaluated wind speed, which is a wind speed value based on the wind direction observation value, is calculated from the two data, the wind direction observation value and the wind speed observation value, and this evaluated wind speed is transmitted as one data to the traffic management device 40. For this reason, the correction amounts in the correction amount tables 262A, 262B, and 262C are for converting the wind speed observation value of the anemometer 30 into an evaluated wind speed, which is the wind speed received by the vehicle 110 based on the wind direction, and are not for correcting the wind received by the vehicle 110 passing near the installation position of the anemometer 30 due to the difference in position between the anemometer 30 and the track 100 as in the first embodiment. In other words, the correction amount tables 262A, 262B, 262C calculated by the method described in Patent Document 1 do not take into account the discrepancy between the wind received by the anemometer 30 and the wind received by a vehicle 110 passing over the track 100 near the installation position of the anemometer 30.
[0062] 9, the anemometer 30 at point X observes wind direction and speed observation data in which the wind speed observation value is 10.0 m / s and the wind direction observation value is 20° (step S72). The wind speed correction device 50X provided in the anemometer 30 obtains a correction amount of 0.45 corresponding to the wind direction of 20°, which is the wind direction observation value, from the overall correction amount table 263, and calculates 4.5 m / s by multiplying the obtained correction amount by the wind speed observation value as the evaluated wind speed (step S73). Then, the wind speed correction device 50X transmits the wind direction and speed correction data including the evaluated wind speed of 4.5 m / s to the traffic management device 40 (step S74).
[0063] In this way, in Patent Document 1, the traffic management device 40 does not transmit passing train information of the vehicle 110 passing through point X to the wind speed correction device 50X. In other words, the wind speed correction device 50X does not know what type of vehicle 110 will pass through point X where the wind direction and anemometer 30 is installed. For this reason, in order to prevent derailment or overturning due to wind no matter which vehicle 110 passes, an overall correction amount table 263 is used in which the largest correction amount among all vehicle type information is obtained for each wind direction.
[0064] Here, an example is taken of a case where the vehicle 110 of the vehicle type information B passes through the installation position of the anemometer 30. In the above example, the wind speed correction device 50X transmits wind direction and wind speed correction data including an evaluated wind speed of 4.5 m / s to the traffic management device 40. On the other hand, referring to the correction amount table 262B of the vehicle type information B, the correction amount when the wind direction is 20° is 0.39. Therefore, the evaluated wind speed is 3.9 m / s. Thus, in Patent Document 1, the correction is excessive for the vehicle 110 of the vehicle type information B. There is no problem with the wind speed in the example because it does not correspond to the driving restriction, but in the case of a wind speed that corresponds to the driving restriction, for a certain vehicle type information, the evaluated wind speed is such that derailment or overturning does not occur, but the driving restriction is issued for the vehicles 110 of all the vehicle type information. In this case, when a certain vehicle type information passes near the installation position of the anemometer 30, excessive driving restriction is imposed, and the driving efficiency decreases. This is because the operation management device 40 does not notify the wind speed correction device 50X of passing train information, and the wind speed correction device 50X has an overall correction amount table 263 that has been generated to ensure safe operation for all vehicle type information A, B, and C.
[0065] On the other hand, in the first embodiment, the wind speed correction device 50 acquires from the traffic management device 40 passing train information including vehicle type information and running speed information of the vehicle 110 that will soon pass the installation point of the anemometer 30, and calculates the wind speed correction value that the vehicle 110 that will soon pass will receive while traveling, using the correction amount tables 261A, 261B, 261C generated for each vehicle type information. This makes it possible to perform appropriate driving restrictions for each vehicle type information, and to improve the driving efficiency of the entire line.
[0066] In addition, when calculating the correction amount tables 261A, 261B, 261C, the vehicle detail information may include travel speed information of the vehicle 110 indicated in the vehicle type information. In one example, the travel speed information of the vehicle 110 is an average speed calculated from the bus schedule information of the vehicle 110. In another example, the travel speed information of the vehicle 110 is a plurality of predetermined speeds for each vehicle type information. The plurality of predetermined speeds may be a first speed that is a low speed such as when driving slowly, a second speed such as an average speed calculated from the bus schedule information, a third speed between the first speed and the second speed, and the like.
[0067] In this way, when the correction amount tables 261A, 261B, and 261C are generated using vehicle detail information including the traveling speed information of the vehicle 110, the wind speed correction device 50 can omit the vehicle direction information storage unit 55 and the wind direction and wind speed synthesis unit 57. In this case, the wind speed correction unit 58 acquires a correction amount corresponding to the wind direction observation value from the anemometer 30 from the correction amount tables 261A, 261B, and 261C, and calculates a wind speed correction value. The wind direction and wind speed data transmission unit 59 transmits wind direction and wind speed data including the wind direction observation value and the wind speed correction value to the traffic management device 40. In this case, the wind speed correction unit 58 calculates a wind speed correction value by correcting the wind speed observation value based on the correction amount table corresponding to the vehicle type information and the wind direction observation value.
[0068] Embodiment 2 FIG. 10 is a diagram showing a schematic example of the configuration of a train control management system according to the second embodiment. The same components as those in the first embodiment are given the same reference numerals, and their description will be omitted. Only the parts different from the first embodiment will be described. The train control management system 1A according to the second embodiment further includes a wind direction and speed data storage device 80 in addition to the configuration of the first embodiment. The correction amount table generating device 20, the anemometer 30, the traffic management device 40, the wind speed correction device 50, the information transmission device 70, and the wind direction and speed data storage device 80 are connected to each other via a communication line such as a network.
[0069] The wind direction and speed data accumulation device 80 has a wind direction and speed data receiving unit 81, a wind direction and speed data storage unit 82, a learning unit 83, a learned model storage unit 84, an abnormality determination unit 85, and an alarm information transmission unit 86.
[0070] The wind direction and speed data receiving unit 81 receives wind direction and speed correction data transmitted from the wind speed correction device 50. Here, the wind direction and speed data receiving unit 81 receives a wind direction composite value and a wind speed correction value as the wind direction and speed correction data. The wind direction and speed data receiving unit 81 may also receive driving restriction information including driving restriction content data indicating the content of driving restrictions when driving restrictions are issued from the traffic management device 40, the installation position of the anemometer 30 corresponding to the driving restriction, and the date and time when the driving restriction was issued.
[0071] The wind direction and wind speed data storage unit 82 stores the wind direction and wind speed correction data received by the wind direction and wind speed data receiving unit 81. When storing the wind direction and wind speed correction data, the wind direction and wind speed data storage unit 82 desirably also stores the date and time of the wind direction and wind speed correction data. The date and time of the wind direction and wind speed correction data may be the observation date and time of the wind direction observation value and the wind speed observation value by the anemometer 30, the transmission date and time of the wind direction composite value and the wind speed correction value by the anemometer 50, or the reception date and time of the wind direction composite value and the wind speed correction value by the anemometer 81. Furthermore, since a plurality of anemometers 30 are generally provided, the wind direction and wind speed correction data is stored in association with the anemometer 30. Furthermore, when the wind direction and wind speed data storage unit 82 receives driving restriction information from the operation management device 40, it also stores the driving restriction information.
[0072] The learning unit 83 uses learning data including wind direction and wind speed correction data having a wind direction observation value and a wind speed correction value to generate a trained model for inferring the occurrence of an event that may affect the operation of the vehicle 110 from the wind direction and wind speed data. Here, a composite wind direction value is used as the wind direction observation value. In this example, the learning unit 83 performs machine learning using accumulated information including the wind direction and wind speed correction data accumulated in the wind direction and wind speed data storage unit 82, and generates a trained model for inferring the occurrence of an event that is likely to affect the operation of the vehicle 110 from the wind direction and wind speed correction data.
