Failure sign diagnosis apparatus, failure sign diagnosis method, and program

Abstract

A failure sign diagnosis apparatus 100 detects failure signs of driving-related devices 21 mounted on a plurality of moving bodies 20a and 20b. The failure sign diagnosis apparatus 100 comprises: an operation history data calculation unit 120 for calculating operation history data including operating times of the driving-related devices 21 and an output usage condition calculated on the basis of operation information of output-using devices 22a and 22b using outputs of the driving-related devices 21; an operation history correction unit 130 for correcting the operation history data through correction processing for correcting an individual difference in each of the travel-related devices 21 and calculating corrected operation history data; and a failure sign diagnosis unit 150 for diagnosing the failure signs of the driving-related devices 21 by comparing the corrected operation history data with a past corrected operation history data group of the plurality of moving bodies 20a and 20b that satisfies a predetermined condition based on the output usage condition. Thus, the present invention provides a failure sign diagnosis apparatus, a failure sign diagnosis method, and a program that take into account the individual difference in the driving-related devices, perform the correcting processing on the individual difference, and share data of the driving-related devices, thereby shortening a preliminary preparation period required for introducing a failure sign diagnosis.

Description

Fault预兆diagnosis device, fault预兆diagnosis method, and program 【0001】 The present invention relates to a fault预兆diagnosis device, a fault预兆diagnosis method, and a program. In particular, the present invention relates to a fault预兆diagnosis device, a fault预兆diagnosis method, and a program for detecting a fault预兆 of a running-related device mounted on a moving body. 【0002】 For example, from the viewpoint of improving the efficiency and labor-saving of the maintenance work of railway vehicles, the development of a fault预兆diagnosis technology that utilizes the data acquired from running-related devices is required. Among running-related devices, an air compressor that generates compressed air used in an air brake or an air spring has a high need for a fault预兆diagnosis technology that detects and diagnoses the预兆 before a fault occurs because it also affects the operation of other devices such as an air brake when a fault occurs. 【0003】 Patent Document 1 discloses a state monitoring device including an acquisition unit, a totaling unit, and a gradation processing unit. The acquisition unit acquires operation information within a range that meets specified conditions regarding a plurality of devices. The totaling unit calculates a total value for a predetermined time of the operation information for each of the plurality of devices. The gradation processing unit performs gradation processing on the total value for each of the plurality of devices. 【0004】 Japanese Unexamined Patent Application Publication No. 2021-107949 【0005】 For example, in a railway vehicle, a plurality of formations made with the same specifications run on one route. Therefore, the running-related devices mounted are common to each formation, but there is a characteristic that even the same running-related device has a different reference operation time for each formation due to individual differences or assembly errors of each running-related device. In addition, since railway vehicles are obliged to perform periodic disassembly inspections, even the same running-related device may have a fluctuating reference operation time due to assembly errors or the like in the process of reassembly each time an inspection is performed. An object of the present invention is to provide a fault预兆diagnosis device, a fault预兆diagnosis method, and a program that shorten the pre-preparation period for introducing fault预兆diagnosis by considering individual differences of running-related devices, correcting the individual differences, and making the data of running-related devices common. 【0006】To solve the above problems, the present invention provides a fault prediction diagnostic device for detecting signs of failure in driving-related equipment mounted on multiple mobile bodies, comprising: an operation history data calculation unit that calculates operation history data including the operating time of the driving-related equipment and output usage conditions calculated based on the operation information of output-using equipment that uses the output of the driving-related equipment; an operation history correction unit that corrects the operation history data by a correction process that corrects individual differences of each of the driving-related equipment and calculates corrected operation history data; and a fault prediction diagnostic unit that diagnoses signs of failure in the driving-related equipment by comparing the corrected operation history data with a group of corrected operation history data of multiple past mobile bodies that satisfy predetermined conditions based on the output usage conditions. In this case, by considering individual differences of driving-related equipment, correcting individual differences, and standardizing the data of driving equipment, it is possible to provide a fault prediction diagnostic device that shortens the preparation period required for introducing fault prediction diagnostics. 【0007】Here, for example, the moving body is a railway vehicle, the running-related equipment is an air compressor that generates compressed air, the output-using equipment is compressed air-using equipment that uses the compressed air, and the fault prediction diagnostic device diagnoses the air system systems of multiple railway vehicles. In this case, fault prediction diagnostics can be performed on the air system systems of railway vehicles. Furthermore, for example, the system further includes a data input unit that acquires running information from the railway vehicle, including the operating status of the air compressor, the operating status of the compressed air-using equipment, and the position and speed of the railway vehicle. The operation history data calculation unit calculates the operating time of the air compressor from the acquired operating information of the air compressor, and calculates the air consumption conditions during operation of the air compressor from the operating information of the compressed air-using equipment and the running information of the railway vehicle as output usage conditions, and calculates the operating time and the air consumption conditions as operation history data. In this case, it is possible to calculate data that can be compared with normal operation history data. Furthermore, for example, if the compressed air-using equipment includes an air brake system, and the operation history data calculation unit can input information on brake cylinder pressure as the operating state of the air brake system, it uses the integrated value of the brake cylinder pressure that fluctuated while the air compressor was operating as one of the air consumption conditions. In this case, an air consumption condition closer to the actual amount of air used can be used.For example, if the operating history correction unit is correcting the operating time, it uses the standard operating time calculated from the performance of the driving-related equipment, or the operating time of one of the multiple mobile bodies, as the reference time, and for the multiple mobile bodies, it uses the result of statistically processing the difference between each of the operating times to be corrected and the reference time as the correction amount for each of the multiple mobile bodies to correct the operating time. If the target of correction is the output usage conditions, it uses the output usage conditions as the reference output usage conditions, and uses the operating time of one of the multiple mobile bodies as the reference time, and for the multiple mobile bodies, if each of the operating times and the reference time are the same, it uses a value calculated by statistical processing or machine learning so that each of the values ​​included in the output usage conditions and the reference output usage conditions are the same, and uses that value as the correction amount for each of the multiple mobile bodies to correct the output usage conditions. In this case, the operating history can be corrected more accurately. Furthermore, for example, the fault prediction diagnosis unit uses the ratio of the difference between the operating time and the estimated operating time calculated by statistically processing the operating time included in the corrected operating history data group as the degree of abnormality. If the degree of abnormality of any one of the multiple mobile units exceeds a predetermined threshold, the unit determines that there is a sign of abnormality in the driving-related equipment of that mobile unit. In this case, signs of abnormality in the driving-related equipment can be detected with high accuracy. 【0008】Furthermore, the present invention relates to a fault prediction diagnostic method for detecting signs of failure in driving-related equipment mounted on multiple mobile bodies, wherein a processor executes a program recorded in memory to calculate operating history data including the operating time of the driving-related equipment and output usage conditions calculated based on the operation information of output-using equipment that uses the output of the driving-related equipment; the operating history data is corrected by a correction process that corrects individual differences in each of the driving-related equipment to calculate corrected operating history data; and signs of failure in the driving-related equipment are diagnosed by comparing the corrected operating history data with a group of corrected operating history data from multiple past mobile bodies that satisfy predetermined conditions based on the output usage conditions. In this case, by considering individual differences in driving-related equipment, correcting individual differences, and standardizing the data of the driving equipment, it is possible to provide a fault prediction diagnostic method that shortens the preparation period required for introducing fault prediction diagnostics. 