[0073] The learned model storage unit 84 stores the learned model generated by the learning unit 83.
[0074] The abnormality determination unit 85 uses a trained model for inferring the occurrence of an event that may affect the operation of the vehicle 110 from the wind direction and wind speed correction data including the wind direction observation value and the wind speed correction value, and outputs the occurrence of an event that may affect the operation of the vehicle 110 from the wind direction and wind speed correction data. That is, the abnormality determination unit 85 determines the occurrence of an event that may affect the operation of the vehicle 110 from a result obtained by inputting the wind direction and wind speed correction data received by the wind direction and wind speed data receiving unit 81 into the trained model. When the abnormality determination unit 85 determines that an event that may affect the operation of the vehicle 110 will occur, it generates alarm information indicating the occurrence of an event that may affect the operation of the vehicle 110. The content of the alarm information can be determined according to the output obtained by inputting the wind direction and wind speed data into the trained model. When the output of the trained model indicates that the wind direction and wind speed correction data has an unusual tendency, the alarm information does not specifically indicate what kind of event will occur, but simply alerts the user. In addition, if the output of the trained model indicates the possibility of an event occurring based on the wind direction and speed corrected data, the alarm information will include specific details about what type of event will occur.
[0075] The alarm information transmission unit 86 transmits the alarm information generated by the abnormality determination unit 85 to the traffic management device 40.
[0076] In addition, since the process performed by the wind direction and speed data accumulation device 80 takes longer than the process performed by the wind speed correction device 50, it is often difficult to execute the process every time the wind direction and speed correction data is received. For this reason, the presence or absence of an event that may affect the operation of the vehicle 110 is determined at a period longer than the reception period of the wind direction and speed correction data, for example, at a one-minute period.
[0077] In the second embodiment, the wind direction and speed data transmitter 59 of the wind speed correction device 50 transmits the wind direction and speed correction data to the traffic management device 40 and also to the wind direction and speed data storage device 80.
[0078] The traffic management device 40 receives wind direction and wind speed correction data including a wind direction composite value and a wind speed correction value from the wind speed correction device 50, and receives alarm information from the wind direction and wind speed data storage device 80. When the traffic management device 40 receives only the wind direction and wind speed correction data, it determines whether driving restrictions are necessary based on the wind direction and wind speed correction data. When the traffic management device 40 receives only the alarm information, it implements driving restrictions based on the alarm information. When the traffic management device 40 receives both the wind direction and wind speed correction data and the alarm information, it compares driving restrictions based on the wind direction and wind speed correction data with driving restrictions based on the alarm information, selects driving restrictions that allow the vehicle 110 to operate more safely, and implements the selected driving restrictions.
[0079] The wind speed correction method in the wind speed correction device 50 is the same as that explained in Fig. 5 of the first embodiment, and therefore the explanation thereof will be omitted. However, it is different in that in step S42, the wind direction and wind speed data transmission unit 59 of the wind speed correction device 50 transmits the wind direction composite value and the wind speed correction value to the traffic management device 40 and the wind direction and wind speed data accumulation device 80.
[0080] Here, an example will be described of the machine learning process by the learning unit 83 of the wind direction and speed data storage unit 82 and the inference process by the anomaly determination unit 85. Below, an example will be described of (1) a case where only the wind direction composite value and the wind speed correction value are stored in the wind direction and speed data accumulation device 80, and (2) a case where the wind direction composite value, the wind speed correction value, and operation restriction information are stored.
[0081] (1) When only the composite wind direction value and the wind speed correction value are stored <Learning Phase> 11 is a diagram illustrating an example of the configuration of a learning unit of a wind direction and speed data accumulation device. The learning unit 83 includes a data acquisition unit 831a and a model generation unit 832a.
[0082] The data acquiring unit 831a acquires learning data including a wind direction observation value and a wind speed correction value. The data acquiring unit 831a acquires the wind direction composite value and the wind speed correction value as learning data from the wind direction and wind speed data storage unit 82. The data acquiring unit 831a corresponds to a learning data acquiring unit.
[0083] The model generating unit 832b uses the learning data to generate a trained model for inferring whether or not there is a deviation of the wind direction and wind speed corrected data from the normal value based on the wind direction observation value and the wind speed correction value. In this example, the wind direction composite value output from the data acquiring unit 831a is used as the wind direction observation value. That is, a trained model is generated for inferring whether or not there is a deviation of the wind direction and wind speed corrected data from the wind direction composite value and the wind speed correction value of the wind direction and wind speed data storage device 80. Here, the learning data is data in which the wind direction composite value and the wind speed correction value are associated with each other.
[0084] In one example, the trained model is configured as a model for classifying, i.e., clustering, a wind direction composite value and a wind speed correction value in a normal state when no driving restrictions are imposed on the vehicle 110, and a wind direction composite value and a wind speed correction value in an abnormal state when driving restrictions are imposed on the vehicle 110. By using driving restriction content data including the content of driving restrictions associated with the wind direction observation value and the wind speed correction value, it is possible to associate the clustered items with the content of the driving restrictions.
[0085] The learning algorithm used by the model generation unit 832a may be a known algorithm such as supervised learning, unsupervised learning, or reinforcement learning. As an example, a case where the K-means method, which is unsupervised learning, is applied will be described. The K-means method is also called clustering. Unsupervised learning refers to a method of learning features in learning data by providing the learning unit 83 with learning data that does not include a label, which is the result of the learning.
[0086] In one example, the model generating unit 832a learns whether or not there is a deviation from a normal value of the corrected wind direction and speed data by so-called unsupervised learning in accordance with a grouping technique using the K-means method.
[0087] K-means is a non-hierarchical clustering algorithm that uses the mean of the clusters to classify a given number of clusters into k.
[0088] Specifically, the K-means algorithm is processed as follows: First, a cluster is randomly assigned to each data xi. Next, the center Vj of each cluster is calculated based on the assigned data. Next, the distance between each xi and each Vj is calculated, and xi is reassigned to the closest central cluster. Then, if the cluster assignment for all xi remains unchanged through the above process, or if the amount of change falls below a certain threshold set in advance, it is determined that convergence has occurred and the process ends.
[0089] In the second embodiment, the presence or absence of deviation of the wind direction and wind speed corrected data from the normal value is learned by unsupervised learning according to learning data created based on a combination of the wind direction composite value and the wind speed corrected value acquired by the data acquisition unit 831a.
[0090] The model generation unit 832a executes the above-described learning to generate and output a trained model.
[0091] Next, the learning process performed by the learning unit 83 will be described with reference to Fig. 12. Fig. 12 is a flowchart showing an example of a processing procedure of the learning method performed by the learning unit.
[0092] First, the data acquisition unit 831a acquires a wind direction composite value and a wind speed correction value (step S91). Note that, although the wind direction composite value and the wind speed correction value are acquired simultaneously, it is sufficient that the wind direction composite value and the wind speed correction value are input in association with each other, and the data of the wind direction composite value and the wind speed correction value may be acquired at different times.
[0093] Next, the model generation unit 832a learns whether or not there is a deviation in the wind direction and wind speed correction data from the normal value through unsupervised learning in accordance with the learning data created based on the combination of the wind direction composite value and the wind speed correction value acquired by the data acquisition unit 831a, and generates a learned model (step S92).
[0094] After that, the trained model storage unit 84 stores the trained model generated by the model generation unit 832a (step S93). This ends the learning process.
[0095] <Utilization phase> 13 is a diagram illustrating an example of the configuration of an abnormality determination unit of a wind direction and speed data accumulation device. The abnormality determination unit 85 corresponds to an inference device. The abnormality determination unit 85 includes a data acquisition unit 851a and an inference unit 852a.
[0096] The data acquiring unit 851a acquires a wind direction observation value and a wind speed correction value. In this example, the data acquiring unit 851a acquires a wind direction composite value and a wind speed correction value received by the wind direction and wind speed data receiving unit 81. In the learning phase, past data accumulated in the wind direction and wind speed data storage unit 82 is used as the wind direction composite value and the wind speed correction value, but in the utilization phase, data received in real time from the wind speed correction device 50 is used as the wind direction composite value and the wind speed correction value. In other words, the data used in the utilization phase is data that is later in time than the data used in the learning phase.
[0097] The inference unit 852a uses a trained model for inferring the presence or absence of deviation of the wind direction and wind speed correction data including the wind direction observation value and the wind speed correction value from the wind direction observation value and the wind speed correction value from the wind direction observation value and the wind speed correction value, and outputs the presence or absence of deviation of the wind direction and wind speed correction data from the normal value from the wind direction observation value and the wind speed correction value acquired by the data acquisition unit 851a. That is, the inference unit 852a infers the presence or absence of deviation of the wind direction and wind speed correction data obtained by using the trained model stored in the trained model storage unit 84 from the normal value. In this example, the wind direction composite value is used as the wind direction observation value. That is, by inputting the wind direction composite value and the wind speed correction value acquired by the data acquisition unit 851a into this trained model, it is possible to infer which cluster the wind direction composite value and the wind speed correction value belong to, and output the inference result as the presence or absence of deviation of the wind direction and wind speed correction data from the normal value.
[0098] In one example, the inference unit 852a judges whether the wind direction composite value and the wind speed correction value input to the trained model belong to a cluster indicating that the wind direction and wind speed correction data are not deviated from the normal value, i.e., normal, or whether the wind direction and wind speed correction data are deviated from the normal value, i.e., abnormal. If the wind direction and wind speed data belong to a cluster indicating normal, the inference unit 852a infers that the wind direction composite value and the wind speed correction value are not deviated from the normal value. On the other hand, if the wind direction and wind speed data belong to a cluster indicating abnormal, the inference unit 852a infers that the wind direction composite value and the wind speed correction value are deviated from the normal value.
[0099] In the second embodiment, it has been described that the presence or absence of deviation of the wind direction and speed correction data from the normal value is output using a learned model learned by the model generation unit 832a of the wind direction and speed data accumulation device 80. However, it is also possible to obtain a learned model from an external source, such as another wind direction and speed data accumulation device 80, and output the presence or absence of deviation of the wind direction and speed correction data from the normal value based on this learned model.
[0100] In this way, the inference unit 852a determines whether or not there is a deviation from the normal value of the wind direction and wind speed corrected data obtained based on the wind direction composite value and the wind speed corrected value, and if there is a deviation from the normal value of the wind direction and wind speed corrected data, generates alarm information indicating that the wind direction and wind speed corrected data deviates from the normal value. The inference unit 852a outputs the alarm information to the alarm information transmission unit 86. The inference unit 852a may also output the presence or absence of deviation from the normal value of the wind direction and wind speed corrected data to a display or the like.
[0101] Next, an inference process for determining whether or not there is a deviation from a normal value of the corrected wind direction and speed data using the abnormality determination unit 85 will be described with reference to Fig. 14. Fig. 14 is a flowchart showing an example of a processing procedure of an abnormality determination method of the abnormality determination unit.
[0102] First, the data acquisition unit 851a acquires a wind direction composite value and a wind speed correction value from the wind direction and wind speed data receiving unit 81 (step S111).
[0103] Next, the inference unit 852a inputs the wind direction composite value and the wind speed correction value to the trained model stored in the trained model storage unit 84, and obtains whether or not there is a deviation of the wind direction and wind speed correction data from the normal value (step S112). The obtained whether or not there is a deviation of the wind direction and wind speed correction data from the normal value becomes the result of inference.
[0104] Thereafter, the inference unit 852a judges whether the wind direction and wind speed correction data deviates from the normal value (step S113). If the wind direction and wind speed correction data deviates from the normal value (Yes in step S113), the inference unit 852a judges whether the obtained inference result differs from the previous inference result (step S114). If there is a difference from the previous inference result (Yes in step S114), the inference unit 852a generates alarm information indicating that the wind direction and wind speed correction data deviates from the normal value (step S115). The alarm information also includes information for identifying the anemometer 30 that is the subject of the alarm information.
[0105] Then, the alarm information transmission unit 86 transmits the alarm information to the traffic management device 40 (step S116). The traffic management device 40 uses the alarm information indicating that the received wind direction and wind speed correction data deviates from the normal value to implement driving restrictions on the vehicle 110 that is about to pass the corresponding anemometer 30. This makes it possible to prevent the vehicle 110 that is about to pass the installation position of the anemometer 30 from being affected by derailment, overturning, etc. due to wind that is different from normal. Then, the process returns to step S111.
[0106] If there is no difference from the previous inference result in step S114 (No in step S114), or if there is no deviation from the normal value of the wind direction and speed correction data in step S113 (No in step S113), the process returns to step S111. If there is no difference from the previous inference result in step S114, alarm information with the same contents has already been transmitted, so in order to prevent duplicate transmission, no alarm information is transmitted.
[0107] The learning algorithm used by the learning unit 83 may be deep learning, which learns to extract the feature amount itself, or other known methods.
[0108] In addition, when implementing unsupervised learning, other known methods capable of clustering can be used in addition to the non-hierarchical clustering using the K-means method described above. For example, hierarchical clustering such as the shortest distance method can be used.
[0109] In this example, the learning unit 83 and the abnormality determination unit 85 may be connected to the wind direction and speed data accumulation device 80 via a network, and may be separate devices from the wind direction and speed data accumulation device 80. Also, the learning unit 83 and the abnormality determination unit 85 may be built into the wind direction and speed data accumulation device 80 as shown in Fig. 10. Furthermore, the learning unit 83 and the abnormality determination unit 85 may exist on a cloud server.
[0110] The model generating unit 832a may learn whether or not there is a deviation from the normal value of the wind direction and wind speed correction data according to learning data created for multiple wind direction and wind speed data storage devices 80. The model generating unit 832a may acquire learning data from multiple wind direction and wind speed data storage devices 80 used in the same area, or may learn whether or not there is a deviation from the normal value of the wind direction and wind speed correction data by using learning data collected from multiple wind direction and wind speed data storage devices 80 operating independently in different areas. It is also possible to add or remove a wind direction and wind speed data storage device 80 that collects learning data to or from the target midway. Furthermore, the learning unit 83 that has learned whether or not there is a deviation from the normal value of the wind direction and wind speed correction data for a certain wind direction and wind speed data storage device 80 may be applied to another wind direction and wind speed data storage device 80, and the learning unit 83 may re-learn whether or not there is a deviation from the normal value of the wind direction and wind speed correction data for the other wind direction and wind speed data storage device 80 and update the data.
[0111] (2) When the composite wind direction value, the wind speed correction value, and the operation restriction information are stored <Learning Phase> In this case, when the traffic management device 40 implements a driving restriction on the vehicle 110, the traffic management device 40 transmits driving restriction information including driving restriction content data indicating the content of the driving restriction, information indicating the anemometer 30 corresponding to the location subject to the driving restriction, and date and time data to the wind direction and speed data accumulation device 80. That is, the wind direction and speed data receiving unit 81 receives the driving restriction information from the traffic management device 40 and accumulates the driving restriction information in the wind direction and speed data storage unit 82. In one example, the wind direction composite value and the wind speed correction value can be associated with the driving restriction content data using the date and time data.
[0112] 15 is a diagram showing an example of the configuration of a learning unit of the wind direction and speed data accumulation device. The learning unit 83 includes a data acquisition unit 831b and a model generation unit 832b.