【0009】Here, for example, the moving body is a railway vehicle, the running-related equipment is an air compressor that generates compressed air, and the output-using equipment is a compressed air-using equipment that uses the compressed air, and the air system systems of multiple railway vehicles are to be diagnosed. In this case, fault prediction diagnosis of the air system systems of railway vehicles can be performed. Furthermore, for example, running information including the operating state of the air compressor, the operating state of the compressed air-using equipment, and the position and speed of the railway vehicle is further acquired from the railway vehicle, the operating time of the air compressor is calculated from the acquired operating information of the air compressor, and the air consumption conditions during the operation of the air compressor are calculated as the output usage conditions from the operating information of the compressed air-using equipment and the running information of the railway vehicle, and the operating time and the air consumption conditions are calculated as operation history data. In this case, data that can be compared with normal operation history data can be calculated. Furthermore, for example, if the compressed air-using equipment includes an air brake system and information on brake cylinder pressure can be input as the operating state of the air brake system, the integrated value of the brake cylinder pressure that fluctuated while the air compressor was operating is used as one of the air consumption conditions. In this case, an air consumption condition closer to the actual amount of air used can be used. Furthermore, for example, if the compressed air equipment includes an air spring device and information on the air suspension pressure can be input as the operating state of the air spring device, the cumulative value of the air suspension pressure that fluctuated while the air compressor was operating can be used as one of the air consumption conditions. In this case, an air consumption condition closer to the actual amount of air used can be used.For example, if the target of correction is the operating time, the standard operating time calculated from the performance of the driving-related equipment, or the operating time of one of the multiple mobile bodies, is used as the reference time. For the multiple mobile bodies, the operating time is corrected by statistically processing the difference between each of the operating time targets for correction and the reference time, and using the result as the correction amount for each of the multiple mobile bodies. If the target of correction is the output usage conditions, the output usage conditions are used as the reference output usage conditions, and the operating time of one of the multiple mobile bodies is used as the reference time. For the multiple mobile bodies, if each of the operating time targets the reference time, the output usage conditions are corrected by statistical processing or machine learning so that each value included in the output usage conditions and the reference output usage conditions are the same, and using the values ​​calculated as the correction amount for each of the multiple mobile bodies. In this case, the operating history can be corrected more accurately. Also, for example, the ratio of the difference between the operating time and the estimated operating time calculated by statistically processing the operating time included in the corrected operating history data group is defined as the degree of abnormality, and if the degree of abnormality of one of the multiple mobile bodies exceeds a predetermined threshold, it is determined that there is a sign of abnormality in the driving-related equipment of that mobile body. In this case, signs of malfunction in the driving-related equipment can be detected with high accuracy. 【0010】 Furthermore, the present invention is a program for detecting signs of failure in driving-related equipment mounted on multiple mobile bodies, and provides a computer with the following functions: a function to calculate operating history data including the operating time of the driving-related equipment and output usage conditions calculated based on the operation information of output-using equipment that uses the output of the driving-related equipment; a function to correct the operating history data by a correction process that corrects individual differences in each of the driving-related equipment and calculate corrected operating history data; and a function to diagnose signs of failure in the driving-related equipment by comparing the corrected operating history data with a group of corrected operating history data from multiple past mobile bodies that satisfy predetermined conditions based on the output usage conditions. In this case, by considering individual differences in driving-related equipment, correcting those individual differences, and standardizing the data of the driving equipment, the computer can implement a function that shortens the preparation period required for introducing failure prediction diagnosis. 【0011】 According to the present invention, by considering individual differences in driving-related equipment, correcting for these individual differences, and standardizing the data of driving-related equipment, it is possible to provide a fault prediction diagnostic device, a fault prediction diagnostic method, and a program that shorten the preparation period required for introducing fault prediction diagnostics. 【0012】 This diagram shows the overall configuration of a fault prediction and diagnosis system, including the moving object. This diagram shows the overall configuration of the fault prediction and diagnosis system when the moving object is a railway vehicle. This diagram shows an example of operational data input to the fault prediction and diagnosis device. This diagram shows the case where an external storage area is provided. This diagram shows an example of operational history data used within the fault prediction and diagnosis device. This diagram shows the processing flow of the operational history data calculation unit. This diagram shows the processing flow of the operational history correction unit. This diagram shows the processing flow of the operational history storage unit. This diagram shows the processing flow of the fault prediction and diagnosis unit. This diagram shows the processing flow of the same condition data extraction process. This diagram shows the flow for estimating air consumption based on occupancy rate information. This diagram shows the overall configuration of the fault prediction and diagnosis system when the moving object is a railway vehicle. This diagram shows an example of operational history data used within the fault prediction and diagnosis device. This diagram shows the processing flow of the operational history data calculation unit. This diagram shows the processing flow of the operational history correction unit. 【0013】 The embodiments of the present invention will be described in detail below with reference to the attached drawings. The present invention will be described below with reference to Examples 1 and 2. 【0014】In this embodiment, a first embodiment of the mobile body fault prediction and diagnosis device will be described with reference to Figures 1 to 10. Figure 1 is a diagram showing the overall configuration of the fault prediction and diagnosis system including the mobile body. The fault prediction and diagnosis system consists of mobile bodies 20a and 20b, a fault prediction and diagnosis device 100, and a diagnosis result notification function 500. The fault prediction and diagnosis device 100 consists of a data input unit 110, an operation history data calculation unit 120, an operation history correction unit 130, an operation history storage unit 140, and a fault prediction and diagnosis unit 150. The mobile body 20a consists of driving-related equipment 21, output-using equipment 22a and 22b, a driving information acquisition unit 23, and a data output unit 24. Although not shown in the diagram, the mobile body 20b has a similar configuration to the mobile body 20a. 【0015】 In this embodiment, the configuration of the mobile bodies 20a, 20b, running-related equipment 21, output-using equipment 22a, 22b, and running information acquisition unit 23 shown in Figure 1 is set to the configuration shown in Figure 2, and the air system systems of multiple railway vehicles are used as the target for diagnosis. Specifically, the mobile bodies 20a and 20b are railway vehicles 200a and 200b. The running-related equipment 21 is an air compressor 210 that generates compressed air used for train running control, such as an air brake system 220 and an air spring system 230. The running-related equipment 21 may also be an air tank. The output-using equipment 22a and 22b are compressed air-using equipment that commonly uses compressed air in general railway vehicles, and more specifically, an air brake system 220 and an air spring system 230. The air system refers to a system including the air compressor 210 or air tank and compressed air-using equipment. The running information acquisition unit 23 is a vehicle information control device 240, which is commonly used in railway vehicles for information aggregation and control of various on-board devices. This enables fault prediction diagnosis of the air system of railway vehicles 200a and 200b. The output devices 22a and 22b in this embodiment are not limited to the devices described above; if there are other devices that use compressed air generated by the air compressor 210, those devices may also be used as output devices. This is also true in other embodiments. Details of the input / output and operation of each component will be described later. 【0016】Example 1 describes an example of a fault prediction diagnostic device in which, for multiple mobile bodies, the operating time (operating time) calculated from the operating state of the driving-related equipment of each mobile body, and the output usage conditions calculated from the operating state of the output-related equipment and the driving state of the multiple mobile bodies (hereinafter, operating time and output usage conditions are collectively referred to as operating history data), are used to perform a correction process on the operating time of each driving-related equipment 21 using a predetermined correction amount to eliminate individual differences, and are treated as common corrected operating history data. The operating time in the newly acquired corrected operating history data is then compared with the estimated operating time calculated using the operating time of past corrected operating history data sets under the same output usage conditions at that time, and the degree of abnormality is calculated, thereby shortening the period until the diagnosis of fault prediction signs of driving-related equipment can be started. 