[0113] The data acquisition unit 831b acquires learning data including a wind direction observation value, a wind speed correction value, and driving restriction content data including the details of driving restrictions. In this example, the data acquisition unit 831b acquires the wind direction composite value, the wind speed correction value, and the driving restriction content data as learning data from the wind direction and wind speed data storage unit 82. The data acquisition unit 831b corresponds to a learning data acquisition unit.
[0114] The model generating unit 832a uses the learning data to generate a trained model for inferring driving restriction details from the wind direction observation value and the wind speed correction value. In this example, the wind direction composite value output from the data acquiring unit 831b is used as the wind direction observation value. That is, a trained model is generated that infers optimal driving restriction details from the wind direction composite value and wind speed correction value of the wind direction and wind speed data storage device 80 and the driving restriction details data. Here, the learning data is data in which the wind direction composite value, the wind speed correction value, and the driving restriction details data are associated with each other.
[0115] In one example, a set of data of the wind direction composite value and the wind speed correction value going back a specified time from the time when the driving restriction was issued, and the driving restriction content data issued at the above-mentioned time, which is the correct answer data, is used as the learning data. The specified time can be one hour, for example. Note that the driving restriction content information is received by the wind direction and wind speed data storage device 80 when the driving restriction content is issued, so the correct answer data associated with the wind direction composite value and the wind speed correction value at the time when there is no driving restriction content information is assumed to be no driving restriction.
[0116] The learning unit 83 and the abnormality determination unit 85 are used to learn the operation restriction contents of the wind direction and speed data storage device 80. Such learning unit 83 and the abnormality determination unit 85 may be built into the wind direction and speed data storage device 80 as shown in Fig. 10. In another example, the learning unit 83 and the abnormality determination unit 85 may be connected to the wind direction and speed data storage device 80 via a network and may be separate devices from the wind direction and speed data storage device 80. Furthermore, the learning unit 83 and the abnormality determination unit 85 may exist on a cloud server.
[0117] The learning algorithm used by the model generating unit 832b may be a known algorithm such as so-called supervised learning, unsupervised learning, reinforcement learning, etc. As an example, a case where a neural network is applied will be described.
[0118] In one example, the model generation unit 832b learns the details of driving regulations according to a neural network model by so-called supervised learning. Here, supervised learning refers to a method of providing the learning unit 83 with a set of data of input and result as a label, learning the features of the learning data, and inferring the result from the input.
[0119] A neural network is composed of an input layer consisting of multiple neurons, an intermediate layer, which is a hidden layer consisting of multiple neurons, and an output layer consisting of multiple neurons. The intermediate layer may be one layer or two or more layers.
[0120] FIG. 16 is a diagram showing an example of a neural network used by the model generation unit. In the case of a three-layered neural network as shown in FIG. 16, when multiple inputs are input to the input layers X1-X3, these values are multiplied by weights indicated by w11-w16 and input to the intermediate layers Y1-Y2. When the weights w11-w16 are not individually distinguished, they are called weight w1. Furthermore, the results of the intermediate layers Y1-Y2 are further multiplied by weights indicated by w21-w26 and output to the output layers Z1-Z3. When the weights w21-w26 are not individually distinguished, they are called weight w2. The output results of the output layers Z1-Z3 vary depending on the values of the weights w1 and w2.
[0121] In this example, the neural network learns the details of driving restrictions through supervised learning in accordance with learning data created based on a combination of the wind direction composite value and wind speed correction value acquired by the data acquisition unit 831b and the driving restriction details data.
[0122] That is, the neural network learns by inputting the composite wind direction value and the corrected wind speed value to the input layer and adjusting the weights w1 and w2 so that the results output from the output layer approach the driving restriction details data.
[0123] The model generation unit 832b executes the above-described learning to generate and output a trained model.
[0124] Next, the learning process performed by the learning unit 83 will be described with reference to Fig. 17. Fig. 17 is a flowchart showing an example of the processing procedure of the learning method performed by the learning unit.
[0125] First, the data acquisition unit 831b acquires the wind direction composite value, the wind speed correction value, and the driving restriction content data (step S131). Note that the wind direction composite value, the wind speed correction value, and the driving restriction content data are acquired simultaneously. However, as long as the wind direction composite value, the wind speed correction value, and the driving restriction content data are input in association with each other, the wind direction composite value, the wind speed correction value, and the driving restriction content data may be acquired at different times. In one example, the wind direction composite value and the wind speed correction value may be a data group of the wind direction composite value and the wind speed correction value that are obtained a predetermined time before the time when the driving restriction was issued.
[0126] Next, the model generation unit 832b learns the driving restriction details through supervised learning in accordance with learning data created based on a combination of the wind direction composite value and wind speed correction value acquired by the data acquisition unit 831b and the driving restriction details data, and generates a learned model (step S132).
[0127] After that, the trained model storage unit 84 stores the trained model generated by the model generation unit 832b (step S133). This ends the learning process.
[0128] <Utilization phase> 18 is a diagram illustrating an example of the configuration of the abnormality determination unit of the wind direction and speed data accumulation device. The abnormality determination unit 85 includes a data acquisition unit 851b and an inference unit 852b.
[0129] The data acquisition unit 851b acquires a wind direction observation value and a wind speed correction value. In this example, the data acquisition unit 851b acquires a wind direction composite value and a wind speed correction value. Here, the data acquisition unit 851b acquires a wind direction composite value and a wind speed correction value received by the wind direction and wind speed data receiving unit 81, and a wind direction composite value and a wind speed correction value acquired from the wind direction and wind speed data storage unit 82 within a range going back a set time from the time when the wind direction composite value and the wind speed correction value were received. In the learning phase, past data accumulated in the wind direction and wind speed data storage unit 82 is used as the wind direction composite value and the wind speed correction value, but in the utilization phase, data received from the wind speed correction device 50 is used as the wind direction composite value and the wind speed correction value. In other words, the data used in the utilization phase is data that is later in time than the data used in the learning phase.
[0130] The inference unit 852b uses a trained model for inferring driving restriction contents from the wind direction observation value and the wind speed correction value, and outputs driving restriction contents from the wind direction observation value and the wind speed correction value input from the data acquisition unit 851b. That is, the inference unit 852b infers driving restriction contents obtained by using the trained model. In this example, the wind direction composite value is used as the wind direction observation value. That is, by inputting the wind direction composite value and the wind speed correction value acquired by the data acquisition unit 851b to this trained model, it is possible to output driving restriction contents inferred from the wind direction composite value and the wind speed correction value. As an example, the driving restriction contents can be no restriction, slow driving, and driving stop.
[0131] In this example, the driving restriction contents are output using a learned model learned by the model generation unit 832b of the wind direction and wind speed data accumulation device 80. However, it is also possible to obtain a learned model from an external source, such as another wind direction and wind speed data accumulation device 80, and output the driving restriction contents based on this learned model.
[0132] Next, a process for obtaining details of driving restrictions using the abnormality determination unit 85 will be described with reference to Fig. 19. Fig. 19 is a flowchart showing an example of a process procedure of an abnormality determination method of the abnormality determination unit.
[0133] First, the data acquisition unit 851b acquires a wind direction composite value and a wind speed correction value (step S151).
[0134] Next, the inference unit 852b inputs the wind direction composite value and the wind speed correction value to the learned model stored in the learned model storage unit 84, and obtains the details of the driving restriction (step S152). The obtained details of the driving restriction become the result of the inference.
[0135] Thereafter, the inference unit 852b judges whether the driving restriction content is slow driving or driving stop (step S153). If the driving restriction content is slow driving or driving stop (Yes in step S153), the inference unit 852b judges whether the obtained driving restriction content differs from the result of the previous inference (step S154). If there is a difference from the result of the previous inference (Yes in step S154), the inference unit 852b generates alarm information including the obtained driving restriction content (step S155). The alarm information also includes information for identifying the anemometer 30 that is the subject of the alarm information.
[0136] Then, the alarm information transmission unit 86 transmits the alarm information to the traffic management device 40 (step S156). The traffic management device 40 uses the received alarm information including the details of the traffic restriction to implement traffic restrictions on the vehicles 110 that are about to pass the corresponding anemometer 30. This makes it possible to prevent the vehicles 110 that are about to pass the installation position of the anemometer 30 from being affected by derailment, overturning, or the like due to strong winds. Then, the process returns to step S151.
[0137] If there is no difference from the previous inference result in step S154 (No in step S154), or if the driving restriction content in step S153 is not slow driving or driving suspension (No in step S153), the process returns to step S151.
[0138] In this example, the model generating unit 832b uses a learning algorithm that is supervised learning, but the present invention is not limited to this. As for the learning algorithm, it is also possible to use reinforcement learning, unsupervised learning, semi-supervised learning, or the like, in addition to supervised learning.
[0139] The model generating unit 832b may learn the driving restriction contents according to learning data created for multiple wind direction and wind speed data storage devices 80. The model generating unit 832b may acquire learning data from multiple wind direction and wind speed data storage devices 80 used in the same area, or may learn the driving restriction contents using learning data collected from multiple wind direction and wind speed data storage devices 80 operating independently in different areas. It is also possible to add or remove a wind direction and wind speed data storage device 80 that collects learning data from the target midway. Furthermore, the learning unit 83 that has learned the driving restriction contents for a certain wind direction and wind speed data storage device 80 may be applied to another wind direction and wind speed data storage device 80, and the driving restriction contents for the other wind direction and wind speed data storage device 80 may be re-learned and updated.
[0140] In addition, the learning algorithm used in the model generation unit 832b may be deep learning, which learns to extract the features themselves, or machine learning may be performed according to other known methods, such as genetic programming, functional logic programming, and support vector machines.
[0141] Furthermore, in this example, the input data are the wind direction composite value and the wind speed correction value received by the wind direction and wind speed data receiving unit 81, and the wind direction composite value and the wind speed correction value acquired from the wind direction and wind speed data storage unit 82 within a range going back a set time from the time when the wind direction composite value and the wind speed correction value were received, but the distribution tendency of these wind direction composite value and wind speed correction value may be used as the input data. The distribution tendency of the wind direction composite value and the wind speed correction value may be obtained by statistically processing or approximating a data group within a set range.
[0142] In the above description, a trained model was generated to learn the driving restriction content using the wind direction composite value and the wind speed correction value in a period going back a certain time from a certain time and the driving restriction content data at a certain time as the correct answer data. In addition, various learning processes can be performed using the data stored in the wind direction and speed data storage device 80. In one example, the relationship between the tendency of the wind direction composite value and the wind speed correction value in a certain period and the wind speed considered to be a strong wind at a certain time after a certain period has passed, for example, the strong wind occurrence probability, which is the probability that a wind of 20 m / s or more will blow, or the wind speed of the wind blowing at a certain time, may be trained. In this case, the strong wind occurrence probability or the wind speed at the time when the vehicle 110 passes the installation position of the anemometer 30 is notified to the traffic management device 40, and a decision on driving restriction is made based on the inferred value of the strong wind occurrence probability or the wind speed.
[0143] In the second embodiment, the wind direction composite value and the wind speed correction value corrected by the wind speed correction device 50 are stored in the wind direction and wind speed data storage device 80, and the stored wind direction composite value and wind speed correction value are used to generate a trained model that infers the effect of wind on the vehicle 110. The trained model is then input with the wind direction composite value and the wind speed correction value received from the wind speed correction device 50, and the result is transmitted to the traffic management device 40, and the traffic management device 40 determines the driving restriction based on the contents. In one example, the wind direction composite value and the wind speed correction value are detected to be outside of a normal range in which driving restrictions are not issued, and alarm information is transmitted to the traffic management device 40, or alarm information including the contents of the driving restriction at a certain time estimated from the data of the wind direction composite value and the wind speed correction value at a specified time from a certain time is transmitted to the traffic management device 40. This makes it possible to determine the driving restriction of the vehicle 110 using not only the wind speed observed in real time, but also the occurrence of wind that is likely to affect the vehicle 110, which is inferred from past experience.
[0144] 1 and 10 show a case where the correction amount table generating device 20 is provided separately from the wind speed correcting device 50, but the wind speed correcting device 50 may have the function of the correction amount table generating device 20.
[0145] In addition, in FIG. 10, the wind direction and speed data accumulation device 80 is provided separately from the wind speed correction device 50, but the wind speed correction device 50 may have the functions of the wind direction and speed data accumulation device 80.
[0146] 1 and 10, a system including the wind speed correction device 50 and the correction amount table generation device 20 corresponds to a wind speed correction system that can correct the wind speed of the vehicle 110. In other words, the train control management system 1 can be considered to be a system including the anemometer 30, the traffic management device 40, and the wind speed correction system, and the train control management system 1A can be considered to be a system including the anemometer 30, the traffic management device 40, the wind speed correction system, and the wind direction and speed data accumulation device 80.
[0147] 10, the system including the wind speed correction device 50 and the wind direction and speed data storage device 80 corresponds to a wind speed correction system that can correct the wind speed of the vehicle 110. In other words, the train control management system 1A can be considered as a system including the correction amount table generation device 20, the wind direction and speed meter 30, the traffic management device 40, and the wind speed correction system.
[0148] 1 and 10, a system including the wind speed correction device 50 and the traffic management device 40 corresponds to a traffic management system that can manage the traffic of the vehicle 110 using a wind speed correction value obtained by correcting the wind speed of the vehicle 110. In other words, the train control management system 1 can be considered to be a system including the correction amount table generation device 20, the anemometer 30, and a traffic management system, and the train control management system 1A can be considered to be a system including the correction amount table generation device 20, the anemometer 30, the traffic management system, and the wind direction and speed data storage device 80.
[0149] 10, a system including the wind speed correction device 50, the traffic management device 40, and the wind direction and speed data storage device 80 corresponds to a traffic management system that can manage the traffic of the vehicle 110 using a wind speed correction value obtained by correcting the wind speed of the vehicle 110. In other words, the train control management system 1A can be considered to be a system including the correction amount table generation device 20, the wind direction and speed meter 30, and the traffic management system.
[0150] Next, a description will be given of the hardware configurations of the correction amount table generating device 20, the traffic management device 40, the wind speed correction device 50, and the wind direction and speed data storage device 80 used in the train control management systems 1 and 1A. In the wind speed correction device 50 of the first and second embodiments, a computer system functions as the wind speed correction device 50 by executing a program, which is a computer program describing the processing in the wind speed correction device 50, on the computer system.
[0151] Fig. 20 is a block diagram showing an example of the configuration of a computer system that realizes the wind speed correction device according to embodiment 1. As shown in Fig. 20, this computer system includes a control unit 901, an input unit 902, a storage unit 903, a display unit 904, a communication unit 905, and an output unit 906, which are connected via a system bus 907.