【0017】 According to Example 1, the fault prediction diagnostic device 100 has the following functions, which enable it to shorten the time until the start of fault prediction diagnosis for the driving-related equipment: A function to calculate a standard operating time (standard operating time) from the performance specifications of the driving-related equipment, and to calculate a correction amount for each of the multiple moving units from the difference between the standard operating time and the operating time included in the operating history data of each of the multiple moving units. A function to select one of the multiple moving units and calculate a correction amount from the difference between the standard operating time calculated based on the operating time of the selected unit over a certain period and the operating time of other units over the same period. A function to calculate corrected operating history data by applying the correction amount to the operating time or output usage conditions of multiple units to eliminate the differences between units, and to share the corrected operating history data among the units. A function to refer to the corrected operating history data of multiple moving units from the past for the output usage conditions included in the newly acquired corrected operating history data, and to extract a group of data where the output usage conditions are the same as a group of corrected operating history data. A function to use the average value of the operating time in the group of corrected operating history data as the estimated operating time. This function uses the ratio of the difference between the estimated operating time and the operating time of the corrected operating history data as the anomaly score. It also determines that a failure is likely if the daily average of the anomaly score, when moved over a period of days sufficient to account for operational differences, is greater than the maximum value under normal conditions. 【0018】 Furthermore, according to Embodiment 1, the fault prediction diagnostic device 100 has the following functions, enabling it to identify the mobile unit in which a fault prediction has occurred among multiple mobile units: A function to use a common detection threshold for each running-related equipment 21 by eliminating individual differences through correction processing. A function to identify which train set number's corrected operation history data detected the fault prediction by including a train set number in the corrected operation history data. 【0019】 (Block Configuration) Figure 2 shows the overall configuration of the fault prediction and diagnosis system when the moving objects are railway vehicles 200a and 200b. The railway vehicles 200a and 200b are equipped with an air compressor 210, an air brake device 220, an air spring device 230, a vehicle information control device 240, and a data output unit 250. The fault prediction and diagnosis device 100 is equipped with a data input unit 110, an operation history data calculation unit 120, an operation history correction unit 130, an operation history storage unit 140, and a fault prediction and diagnosis unit 150. The fault prediction and diagnosis device 100 is also connected to a diagnosis result notification function 500. 【0020】The inputs and outputs of the railway vehicle 200a are shown below. The air compressor 210 outputs operation information to the data output unit 250 indicating whether it is operating or stopped. The air brake system 220 outputs BC pressure (brake cylinder pressure) to the data output unit 250. The air spring system 230 uses AS pressure (air suspension pressure) to control vehicle height and vehicle body vibration, but in this embodiment, it is assumed that AS pressure information cannot be obtained from the air spring system 230. In Figure 2, "×" means that AS pressure cannot be obtained from the air spring system 230. As shown, the vehicle information control device 240 outputs the train set number of its own vehicle and the position and speed information of the railway vehicle 200a, calculated based on the wheel diameter and rotation speed of the wheels obtained from the speed generator of the railway vehicle 200a (not shown), as running information to the data output unit 250. The data output unit 250 outputs operational data (see Figure 3, described later) consisting of time-series information on the operation of the air compressor 210, BC pressure, train number, position, and speed to the data input unit 110 of the fault prediction diagnostic device 100. The input and output of the railway vehicle 200b are the same. Furthermore, the method of outputting from the data output unit 250 to the data input unit 110 can be as follows: if the railway vehicle 200a or railway vehicle 200b can be directly connected to the fault prediction diagnostic device 100, output may be performed using known wired communication means such as Ethernet communication; if direct connection to the fault prediction diagnostic device 100 is not possible, output may be performed using a public communication network such as LTE or 5G to an operational data storage area 311 consisting of memory installed in an external storage area 300 built on a server or cloud environment, or a hard disk or SSD (Solid State Drive), as shown in Figure 4, described later. 【0021】The input and output of the fault prediction and diagnosis device 100 are shown below. The data input unit 110 receives operational data from the data output units 250 of the railway vehicles 200a and 200b, respectively, and outputs the operational data to the operational history data calculation unit 120. As described above, the data input method for the data input unit 110 is as follows: if the fault prediction and diagnosis device 100 can be directly connected to the railway vehicle 200, the data may be input by known wired communication; if a direct connection is not possible, the operational data stored in the operational data storage area 311 installed in the external storage area 300 may be input by reading it via wired or wireless communication, as shown in Figure 4 which will be described later. 【0022】The operation history data calculation unit 120 receives operation data for each train set from the data input unit 110. From the operation information of the air compressor 210 included in the operation data, it calculates the time the equipment was in operation (operating time). Furthermore, the operation history data calculation unit 120 calculates the conditions under which the output (compressed air) generated by the air compressor 210 was consumed (air consumption conditions), including the cumulative value of the fluctuation amount of BC pressure while the air compressor 210 was in operation (total BC pressure fluctuation amount), the train position (start position) and train speed (start speed) when the air compressor 210 started operating, and the train position (end position) and train speed (end position) when it stopped operating. In this case, the air consumption conditions are an example of output usage conditions. In this way, it is possible to calculate data that can be compared with normal operation history data. The air consumption conditions are calculated from the operation information of the air brake system 220, which is a compressed air-using device (information indicating whether it is operating or stopped) and the running information of the railway vehicle 200a (information on the position and speed of the railway vehicle 200a). The operation history data calculation unit 120 then outputs the operation history data (see Figure 5, described later) by combining the operation time, air consumption conditions, and the train set number included in the operation data to the operation history correction unit 130 and the operation history storage unit 140. Details of the operation of the operation history data calculation unit 120 will be described later. The reason for including the start and end positions of operation in addition to the total fluctuation amount of BC pressure as air consumption conditions is that the conditions for air consumption by the air spring device 230 are either changes in sprung weight (number of passengers), vibrations during running, or the tilt of the train, or both. If limited to during running, the change in sprung weight (number of passengers) is small and is thought to depend on the terrain traveled while the air compressor was operating. Therefore, the fault prediction and diagnosis unit 150, described later, is limited to data where the start speed and end speed are greater than 0. The method for considering data where the start speed and end speed are 0 will be described in a modified example. 【0023】The operation history correction unit 130 receives operation history data from the operation history data calculation unit 120 and identifies whether the data belongs to railway vehicle 200a or railway vehicle 200b based on the train set number included in the operation history data. Depending on the identification result, the unit performs a correction process on each value of the operating time or air consumption condition included in the operation history data using the correction amount Ca for railway vehicle 200a and the correction amount Cb for railway vehicle 200b, which have been defined in advance and stored in the operation history storage unit 140, and outputs the corrected operation history data, which includes the corrected operating time or corrected air consumption condition, to the operation history storage unit 140 and the fault prediction diagnosis unit 150. In this embodiment, the case of correcting the operating time will be explained as an example, and the case of correcting the air consumption condition will be explained in Embodiment 2. Details of the operation of the operation history correction unit 130 will be described later. 【0024】 The operation history storage unit 140 is composed of a storage area such as memory, a hard disk, or an SSD, and stores the operation history data input from the operation history data calculation unit 120 and the corrected operation history data input from the operation history correction unit 130 within its storage area. In addition, it receives air consumption conditions from the fault prediction diagnosis unit 150, extracts all corrected operation history data that meet predetermined conditions based on the air consumption conditions, and outputs the corrected operation history data group to the fault prediction diagnosis unit 150. The corrected operation history data that meets predetermined conditions will be described in detail in Figure 8, but for example, it is corrected operation history data with similar air consumption conditions. The operation details of the operation history storage unit 140 will be described later. Here, the operation history storage unit 140 is not limited to the configuration shown in Figure 1, and as shown in Figure 4 later, if the fault prediction diagnosis device 100 can connect to an external storage area 300 built on an external server or cloud environment via wired or wireless communication, the data may be stored in the operation history storage unit 140 installed in that storage area. 【0025】The fault prediction and diagnosis unit 150 first uses a condition data extraction process to narrow down the data to those where the start speed and end speed are greater than 0, outputs the air consumption conditions of the corrected operation history data obtained from the operation history correction unit 130 to the operation history storage unit 140, and inputs a group of corrected operation history data that satisfy predetermined conditions based on the air consumption conditions from the operation history storage unit 140. Next, it calculates the estimated operation time using the corrected operation time of the input group of corrected operation history data through an estimated operation time calculation process. Next, it compares the corrected operation time of the corrected operation history data with the estimated operation time to calculate the degree of abnormality. Finally, it diagnoses fault signs using the calculated degree of abnormality and outputs the diagnosis result to the diagnosis result notification function 500. Details of the operation of the fault prediction and diagnosis unit 150 and each process will be described later. 