[0152] In FIG. 20, the control unit 901 is, in one example, a processor such as a CPU (Central Processing Unit) and executes a program in which the processing in the wind speed correction device 50 of the first and second embodiments is described. The input unit 902 is, in one example, composed of a keyboard, a mouse, and the like, and is used by a user of the computer system to input various information. The storage unit 903 includes various memories such as a RAM (Random Access Memory) and a ROM (Read Only Memory) and a storage device such as a hard disk, and stores the program to be executed by the control unit 901, necessary data obtained in the process of processing, and the like. The storage unit 903 is also used as a temporary storage area for the program. The display unit 904 is composed of a display, a liquid crystal display panel, and the like, and displays various screens to the user of the computer system. In one example, the input unit 902 and the display unit 904 may be composed of a touch panel in which the input unit 902 and the display unit 904 are integrally formed. The communication unit 905 is a receiver and a transmitter that perform communication processing. The output unit 906 is a printer, a speaker, and the like. Note that FIG. 20 is just an example, and the configuration of the computer system is not limited to the example of FIG.
[0153] Here, an example of the operation of the computer system until the program becomes executable will be described. In the computer system having the above-mentioned configuration, for example, a program is installed in the storage unit 903 from a CD-ROM or DVD-ROM set in a CD (Compact Disc)-ROM drive or DVD (Digital Versatile Disc)-ROM drive (not shown). Then, when the program is executed, the program read from the storage unit 903 is stored in the main storage area of the storage unit 903. In this state, the control unit 901 executes the process as the wind speed correction device 50 of the first and second embodiments according to the program stored in the storage unit 903.
[0154] In the above description, a program describing the processing in the wind speed correction device 50 is provided on a CD-ROM or DVD-ROM as a recording medium. However, this is not limiting, and depending on the configuration of the computer system and the capacity of the program to be provided, for example, a program provided over a transmission medium such as the Internet via the communication unit 905 may be used.
[0155] The correction amount table generating device 20, the traffic management device 40 and the wind direction and speed data storage device 80 are also realized by computer systems having similar hardware configurations, so their explanations are omitted.
[0156] The correction amount table selection unit 56, the wind direction and wind speed synthesis unit 57, and the wind speed correction unit 58 of the wind speed correction device 50 shown in Fig. 1 and Fig. 10 are realized by executing a program stored in the storage unit 903 shown in Fig. 20 by the control unit 901 shown in Fig. 20. The correction amount table selection unit 56, the wind direction and wind speed synthesis unit 57, and the wind speed correction unit 58 are realized by the storage unit 903 shown in Fig. 20. The correction amount table storage unit 54 and the vehicle direction information storage unit 55 are realized by the storage unit 903 shown in Fig. 20. The passing train information receiving unit 51, the wind direction and wind speed observation data receiving unit 52, the correction amount table receiving unit 53, and the wind direction and wind speed data transmitting unit 59 are realized by the communication unit 905 shown in Fig. 20.
[0157] The difference determination unit 25 and the correction amount table generation unit 26 of the correction amount table generation device 20 shown in Fig. 1 and Fig. 10 are realized by executing a program stored in the storage unit 903 shown in Fig. 20 by the control unit 901 shown in Fig. 20. The difference determination unit 25 and the correction amount table generation unit 26 are realized by the storage unit 903 shown in Fig. 20. The surrounding area information storage unit 22 and the vehicle detailed information storage unit 24 are realized by the storage unit 903 shown in Fig. 20. The surrounding area information receiving unit 21, the vehicle detailed information receiving unit 23, and the correction amount table transmission unit 27 are realized by the communication unit 905 shown in Fig. 20.
[0158] The learning unit 83 and the abnormality determination unit 85 of the wind direction and wind speed data accumulation device 80 shown in Fig. 10 are realized by executing a program stored in the memory unit 903 shown in Fig. 20 by the control unit 901 shown in Fig. 20. The learning unit 83 and the abnormality determination unit 85 are also realized by the memory unit 903 shown in Fig. 20. The wind direction and wind speed data memory unit 82 and the learned model memory unit 84 are realized by the memory unit 903 shown in Fig. 20. The wind direction and wind speed data receiving unit 81 and the alarm information transmitting unit 86 are realized by the communication unit 905 shown in Fig. 20.
[0159] The configurations shown in the above embodiments are merely examples, and may be combined with other known technologies, or the embodiments may be combined with each other. Also, parts of the configurations may be omitted or modified without departing from the spirit of the invention.
[0160] Various aspects of the present disclosure are summarized below as appendices.
[0161] [Appendix 1] a wind direction and speed observation data receiving unit that receives wind direction and speed observation data including wind direction observation values and wind speed observation values that are wind direction and wind speed values observed by an anemometer installed along a railway track; a passing train information receiving unit that receives passing train information including vehicle type information that is a type of vehicle passing through the track; a wind speed correction unit that calculates a wind speed correction value by correcting the wind speed observation value based on the wind direction observation value and correction amount information corresponding to the vehicle type information among a plurality of correction amount information in which a wind direction and a correction amount for correcting the wind speed observation value are associated with each vehicle type; A wind speed correction device comprising: [Appendix 2] The wind speed correction device described in Appendix 1, characterized in that the correction amount information is information generated for each type of vehicle based on the vehicle type information, vehicle detail information including the weight and shape of the vehicle corresponding to the vehicle type information, and surrounding area information including map information of the surrounding area including the installation position of the anemometer and structure information which is information including the size of structures. [Appendix 3] The wind speed correction device described in Appendix 2, further comprising a correction amount information generation unit that calculates a correction amount, which is a ratio of the wind speed received by the vehicle to the wind speed at the installation position of the anemometer in each wind direction, by simulation based on the vehicle detail information and the surrounding area information, and generates the correction amount information for each type of vehicle. [Appendix 4] The passing train information further includes travel speed information of the vehicle, a wind direction and wind speed synthesis unit that calculates a synthetic wind direction value and a synthetic wind speed value, which are a wind direction and a wind speed based on information on a traveling speed of the vehicle, from the wind direction observation value and the wind speed observation value, The wind speed correction device according to any one of claims 1 to 3, characterized in that the wind speed correction unit calculates the wind speed correction value by correcting the wind speed composite value based on the correction amount information corresponding to the vehicle type information and the wind direction composite value. [Appendix 5] The wind speed correction device according to any one of appendices 1 to 4, further comprising an abnormality determination unit that uses a trained model for inferring the occurrence of an event that may affect the operation of the vehicle from wind direction and wind speed corrected data including the wind direction observation value and the wind speed correction value, and outputs the occurrence of an event that may affect the operation of the vehicle from the wind direction and wind speed corrected data. [Appendix 6] The abnormality determination unit is A data acquisition unit that acquires the wind direction observation value and the corrected wind speed value; an inference unit that outputs, using a trained model for inferring whether or not the wind direction and wind speed corrected data, including the wind direction observation value and the wind speed corrected value, deviates from a normal value from the wind direction observation value and the wind speed corrected value acquired by the data acquisition unit; 6. The wind speed correction device according to claim 5, further comprising: [Appendix 7] The abnormality determination unit is A data acquisition unit that acquires the wind direction observation value and the corrected wind speed value; an inference unit that outputs the details of driving restrictions from the wind direction observation value and the wind speed correction value input from the data acquisition unit, using a trained model for inferring details of driving restrictions from the wind direction observation value and the wind speed correction value; 6. The wind speed correction device according to claim 5, further comprising: [Appendix 8] The wind speed correction device according to any one of appendices 1 to 7, further comprising a learning unit that generates a trained model for inferring the occurrence or nonoccurrence of an event that may affect the operation of the vehicle from the wind direction and wind speed correction data, using learning data including wind direction and wind speed correction data having the wind direction observation value and the wind speed correction value. [Appendix 9] The learning unit is a learning data acquisition unit that acquires learning data including the wind direction observation value and the corrected wind speed value; a model generation unit that generates a trained model for inferring the presence or absence of deviation of the wind direction and wind speed corrected data from a normal value based on the wind direction observation value and the wind speed corrected value, using the learning data; 9. The wind speed correction device according to claim 8, comprising: [Appendix 10] The learning unit is a learning data acquisition unit that acquires learning data including the wind direction observation value, the corrected wind speed value, and driving restriction content data including details of driving restrictions; a model generation unit that generates a trained model for inferring details of driving restrictions from the wind