【0026】 The diagnostic result notification function 500 receives diagnostic results from the fault prediction diagnostic unit 150 and notifies the driver and the train depot of the detection of a fault precursor in the air compressor 210 through screen display and audible alarm. Details of the method of notifying the diagnostic results in the diagnostic result notification function 500 will be described later. 【0027】 Figure 3 shows an example of operational data input to the fault prediction and diagnosis device 100. The operational data includes, for example, the date, time, train set number, train speed, train position, operation information (ON / OFF) of the air compressor 210, and information regarding BC pressure. 【0028】The date and time indicate the date and time the operational data was created. The train set number indicates the train set number of the railway vehicle 200a. The train speed indicates the speed of the railway vehicle 200a. The train position indicates the position of the railway vehicle 200a. The air compressor operation information indicates that "ON" means the air compressor 210 is operating and "OFF" means it is stopped. BC pressure indicates the brake cylinder pressure value [Pa]. In this case, it can also be said that the operational data is acquired by the data input unit 110 from the railway vehicle 200a and includes the operating status of the air compressor 219 (in this case, the operating information of the air compressor 210 (ON / OFF)), the operating status of the air brake device 220 which is a compressed air-using device (in this case, BC pressure), and running information including the position and speed of the railway vehicle 200a (in this case, train speed, train position). 【0029】 Figure 4 shows a case where an external storage area 300 is provided. The external storage area 300 is a storage device provided outside the railway vehicles 200a, 200b and the fault prediction and diagnosis device 100, and can be, for example, a server or data center located on the cloud. In this case, the operational data is temporarily stored from the data output unit 250 of the railway vehicles 200a, 200b to the operational data storage area 310 of the external storage area 300. Then, at a predetermined timing, the external storage area 300 outputs the data from the operational data storage area 310 to the data input unit 110 of the fault prediction and diagnosis device 100. In this case, the operational history storage unit 140 is provided in the external storage area 300, rather than inside the fault prediction and diagnosis device 100. Data transmission between the external storage area 300 and the railway vehicles 200a, 200b and the fault prediction and diagnosis device 100 may use a dedicated line, but public communication networks such as LTE or 5G can also be used. 【0030】 Figure 5 shows an example of operation history data used in the fault prediction and diagnosis device 100. The operation history data includes, for example, information on the date, train number, operating time, starting speed, ending speed, starting position, ending position, and total BC pressure fluctuation. 【0031】The date indicates the date the operational history data was created. The train set number indicates the train set number of railway vehicle 200a. The operating time indicates the time [seconds] that the air compressor 210 was operating. The starting speed indicates the speed of railway vehicle 200a when the air compressor 210 started operating. The ending speed indicates the speed of railway vehicle 200a when the air compressor 210 stopped operating. The starting speed and ending speed are calculated from the date, time, and train speed of the operational data in Figure 3 described above. 【0032】 The starting position indicates the position of the railway vehicle 200a when the air compressor 210 starts operating. The ending position indicates the position of the railway vehicle 200a when the air compressor 210 stops operating. The starting and ending positions are calculated from the date, time, and train position of the operating data in Figure 3 described above. The total BC pressure fluctuation indicates the total value of the fluctuation in brake cylinder pressure during the above operating time. In this case, the operating history data includes the operating time and air consumption conditions. The air consumption conditions are the cumulative value of the BC pressure fluctuation while the air compressor 210 is operating (total BC pressure fluctuation), the train position (starting position) and train speed (starting speed) when the air compressor 210 starts operating, and the train position (end of operation position) and train speed (end of operation position) when it stops operating. By using the cumulative value of the BC pressure fluctuation (total BC pressure fluctuation), it is possible to use an air consumption condition that is close to the actual amount of air used. 【0033】(Operation of the Operation History Data Calculation Unit 120) Figure 6 shows the processing flow of the Operation History Data Calculation Unit 120. The Operation History Data Calculation Unit 120 is activated each time it acquires operation data for the railway vehicles 200a and 200b from the data input unit 110. Step 1201 acquires operation data from the data input unit 110 and proceeds to step 1202. Step 1202 determines whether the operation information of the air compressor in the operation data has changed from OFF to ON. If Yes, proceeds to step 1203; if No, proceeds to step 1204. Step 1203 sets the time in the operation data as the start time, the train speed at that time as the start speed, and the train position at that time as the start position, and proceeds to step 1204. 【0034】 Step 1204 determines whether the operation information of the air compressor in the operation data has changed from ON to OFF. If Yes, proceed to step 1205; otherwise, proceed to step 1209. Step 1205 sets the time at that time in the operation data as the end time of operation, the train speed at that time as the end speed of operation, and the train position at that time as the end position of operation, and proceeds to step 1206. Step 1206 calculates the operating time using the following formula and proceeds to step 1207: Operating time = End time of operation - Start time of operation. Step 1207 converts the train number, operating time, start speed of operation, end speed of operation, start position of operation, end position of operation, and total BC pressure fluctuation into the format of operation data (see Figure 5), and proceeds to step 1208. Step 1208 outputs the operation history data to the operation history correction unit 130 and the operation history storage unit 140, and terminates the process. 【0035】Step 1209 determines whether the air compressor operation information is ON and whether the current BC pressure is greater than the BC pressure from one cycle ago. If Yes, proceed to step 1210; otherwise, terminate the process. Step 1210 updates the total BC pressure fluctuation using the following formula and terminates the process: Total BC pressure fluctuation = Total BC pressure fluctuation from one cycle ago + (Current BC pressure - BC pressure from one cycle ago). Note that this embodiment is an example where there is only one BC pressure data in the vehicle. If there are multiple BC pressure data in the vehicle, steps 1209 and 1210 are repeated the number of times corresponding to the number of data points, and the total BC pressure fluctuation is the sum of those values. This is the same in other embodiments. 【0036】 (Operation of the Operation History Correction Unit 130) Figure 7 shows the processing flow of the Operation History Correction Unit 130. Step 1301 is to obtain operation history data from the operation history data calculation unit 120 and proceed to step 1302. Step 1302 is to read the train set number included in the operation history data, obtain the correction amount for the same train set from the operation history storage unit 140 and proceed to step 1303. Details of the correction amount will be described later. Step 1303 is to calculate the corrected operation time by multiplying the operation time included in the operation history data with the obtained correction amount and proceed to step 1304. In this embodiment, the case of multiplication has been described, but depending on the characteristics of the object to be corrected, the correction amounts may be added or subtracted. This is the same in other embodiments as well. Step 1304 is to define the operation history data with the operation time updated to the corrected operation time as the corrected operation history data and proceed to step 1305. Step 1305 outputs the corrected operation history data to the operation history storage unit 140 and the fault prediction diagnosis unit 150, and then terminates the process. 【0037】(Method for calculating the correction amount) The correction amount is set to a value that eliminates individual differences between the equipment installed in each train set. For example, in the case of the air compressor 210, the specified air discharge amount is fixed, but the actual air discharge amount differs slightly from one unit to another. Therefore, even when compressing the same amount of air, the operating time will differ. Consequently, even with the same air consumption, a unit with a high air discharge amount can compress the air to the specified value faster, while a unit with a low air discharge amount will take longer to compress the air to the specified value. To eliminate these individual differences, for example, the operating time calculated from the specified air discharge amount and the capacity of the air tank may be set as the base time, and the ratio of the difference between this and the actual operating time obtained from inspections such as when the equipment for each train set is shipped may be used as the correction amount. Alternatively, one vehicle may be selected from multiple train sets, and the result of statistical processing of the operating time of the selected train set over a certain period (e.g., the average value) may be used as the base time, and the ratio of the difference between this and the result of statistical processing of the operating time of the other train sets over the same period may be used as the correction amount. 【0038】 In other words, when the operating history correction unit 130 is correcting operating time, it uses either the standard operating time calculated from the performance of the running-related equipment (in this case, the capacity of the air tank) or the operating time of one of the multiple moving objects (in this case, the railway vehicles 200a and 200b) as the reference time, and for the multiple moving objects, it uses the result of statistically processing the difference between the operating time of each object being corrected and the reference time as the correction amount for each of the multiple moving objects to correct the operating time. This allows for more accurate correction of the operating history. 【0039】 (Operation of the Operation History Storage Unit 140) Figure 8 shows the processing flow of the Operation History Storage Unit 140. The operation of the Operation History Storage Unit 140 is triggered when data is input from each unit in step 1401. Step 1401 determines which of the following three pieces of data has been input: 1. Organization number from the Operation History Correction Unit 130 2. Corrected operation history data from the Operation History Correction Unit 130 3. Air consumption conditions from the Fault Prediction Diagnosis Unit 150 【0040】1. If you enter , proceed to step 1402. 2. If you enter , proceed to step 1403. 3. If you enter , proceed to step 1404. If you do not enter anything, proceed to No and terminate the process. 【0041】 Step 1402 extracts the correction amount for the train set number that matches the acquired train set number from the storage area, outputs it to the operation history correction unit 130, and terminates the process. Step 1403 saves the acquired operation history data to the storage area and terminates the process. Steps 1404 to 1405 are loop processes that search for all operation history data stored in the storage area one by one. Step 1404 determines whether the operation history data being referenced satisfies all of the following conditions 1 to 3. If the answer is Yes, proceed to step 1405; otherwise, proceed to the next step in the loop process. Conditions 1 to 3 are examples of predetermined conditions based on output usage conditions (in this case, air consumption conditions). 【0042】 Condition 1: Is the start position of the operation history data being referenced within the error of the position threshold from the start position of the air consumption condition? This can be determined by the condition |start position of air consumption condition - start position| < position threshold. Condition 2: Is the end position of the operation history data being referenced within the error of the position threshold from the end position of the air consumption condition? This can be determined by the condition |end position of air consumption condition - end position| < position threshold. Condition 3: Is the total BC pressure fluctuation of the operation history data being referenced within the error of the pressure threshold from the total BC pressure fluctuation of the air consumption condition? This can be determined by the condition |total BC pressure fluctuation of air consumption condition - total BC pressure fluctuation| < pressure threshold. Details of the position threshold and pressure threshold will be described later. 【0043】Step 1405 stores the corrected operation history data being referred to in packets for data transmission and proceeds to the next step of the loop process. When the search has been completed for all the corrected operation history data, the loop process ends and the process proceeds to Step 1406. Step 1406 outputs all the corrected operation history data stored in the packets for data transmission to the fault prediction diagnosis unit 150 and ends the process. 【0044】 (Position threshold value) The position threshold value is a threshold value for determining whether the mobile body is in the same position when the air compressor 210 starts and ends its operation. Therefore, in order to indicate that at least a part of the mobile body is in the same position, it is set to a value equal to or less than the size of the mobile body. For example, as in this embodiment, when the mobile body is a railway vehicle, any value equal to or less than the length of the railway vehicle is set. 【0045】 (Pressure threshold value) The pressure threshold value is a threshold value for determining that the total change amount of the BC pressure is within the range of fluctuations of the value caused by the sensor error at the time of BC pressure acquisition. Therefore, a value obtained by doubling the value obtained by multiplying the value of the sensor error of the BC pressure by the number of acquisitions of the BC pressure is used. The reason for doubling is that when adding up the differences in pressure values, when taking the difference between the data with the error on the side where the pressure value increases and the data with the error on the side where the pressure value decreases, an error up to twice the sensor error may occur. 【0046】(Operation of the Fault Prediction and Diagnosis Unit 150) Figure 9 shows the processing flow of the Fault Prediction and Diagnosis Unit 150. Step 1501 obtains corrected operation history data from the operation history correction unit 130 and proceeds to step 1502. Step 1502 performs the same condition data extraction process and proceeds to step 1503. Details of the same condition data extraction process will be described later. Step 1503 performs the estimated operating time calculation process and proceeds to step 1504. Details of the estimated operating time calculation process will be described later. Step 1504 determines whether the estimated operating time is an error value. If Yes, proceeds to step 1505; if No, the process ends. Step 1505 calculates the degree of abnormality from the corrected operating time of the corrected operation history data and the estimated operating time using the following formula and proceeds to step 1506. Degree of abnormality = |Operating time - Estimated operating time| ÷ Estimated operating time Step 1506 performs the fault prediction and diagnosis process and proceeds to step 1507. Details of the fault prediction and diagnosis process will be described later. Step 1507 determines whether the fault prediction diagnosis process has outputted a fault prediction. If the answer is Yes, the process proceeds to step 1508; otherwise, the process ends. Step 1508 outputs the train set number included in the corrected operation history data being referenced, along with the diagnosis result, to the diagnosis result notification function 500, and then the process ends. 【0047】 (Data extraction process under the same conditions) Figure 10 shows the processing flow of the data extraction process under the same conditions. Step 1601 acquires corrected operation history data from the operation history correction unit 130 and proceeds to step 1602. Step 1602 determines whether both the start speed and end speed of the acquired corrected operation history data are greater than 0. This can be determined by whether the start speed > 0 and the end speed > 0. If Yes, proceed to step 1603; otherwise, terminate the process. Step 1603 extracts the start position, end position, and total BC pressure fluctuation amount from the acquired operation history data, sets them as air consumption conditions, and proceeds to step 1604. Step 1604 outputs the air consumption conditions to the operation history storage unit 140 and proceeds to step 1605. 【0048】Step 1605 determines whether a corrected operation history data group is acquired from the operation history storage unit 140 based on the air consumption condition output in Step 1604. If Yes, the process proceeds to Step 1606; if No, the process ends. Step 1606 reads the corrected operation hours of all the acquired corrected operation history data, outputs the read corrected operation hours to the estimated operation hour calculation process, and ends the process. In this embodiment, since the air consumption amount in the air spring device 230 is replaced by the train position, the process of Step 1602 is included because it is necessary to limit the data to that during running. However, as shown in Embodiment 2, if the air consumption amount of the air spring device 230 can be estimated from the AS pressure or the like, Step 1602 becomes unnecessary. 【0049】 (Estimated operation hour calculation process) The estimated operation hour calculation process is a process of calculating the estimated operation hour by performing known statistical processing on all the corrected operation hours acquired in the same condition data extraction process. As methods of statistical processing, a method of calculating the average value, a method of calculating the median value, a method of calculating the mode value, etc. can be considered. In this embodiment, the average value calculated from the number of data of the acquired corrected operation hours and the total value is used, but using the median value or the mode value does not prevent the present embodiment. If the corrected operation hour cannot be acquired from the same condition data extraction process, an error value is output. 【0050】(Fault Prediction Diagnosis Processing) The fault prediction diagnosis processing is a process that diagnoses whether or not there are signs of a fault in the air compressor 210 by applying known statistical processing to the abnormality level calculated in step 1504 of the fault prediction diagnosis unit 150. Specifically, one possible method is to obtain the maximum value of the abnormality level when the equipment is functioning normally from past data, and use that value as a threshold. If the newly obtained abnormality level shows a value greater than or equal to the threshold multiple times, it is determined that there are signs of a fault. Other possible methods include calculating the average value of the abnormality level for a day, rather than using data from each individual instance, and determining that there are signs of a fault if that value is greater than the maximum value of the daily average value when the equipment is functioning normally. Another possible method is to take a moving average over a number of business days that can comprehensively include all operations, taking into account the differences in daily operations, and determining that there are signs of a fault if that value is greater than the maximum value of the moving average of the abnormality level shown when the equipment is functioning normally. Furthermore, instead of using the maximum value of the abnormality level when the equipment is functioning normally, one can consider a method in which the newly obtained abnormality level shows a large fluctuation or variance over a certain period compared to the fluctuation or variance of the value over a certain period when the equipment is functioning normally, and it is determined that there are signs of a fault. Furthermore, one possible method involves calculating the standard deviation σ of the abnormality level under normal conditions and determining that an abnormality exists if the value deviates from the 3σ value, which is generally considered a statistically significant outlier. In other words, any method that can determine a statistically significant outlier by comparing it to the abnormality level under normal conditions or the trend of the abnormality level is acceptable. This is also true for other examples. 