direction observation value and the corrected wind speed value using the learning data; 9. The wind speed correction device according to claim 8, comprising: [Appendix 11] A wind speed correction system including: a wind speed correction device that corrects a wind speed observed by an anemometer installed along a railway track using correction amount information; and a correction amount information generation device that generates the correction amount information, The correction amount information generating device includes: a correction amount information generation unit that calculates, by simulation, a correction amount, which is the ratio of the wind speed received by the vehicle to the wind speed at the installation position of the anemometer in each wind direction, based on vehicle type information, which is the type of vehicle passing through the track, and vehicle detail information including the weight and shape of the vehicle corresponding to the vehicle type information, and generates a plurality of pieces of correction amount information that associate the wind direction with the correction amount for each type of vehicle; a correction amount information transmitting unit that transmits the correction amount information to the wind speed correction device; having The wind speed correction device is a wind direction and speed observation data receiving unit that receives wind direction and speed observation data including wind direction observation values and wind speed observation values that are wind direction values and wind speed values observed by the anemometer; a passing train information receiving unit for receiving passing train information including the vehicle type information; a wind speed correction unit that calculates a wind speed correction value by correcting the wind speed observation value based on the correction amount information corresponding to the vehicle type information among the plurality of correction amount information and the wind direction observation value; A wind speed correction system comprising: [Appendix 12] A wind speed correction system comprising: a wind speed correction device that corrects wind speed observed by an anemometer installed along a railway track; and a wind direction and speed data storage device that stores wind direction information obtained from the wind speed corrected by the wind speed correction device and the wind direction observed by the anemometer, The wind speed correction device is a wind direction and speed observation data receiving unit that receives wind direction and speed observation data including wind direction observation values and wind speed observation values that are wind direction values and wind speed values observed by the anemometer; a passing train information receiving unit that receives passing train information including vehicle type information that is a type of vehicle passing through the track; a wind speed correction unit that calculates a wind speed correction value by correcting the wind speed observation value based on the wind direction observation value and correction amount information corresponding to the vehicle type information among a plurality of correction amount information in which a wind direction and a correction amount for correcting the wind speed observation value are associated with each vehicle type; a wind direction and wind speed data transmission unit that transmits wind direction and wind speed correction data having the wind direction observation value and the wind speed correction value; having The wind direction and speed data storage device includes: a wind direction and wind speed data receiving unit for receiving the wind direction and wind speed correction data; an abnormality determination unit that outputs, from the wind direction and wind speed corrected data, whether or not an event that may affect the operation of the vehicle has occurred, using a trained model for inferring, from the wind direction and wind speed corrected data, whether or not an event that may affect the operation of the vehicle has occurred; A wind speed correction system comprising: [Appendix 13] A traffic management system including: a wind speed correction device that corrects wind speed observed by an anemometer installed along a railway track; and a traffic management device that manages vehicle operations using wind direction information obtained from the wind speed corrected by the wind speed correction device and the wind direction observed by the anemometer, The wind speed correction device is a wind direction and speed observation data receiving unit that receives wind direction and speed observation data including wind direction observation values and wind speed observation values that are wind direction values and wind speed values observed by the anemometer; a passing train information receiving unit that receives passing train information including vehicle type information that is a type of vehicle passing through the track; a wind speed correction unit that calculates a wind speed correction value by correcting the wind speed observation value based on the wind direction observation value and correction amount information corresponding to the vehicle type information among a plurality of correction amount information in which a wind direction and a correction amount for correcting the wind speed observation value are associated with each vehicle type; a wind direction and wind speed data transmission unit that transmits wind direction and wind speed correction data having the wind direction observation value and the wind speed correction value; Equipped with The traffic management system is characterized in that the traffic management device manages the operation of the vehicle based on the wind direction and wind speed correction data. [Appendix 14] A wind speed correction method for a wind direction and speed correction device that corrects wind direction and speed observation data including wind direction observation values and wind speed observation values, which are wind direction and wind speed values observed by an anemometer installed along a railway track, comprising: a wind direction and wind speed observation data receiving step of receiving the wind direction and wind speed observation data; a passing train information receiving step of receiving passing train information including vehicle type information which is a type of vehicle passing through the track; a wind speed correction process for calculating a wind speed correction value by correcting the wind speed observation value based on the wind direction observation value and correction amount information corresponding to the vehicle type information among a plurality of correction amount information in which a wind direction and a correction amount for correcting the wind speed observation value are associated with each vehicle type; A wind speed correction method comprising: [Appendix 15] a wind direction and speed observation data receiving step of receiving wind direction and speed observation data including wind direction observation values and wind speed observation values, which are wind direction values and wind speed values observed by an anemometer installed along a railway track; a passing train information receiving step of receiving passing train information including vehicle type information which is a type of vehicle passing through the track; a wind speed correction step of calculating a wind speed correction value by correcting the wind speed observation value based on the wind direction observation value and correction amount information corresponding to the vehicle type information among a plurality of correction amount information in which a wind direction and a correction amount for correcting the wind speed observation value are associated with each vehicle type; A wind speed correction program that causes a computer to execute the above. [Explanation of symbols]
[0162] 1,1A Train control management system, 10 Surrounding area information providing device, 20 Correction amount table generating device, 21 Surrounding area information receiving unit, 22 Surrounding area information storage unit, 23 Vehicle detailed information receiving unit, 24 Vehicle detailed information storage unit, 25 Difference determination unit, 26 Correction amount table generating unit, 27 Correction amount table transmitting unit, 30 Wind direction and speed meter, 40 Traffic management device, 50 Wind speed correction device, 51 Passing train information receiving unit, 52 Wind direction and speed observation data receiving unit, 53 Correction amount table receiving unit, 54 Correction amount table storage unit, 55 Vehicle direction information storage unit, 56 Correction amount table selecting unit, 57 Wind direction and speed synthesis unit, 58 Wind speed correction unit, 59 Wind direction and speed data transmitting unit, 70 Information transmission device, 71 Information receiving unit, 72 Information transmitting unit, 80 Wind direction and speed data accumulation device, 81 Wind direction and speed data receiving unit, 82 Wind direction and speed data storage unit, 83 learning unit, 84 learned model storage unit, 85 abnormality determination unit, 86 alarm information transmission unit, 100 tracks, 110 vehicles, 831a, 831b, 851a, 851b data acquisition units, 832a, 832b model generation units, 852a, 852b inference units.
Claims
1. a wind direction and speed observation data receiving unit that receives wind direction and speed observation data including wind direction observation values and wind speed observation values that are wind direction and wind speed values observed by an anemometer installed along a railway track; a passing train information receiving unit that receives passing train information including vehicle type information that is a type of vehicle passing through the track; a wind speed correction unit that calculates a wind speed correction value by correcting the wind speed observation value based on the wind direction observation value and correction amount information corresponding to the vehicle type information among a plurality of correction amount information in which a wind direction and a correction amount for correcting the wind speed observation value are associated with each vehicle type; A wind speed correction device comprising:
2. The wind speed correction device according to claim 1, characterized in that the correction amount information is information generated for each type of vehicle based on the vehicle type information, vehicle detail information including the weight and shape of the vehicle corresponding to the vehicle type information, and surrounding area information including map information of the surrounding area including the installation position of the anemometer and structure information which is information including the size of structures.
3. The wind speed correction device according to claim 2, further comprising a correction amount information generation unit that calculates, by simulation, a correction amount, which is a ratio of the wind speed received by the vehicle to the wind speed at the installation position of the anemometer in each wind direction, based on the vehicle detail information and the surrounding area information, and generates the correction amount information for each type of vehicle.