【0051】 In the process described above using Figures 9 and 10, the fault prediction diagnosis unit 150 uses the ratio of the difference between the operating time and the estimated operating time calculated by statistically processing the operating time included in the corrected operating history data group as the degree of abnormality. If the degree of abnormality of any one of the multiple moving bodies exceeds a predetermined threshold, it determines that there is a sign of abnormality in the driving-related equipment of that moving body. This allows for accurate detection of signs of abnormality in driving-related equipment. 【0052】(Method for notifying diagnostic results in the diagnostic result notification function 500) The details of the method for notifying diagnostic results in the diagnostic result notification function 500 will be explained below. The diagnostic result notification function 500 is a function in which, when the fault prediction diagnostic device 100 detects a fault precursor in the air compressor 210 of either railway vehicle 200a or railway vehicle 200b, it notifies the driver of the vehicle in question by displaying on the screen or by sounding an audible alarm, and further notifies the train depot of the train set number of the vehicle in question and the fact that a fault has been detected. Note that the notification target may be either the driver or the train depot, or just one of them. This is the same in other embodiments. 【0053】 As an example of a specific notification method, if the fault prediction diagnostic device 100 is installed on a train, it may output the diagnostic results to the vehicle information control device 240, and the vehicle information control device 240 may display a message indicating that a fault precursor has been detected on a display in the driver's cab (not shown), or the fault prediction diagnostic device 100 may display a message indicating that a fault precursor has been detected on a display in the driver's cab (not shown). Furthermore, if the fault prediction diagnostic device 100 has a function to transmit the aggregated vehicle data from the vehicle information control device to a ground system located in a train control center or maintenance depot using a public wireless communication network, the diagnostic results may be included in the vehicle data and transmitted to the ground system, thereby notifying the diagnostic results through a screen on the ground system. In addition, if the fault prediction diagnostic device 100 is implemented on a ground system located in a train depot, the train set number of the detected vehicle and the diagnostic results may be displayed on a screen on the ground system, or if the screen is monitoring multiple train sets simultaneously, the display of the detected vehicle may be notified by changing the background color or text color, for example, to indicate that a fault has been detected. This is also true in other examples. 【0054】(Modification 1) In Example 1, the data extraction process under the same conditions assumed a case where AS pressure data could not be obtained, so data when the starting speed and ending speed were 0 was excluded. However, when occupancy rate information can be obtained, or when some AS pressure data can be obtained, it is possible to estimate the amount of air consumed due to changes in sprung weight even when the starting speed and ending speed are 0, i.e., when the train is stopped at a station. Therefore, when the starting speed and ending speed are 0, the total displacement of the occupancy rate during operation or the total fluctuation amount of some of the AS pressure data that can be obtained while the air compressor 210 is operating is used as the air consumption condition, and when the starting speed and ending speed are greater than 0, the starting position and ending position used in Example 1 are used as the air consumption condition, so that data when the speed at which the air compressor 210 is operating is 0 can also be considered. 【0055】 Figure 11 shows the flow chart for estimating air consumption based on occupancy rate information. The differences from Figure 10 are explained below. Specifically, if occupancy rate information can be obtained, as shown in Figure 11, if the result in step 1602 of the same condition data extraction process is No, the process proceeds to step 1603b. In step 1603b, the values ​​of the total displacement of the occupancy rate and the total fluctuation of the BC pressure are extracted from the acquired corrected operating history data, set as air consumption conditions, and the process proceeds to step 1604. In this way, data for cases where the operating start speed and operating end speed are 0 can also be considered. 【0056】(Modification 2) In Example 1, the case in which BC pressure can be obtained was described, but a substitute method for the case in which BC pressure cannot be obtained will be described. If BC pressure cannot be obtained, the amount of compressed air consumed by the air brake system 220 can be estimated if it is known which strength of brake was applied and for how long. Here, generally, the strength of the brakes in railway vehicles can be calculated from the brake notch, but in some railway vehicles, in order to keep the deceleration for each notch constant, the force that presses the brake pads (amount of compressed air output) output for each notch is changed in accordance with the change in vehicle weight (number of passengers). From the above, the operation history data calculation unit 120 can substitute the BC pressure output consumption conditions by obtaining notch information and occupancy rate information as running information from the vehicle information control device, etc., and calculating the following values: ・Brake notch used while the air compressor 210 was in operation ・Usage time for each brake notch ・Occupancy rate while the air compressor 210 was in operation 【0057】 (Effects) According to the fault prediction diagnostic device 100 of the above-described embodiment 1, for multiple mobile units, the operating history data, which consists of the operating time calculated from the operating state of the running-related equipment of each mobile unit and the output usage conditions calculated from the operating state and running state of the output-related equipment, is used to perform a correction process on the operating time between running-related equipment 21 using a predetermined correction amount to eliminate individual differences, thereby treating it as common corrected operating history data. Then, the operating time in the newly acquired corrected operating history data is compared with the estimated operating time calculated using the operating time of past corrected operating history data sets of multiple mobile units under the same output usage conditions at that time, and the degree of abnormality is calculated, thereby shortening the period until the start of fault prediction diagnosis of running-related equipment. 【0058】 Furthermore, even when AS pressure information for the air spring system cannot be obtained, the system separates air consumption conditions into data taken while driving and data taken while stationary. By using occupancy rate information and some of the acquired AS pressure information for the stationary data, more data can be used, resulting in a fault prediction diagnostic device that can shorten the data accumulation period before fault prediction diagnosis can begin. 【0059】 In this embodiment, a second embodiment of the mobile body failure prediction diagnostic device 100 will be described below with reference to Figures 12 to 15. In this embodiment as well, the mobile bodies 20a and 20b are railway vehicles 200a and 200b, respectively, and the internal configuration of the railway vehicles 200a and 200b is the same as that used in Embodiment 1, as shown in Figure 2. In Embodiment 2, the operation history data calculation unit 120 can input information on air suspension pressure (AS pressure) as the operating state of the air spring device 230, and uses the integrated value of the air suspension pressure displaced while the air compressor 210 is operating (in this case, the total AS pressure fluctuation) as one of the air consumption conditions. By using the integrated value of the air suspension pressure (in this case, the total AS pressure fluctuation), it is possible to use an air consumption condition that is close to the actual amount of air used. 【0060】 Figure 12 shows the overall configuration of the fault prediction and diagnosis system when the moving objects are railway vehicles 200a and 200b. In Example 2, as shown in Figure 12, AS pressure information can be obtained from the air spring device 230. An example is described in which the air consumption conditions are the total fluctuation amount of BC pressure and the total fluctuation amount of AS pressure, and the operation history data shown in Figure 13, which will be described later, is used, and the target of correction of the operation history data is the air consumption conditions. Note that the block configuration and input / output are the same as in Example 1 and will therefore be omitted. Here, the contents of the operation history data, which differ in operation from Example 1, and the details of the operation of the operation history data calculation unit 120 and the operation history correction unit 130 will be described. 【0061】 Figure 13 shows an example of operational history data used in the fault prediction and diagnosis device 100. The operational history data includes, for example, information on the date, train number, total BC pressure fluctuation, and total AS pressure fluctuation. 【0062】The date indicates the date the operational history data was created. The train set number indicates the train set number of railway vehicle 200a. The operating time indicates the time [seconds] that the air compressor 210 was operating. The total BC pressure fluctuation indicates the total fluctuation of the brake cylinder pressure. The total AS pressure fluctuation indicates the total fluctuation of the air suspension pressure. In this case, the operational history data includes the operating time and air consumption conditions. The air consumption conditions are the cumulative values ​​of the fluctuations in BC pressure (total BC pressure fluctuation) and AS pressure (total AS pressure fluctuation) while the air compressor 210 is operating. 【0063】 (Operation of the Operation History Data Calculation Unit 120) Figure 14 is a diagram showing the processing flow of the Operation History Data Calculation Unit 120. In the following explanation, only the differences from Figure 6 of Example 1 will be explained. Step 1703 sets the time in the operation data as the start time of operation and proceeds to step 1704. Step 1705 sets the time in the operation data as the end time of operation and proceeds to step 1706. Step 1707 converts the train number, operating time, total BC pressure fluctuation, and total AS pressure fluctuation into the format of operation data (see Figure 5) and proceeds to step 1708. 