4. The passing train information further includes travel speed information of the vehicle, a wind direction and wind speed synthesis unit that calculates a synthetic wind direction value and a synthetic wind speed value, which are a wind direction and a wind speed based on information on a traveling speed of the vehicle, from the wind direction observation value and the wind speed observation value, 4. The wind speed correction device according to claim 1, wherein the wind speed correction unit calculates the wind speed correction value by correcting the wind speed composite value based on the correction amount information corresponding to the vehicle type information and the wind direction composite value.
5. The wind speed correction device according to any one of claims 1 to 3, further comprising an abnormality judgment unit that outputs, from the wind direction and wind speed corrected data including the wind direction observation value and the wind speed correction value, the presence or absence of the occurrence of an event that may affect the operation of the vehicle, using a trained model for inferring the presence or absence of the occurrence of an event that may affect the operation of the vehicle from the wind direction and wind speed corrected data.
6. The abnormality determination unit A data acquisition unit that acquires the wind direction observation value and the corrected wind speed value; an inference unit that outputs, using a trained model for inferring whether or not the wind direction and wind speed corrected data, including the wind direction observation value and the wind speed corrected value, deviates from a normal value from the wind direction observation value and the wind speed corrected value acquired by the data acquisition unit; 6. The wind speed correction device according to claim 5, further comprising:
7. The abnormality determination unit A data acquisition unit that acquires the wind direction observation value and the corrected wind speed value; an inference unit that outputs the details of driving restrictions from the wind direction observation value and the wind speed correction value input from the data acquisition unit, using a trained model for inferring details of driving restrictions from the wind direction observation value and the wind speed correction value; 6. The wind speed correction device according to claim 5, further comprising:
8. The wind speed correction device according to any one of claims 1 to 3, further comprising a learning unit that generates a trained model for inferring the occurrence or nonoccurrence of an event that may affect the operation of the vehicle from the wind direction and wind speed correction data using learning data including wind direction and wind speed correction data having the wind direction observation value and the wind speed correction value.
9. The learning unit is a learning data acquisition unit that acquires learning data including the wind direction observation value and the corrected wind speed value; a model generation unit that generates a trained model for inferring the presence or absence of deviation of the wind direction and wind speed corrected data from a normal value based on the wind direction observation value and the wind speed corrected value, using the learning data; 9. The wind speed correction device according to claim 8, further comprising:
10. The learning unit is a learning data acquisition unit that acquires learning data including the wind direction observation value, the corrected wind speed value, and driving restriction content data including details of driving restrictions; a model generation unit that generates a trained model for inferring details of driving restrictions from the wind direction observation value and the corrected wind speed value using the learning data; 9. The wind speed correction device according to claim 8, further comprising:
11. A wind speed correction system including: a wind speed correction device that corrects a wind speed observed by an anemometer installed along a railway track using correction amount information; and a correction amount information generation device that generates the correction amount information, The correction amount information generating device includes: a correction amount information generation unit that calculates, by simulation, a correction amount, which is the ratio of the wind speed received by the vehicle to the wind speed at the installation position of the anemometer in each wind direction, based on vehicle type information, which is the type of vehicle passing through the track, and vehicle detail information including the weight and shape of the vehicle corresponding to the vehicle type information, and generates a plurality of pieces of correction amount information that associate the wind direction with the correction amount for each type of vehicle; a correction amount information transmitting unit that transmits the correction amount information to the wind speed correction device; having The wind speed correction device is a wind direction and speed observation data receiving unit that receives wind direction and speed observation data including wind direction observation values and wind speed observation values that are wind direction values and wind speed values observed by the anemometer; a passing train information receiving unit for receiving passing train information including the vehicle type information; a wind speed correction unit that calculates a wind speed correction value by correcting the wind speed observation value based on the correction amount information corresponding to the vehicle type information among the plurality of correction amount information and the wind direction observation value; A wind speed correction system comprising:
12. A wind speed correction system comprising: a wind speed correction device that corrects wind speed observed by an anemometer installed along a railway track; and a wind direction and speed data storage device that stores wind direction information obtained from the wind speed corrected by the wind speed correction device and the wind direction observed by the anemometer, The wind speed correction device is a wind direction and speed observation data receiving unit that receives wind direction and speed observation data including wind direction observation values and wind speed observation values that are wind direction values and wind speed values observed by the anemometer; a passing train information receiving unit that receives passing train information including vehicle type information that is a type of vehicle passing through the track; a wind speed correction unit that calculates a wind speed correction value by correcting the wind speed observation value based on the wind direction observation value and correction amount information corresponding to the vehicle type information among a plurality of correction amount information in which a wind direction and a correction amount for correcting the wind speed observation value are associated with each vehicle type; a wind direction and wind speed data transmission unit that transmits wind direction and wind speed correction data having the wind direction observation value and the wind speed correction value; having The wind direction and speed data storage device includes: a wind direction and wind speed data receiving unit for receiving the wind direction and wind speed correction data; an abnormality determination unit that outputs, from the wind direction and wind speed corrected data, whether or not an event that may affect the operation of the vehicle has occurred, using a trained model for inferring, from the wind direction and wind speed corrected data, whether or not an event that may affect the operation of the vehicle has occurred; A wind speed correction system comprising:
13. A traffic management system including: a wind speed correction device that corrects wind speed observed by an anemometer installed along a railway track; and a traffic management device that manages vehicle operations using wind direction information obtained from the wind speed corrected by the wind speed correction device and the wind direction observed by the anemometer, The wind speed correction device is a wind direction and speed observation data receiving unit that receives wind direction and speed observation data including wind direction observation values and wind speed observation values that are wind direction values and wind speed values observed by the anemometer; a passing train information receiving unit that receives passing train information including vehicle type information that is a type of vehicle passing through the track; a wind speed correction unit that calculates a wind speed correction value by correcting the wind speed observation value based on the wind direction observation value and correction amount information corresponding to the vehicle type information among a plurality of correction amount information in which a wind direction and a correction amount for correcting the wind speed observation value are associated with each vehicle type; a wind direction and wind speed data transmission unit that transmits wind direction and wind speed correction data having the wind direction observation value and the wind speed correction value; Equipped with The traffic management system is characterized in that the traffic management device manages the operation of the vehicle based on the wind direction and wind speed correction data.
14. A wind speed correction method for a wind direction and speed correction device that corrects wind direction and speed observation data including wind direction observation values and wind speed observation values, which are wind direction and wind speed values observed by an anemometer installed along a railway track, comprising: a wind direction and wind speed observation data receiving step of receiving the wind direction and wind speed observation data; a passing train information receiving step of receiving passing train information including vehicle type information which is a type of vehicle passing through the track; a wind speed correction process for calculating a wind speed correction value by correcting the wind speed observation value based on the wind direction observation value and correction amount information corresponding to the vehicle type information among a plurality of correction amount information in which a wind direction and a correction amount for correcting the wind speed observation value are associated with each vehicle type; A wind speed correction method comprising:
15. a wind direction and speed observation data receiving step of receiving wind direction and speed observation data including wind direction observation values and wind speed observation values, which are wind direction values and wind speed values observed by an anemometer installed along a railway track; a passing train information receiving step of receiving passing train information including vehicle type information which is a type of vehicle passing through the track; a wind speed correction step of calculating a wind speed correction value by correcting the wind speed observation value based on the wind direction observation value and correction amount information corresponding to the vehicle type information among a plurality of correction amount information in which a wind direction and a correction amount for correcting the wind speed observation value are associated with each vehicle type; A wind speed correction program that causes a computer to execute the above.
Citation Information
Patent Citations
Vane anemometer, method for regulating operation under strong wind, and evaluating method and program
JP2013159259A