【0064】 Step 1711 determines whether the air compressor operation information is ON and whether the current AS pressure is greater than the AS pressure from the previous cycle. If Yes, proceed to step 1712; otherwise, terminate the process. Step 1712 updates the total AS pressure fluctuation using the following formula and terminates the process: Total AS pressure fluctuation = Total AS pressure fluctuation from the previous cycle + (Current AS pressure - AS pressure from the previous cycle). Note that this embodiment is an example where there is only one AS pressure data in the vehicle. If there are multiple AS pressure data in the vehicle, steps 1711 and 1712 are repeated the number of times corresponding to the number of data points, and the total AS pressure fluctuation is the sum of those values. This is the same in other embodiments. 【0065】(Operation of the Operation History Correction Unit 130) Figure 15 shows the processing flow of the Operation History Correction Unit 130. Step 1801 is to input operation history data from the Operation History Data Calculation Unit 120 and proceed to step 1802. Step 1802 reads the train set number included in the operation history data, obtains the correction amount for the same train set from the Operation History Storage Unit 140, and proceeds to step 1803. Details of the correction amount will be described later. Step 1803 calculates the corrected total BC pressure fluctuation amount and the corrected total AS pressure fluctuation amount by multiplying the obtained correction amount by the total BC pressure fluctuation amount and the total AS pressure fluctuation amount, and proceeds to step 1804. In this embodiment, the case of multiplication has been described, but depending on the characteristics of the correction target, the correction amounts may be added or subtracted. This is also true in other embodiments. Step 1804 updates the total BC pressure fluctuation and total AS pressure fluctuation of the operation history data to the corrected total BC pressure fluctuation and corrected total AS pressure fluctuation, respectively, defines them as corrected operation history data, and proceeds to step 1805. Step 1805 outputs the corrected operation history data to the operation history storage unit 140 and the fault prediction diagnosis unit 150, and terminates the process. 【0066】 (Method for calculating correction amount) When correcting air consumption conditions, the correction amount is set to a value that eliminates individual differences in operating time caused by the equipment installed in each train set. For example, in the case of air compressors, as explained in Example 1, there are differences in the amount of air discharged from one unit to another, and even when compressing the same amount of air, the operating time will differ. To eliminate these individual differences, for example, the air consumption conditions and operating time of one train set are set as the standard air consumption conditions and standard operating time, respectively, and the air consumption conditions of other train sets are corrected to match the standard air consumption conditions. Specifically, for each train set, using data from a period when the equipment is functioning normally, such as immediately after shipment, a method can be considered in which correction amounts for each value of the air consumption conditions (in this example, the total fluctuation amount of BC pressure and the total fluctuation amount of AS pressure) are calculated using known statistical processing or machine learning methods, so that when the operating time is the same as the standard operating time, the air consumption conditions become the same as the standard air consumption conditions. 【0067】This can also be described as follows: When the operation history correction unit 130 is correcting output usage conditions (in this case, air consumption conditions), it sets the output usage conditions as the standard output usage conditions (in this case, standard air consumption conditions), and uses the operating time of one of the multiple moving objects (in this case, railway vehicles 200a, 200b) as the standard time. If the operating time and standard time of the multiple moving objects are the same, it corrects the output usage conditions by calculating values ​​using statistical processing or machine learning so that the values ​​included in the output usage conditions and the standard output usage conditions are the same for each of the multiple moving objects, and using these values ​​as the correction amounts for each of the multiple moving objects. This allows for more accurate correction of the operation history. 【0068】 (Effects) As described in Embodiment 2 above, corrected operating history data is calculated by applying a correction process targeting air consumption conditions to the operating history data of the driving-related equipment 21. By sharing the corrected operating history data of the driving-related equipment 21, it is possible to diagnose signs of failure in the driving-related equipment with high accuracy while shortening the preparation period until the method is applied. 【0069】It should be noted that the present invention is not limited to the embodiments described above, and various modifications are included. For example, the embodiments described above are explained in detail to make the present invention easier to understand, and are not necessarily limited to those having all the configurations described. Other embodiments that can be conceivable within the scope of the technical idea of ​​the present invention are also included within the scope of the present invention. Furthermore, it is possible to replace a part of the configuration of one embodiment with the configuration of another embodiment, and it is also possible to add the configuration of another embodiment to the configuration of one embodiment. Furthermore, it is possible to add, delete, or replace parts of the configuration of each embodiment with other configurations. In addition, some or all of the above configurations, functions, processing units, processing means, etc., may be realized in hardware, for example, by designing them as integrated circuits. Furthermore, each of the above configurations, functions, etc., may be realized in software by having a processor interpret and execute a program that realizes each function. Information such as programs, tables, files, etc. that realize each function can be stored in memory, a recording device such as a hard disk or SSD (Solid State Drive), or a recording medium such as an IC card, SD card, or DVD. Furthermore, control lines and information lines are shown only if they are considered necessary for explanation, and not all control lines and information lines are necessarily shown in the product. In practice, it can be assumed that almost all configurations are interconnected. 【0070】 According to the fault prediction diagnostic device 100 described in detail above, it is possible to detect signs of failure in driving-related equipment mounted on multiple mobile vehicles. In this case, the fault prediction diagnostic device 100 can shorten the preparation period required to introduce fault prediction diagnostics by taking into account individual differences in driving-related equipment, correcting for these individual differences, and standardizing the data of the driving-related equipment. 【0071】(Explanation of the failure prediction diagnosis method and program) The processing performed by the failure prediction diagnosis device 100 is realized through the cooperation of software and hardware resources. That is, a processor such as a CPU provided in the failure prediction diagnosis device 100 loads a program that realizes each function of the failure prediction diagnosis device 100 into memory and executes it to realize each of these functions. Therefore, the processing performed by the failure prediction diagnosis device 100 described above can be understood as a failure prediction diagnosis method for detecting failure signs of driving-related equipment mounted on multiple mobile bodies, in which the processor executes a program recorded in memory to calculate operating history data including the operating time of the driving-related equipment and output usage conditions calculated based on the operation information of output-using equipment that uses the output of the driving-related equipment, corrects the operating history data by a correction process that corrects the individual differences of each driving-related equipment 21, calculates corrected operating history data, and diagnoses failure signs of the driving-related equipment by comparing the corrected operating history data with a group of corrected operating history data of multiple past mobile bodies that satisfy predetermined conditions based on the output usage conditions. This allows us to provide a fault prediction diagnostic method that shortens the preparation period required for introducing fault prediction diagnostics by taking into account individual differences in driving-related equipment, correcting for these individual differences, and standardizing the data of driving-related equipment. 【0072】Furthermore, the program operating in the fault prediction diagnostic device 100 is a program for detecting signs of failure in running-related equipment mounted on multiple mobile bodies, and can be understood as a program that enables the computer to perform the following functions: calculate operating history data including the operating time of the running-related equipment and output usage conditions calculated based on the operation information of output-using equipment that uses the output of the running-related equipment; correct the operating history data by a correction process that corrects the individual differences of each running-related equipment 21 and calculate corrected operating history data; and diagnose signs of failure in the running-related equipment by comparing the corrected operating history data with a group of corrected operating history data from multiple past mobile bodies that satisfy predetermined conditions based on the output usage conditions. As a result, the computer can implement a function that shortens the preparation period required for introducing fault prediction diagnostics by considering the individual differences of the running-related equipment, correcting those individual differences, and standardizing the data of the running equipment. 【0073】 20, 20b... Mobile unit, 21... Running-related equipment, 22a, 22b... Output-using equipment, 23... Running information acquisition unit, 24, 250... Data output unit, 100... Fault prediction diagnosis device, 110... Data input unit, 120... Operation history data calculation unit, 130... Operation history correction unit, 140... Operation history storage unit, 150... Fault prediction diagnosis unit, 200a, 200b... Railway vehicle, 220... Air brake system, 230... Air spring system, 240... Vehicle information control device, 500... Diagnosis result notification function

Claims

1. A fault prediction diagnostic device for detecting signs of failure in running-related equipment mounted on multiple mobile bodies, comprising: an operation history data calculation unit that calculates operation history data including the operating time of the running-related equipment and output usage conditions calculated based on the operation information of output-using equipment that uses the output of the running-related equipment; an operation history correction unit that corrects the operation history data by a correction process that corrects individual differences of each of the running-related equipment and calculates corrected operation history data; and a fault prediction diagnostic unit that diagnoses signs of failure in the running-related equipment by comparing the corrected operation history data with a group of corrected operation history data of multiple past mobile bodies that satisfy predetermined conditions based on the output usage conditions.

2. The fault prediction and diagnosis device according to claim 1, wherein the moving body is a railway vehicle, the running-related equipment is an air compressor that generates compressed air, the output-using equipment is a compressed air-using equipment that uses the compressed air, and the fault prediction and diagnosis device is a fault prediction and diagnosis device that diagnoses the air system systems of a plurality of railway vehicles.

3. The fault prediction diagnostic device according to claim 2, further comprising a data input unit that acquires running information from a railway vehicle, including the operating status of the air compressor, the operating status of the compressed air-using equipment, and the position and speed of the railway vehicle, wherein the operation history data calculation unit calculates the operating time of the air compressor from the acquired operating information of the air compressor, calculates the air consumption conditions during operation of the air compressor from the operating information of the compressed air-using equipment and the running information of the railway vehicle as output usage conditions, and calculates the operating time and the air consumption conditions as operation history data.

4. The fault prediction diagnostic device according to claim 3, wherein the compressed air-using equipment includes an air brake device, and if the operating history data calculation unit can input information on brake cylinder pressure as the operating state of the air brake device, it uses the cumulative value of the brake cylinder pressure that fluctuated while the air compressor was in operation as one of the air consumption conditions.

5. The fault prediction diagnostic device according to claim 3, wherein the compressed air-using equipment includes an air spring device, and if the operation history data calculation unit can input information on the air suspension pressure as the operating state of the air spring device, it uses the cumulative value of the air suspension pressure that fluctuated while the air compressor was in operation as one of the air consumption conditions.

6. The fault prediction diagnostic device according to any one of claims 1 to 5, wherein the operation history correction unit, when the target of correction is the operation time, uses the standard operation time calculated from the performance of the driving-related equipment or the operation time of one of the multiple mobile bodies as the reference time, and for the multiple mobile bodies, uses the result of statistical processing of the difference between the operation time of each target of correction and the reference time as the correction amount for each of the multiple mobile bodies, and when the target of correction is the output usage conditions, uses the output usage conditions as the reference output usage conditions and uses the operation time of one of the multiple mobile bodies as the reference time, and for the multiple mobile bodies, when the operation time of each of the multiple mobile bodies is the same, uses the value calculated by statistical processing or machine learning so that the output usage conditions and each value included in the reference output usage conditions are the same as the value for each of the multiple mobile bodies, thereby correcting the output usage conditions.

7. The fault prediction diagnostic device according to any one of claims 1 to 6, wherein the fault prediction diagnostic unit determines that there is a sign of abnormality in the running-related equipment of any one of the multiple moving bodies when the abnormality of any one of the moving bodies exceeds a predetermined threshold, based on the ratio of the difference between the operating time and the estimated operating time calculated by statistically processing the operating time included in the corrected operating history data group, and the fault prediction diagnostic unit determines that there is a sign of abnormality in the running-related equipment of the moving body.

8. A failure prediction diagnostic method for detecting signs of failure in running-related equipment mounted on multiple mobile bodies, comprising: a processor executing a program recorded in memory to calculate operating history data including the operating time of the running-related equipment and output usage conditions calculated based on the operation information of output-using equipment that uses the output of the running-related equipment; correcting the operating history data by a correction process that corrects individual differences of each of the running-related equipment to calculate corrected operating history data; and diagnosing signs of failure in the running-related equipment by comparing the corrected operating history data with a group of corrected operating history data from multiple past mobile bodies that satisfy predetermined conditions based on the output usage conditions.

9. The fault prediction and diagnosis method according to claim 8, wherein the moving body is a railway vehicle, the running-related equipment is an air compressor that generates compressed air, and the output-using equipment is a compressed air-using equipment that uses the compressed air, and the air system systems of a plurality of railway vehicles are the target of diagnosis.

10. A fault prediction diagnosis method according to claim 9, further comprising: acquiring from a railway vehicle the operating status of the air compressor, the operating status of the compressed air-using equipment, the position and speed of the railway vehicle; calculating the operating time of the air compressor from the acquired operating information of the air compressor; calculating the air consumption conditions during operation of the air compressor from the operating information of the compressed air-using equipment and the running information of the railway vehicle as the output usage conditions; and calculating the operating time and the air consumption conditions as operating history data.

11. The fault prediction and diagnosis method according to claim 10, wherein the compressed air-using equipment includes an air brake device, and if information on the brake cylinder pressure can be input as the operating state of the air brake device, the cumulative value of the brake cylinder pressure that fluctuated while the air compressor was in operation is used as one of the air consumption conditions.

12. The fault prediction and diagnosis method according to claim 10, wherein the compressed air-using equipment includes an air spring device, and if information on the air suspension pressure can be input as the operating state of the air spring device, the cumulative value of the air suspension pressure that fluctuated while the air compressor was in operation is used as one of the air consumption conditions.

13. The fault prediction and diagnosis method according to any one of claims 8 to 12, wherein, when the target of correction is the operating time, the operating time is corrected by using the standard operating time calculated from the performance of the driving-related equipment or the operating time of one of the multiple mobile bodies as the reference time, and the difference between each of the multiple mobile bodies that is the target of correction is statistically processed and the result is used as the correction amount for each of the multiple mobile bodies; and when the target of correction is the output usage conditions, the output usage conditions are used as the reference output usage conditions and the operating time of one of the multiple mobile bodies as the reference time, and the output usage conditions are corrected by using values ​​calculated by statistical processing or machine learning so that each of the values ​​included in the output usage conditions and the reference output usage conditions are the same when the operating time of each of the multiple mobile bodies is the same.

14. A failure prediction diagnosis method according to any one of claims 8 to 13, wherein the ratio of the difference between the operating time and the estimated operating time calculated by statistically processing the operating time included in the corrected operating history data group is defined as the degree of abnormality, and if the degree of abnormality of any one of the multiple moving bodies exceeds a predetermined threshold, it is determined that there is a sign of abnormality in the driving-related equipment of that moving body.

15. A program for detecting signs of failure in driving-related equipment mounted on multiple mobile bodies, the program comprising: a function for a computer to calculate operating history data including the operating time of the driving-related equipment and output usage conditions calculated based on the operation information of output-using equipment that uses the output of the driving-related equipment; a function to correct the operating history data by a correction process that corrects individual differences in each of the driving-related equipment, and to calculate corrected operating history data; and a function to diagnose signs of failure in the driving-related equipment by comparing the corrected operating history data with a group of corrected operating history data from multiple past mobile bodies that satisfy predetermined conditions based on the output usage conditions.

Images