Method and system for determining the state of a component of a train in order to determine a maintenance plan for the train

Abstract

The invention relates to a method for determining (S6) the state of a component (4) of a train (2) in order to determine (S8) a maintenance plan for the train (2). The method has the steps of reading (S5) data relating to a vibration of the component (4) from a sensor (6) of the train (2), reading (S4) data relating to a vibration of the component (4) from a mobile terminal (8), and determining (S6) the state of the component (4) on the basis of the read data of the sensor (6) and of the mobile terminal (8). The invention additionally relates to a system comprising a sensor (6) which can be installed on a component (4) of a train (2), a mobile terminal (8), and a server (10), wherein the server (10) is designed to carry out the step of determining (S6) the state of the component (4) and to carry out a step of determining (S8) a maintenance plan for the train (2) on the basis of the determined state, and the system is designed to carry out the method.

Description

[0001] Method and system for determining a condition of a component of a train for determining a maintenance plan of the train

[0002] Technical area

[0003] The present invention relates to a method for determining the condition of a component of a train based on data read from a sensor and a mobile terminal, as well as to a method for determining a maintenance plan for the train depending on the determined condition of the component. Furthermore, the present invention relates to a system for determining the condition of a component of a train and for determining a maintenance plan for the train depending on a determined condition.

[0004] State of the art

[0005] Trains represent a vital component of transportation. Maintenance is necessary for efficient train operation. Maintenance should not be performed too frequently, as each maintenance operation results in reduced train utilization. Maintenance should also not be performed too infrequently, as otherwise consequential damage and unplanned downtimes due to undetected train damage become more likely. For optimal maintenance planning, the condition of the train is determined. W02023043801 describes how the condition of a train is determined based on vibration measurements in the train.

[0006] Description of the invention

[0007] In a first aspect, the present invention relates to a method for determining a condition of a component of a train for determining a maintenance plan for the train. Steps of the method can be carried out, for example, by a computing device, such as a backend formed by a server. The component of the train can be, for example, a wheel or a wheel bearing. The train can be, for example, a passenger train or a freight train. The condition of the component can include information about the wear of the component. The condition of the component can include information about the age of the component. The condition of the component can include information about the use of the component. The maintenance plan of the train can contain information about when the train or components of the train should be serviced. When determining the maintenance plan, it can be determined, for example, which components should be serviced at what time.

[0008] The method comprises reading in data relating to a vibration of the component from a sensor of the train. The sensor can be configured to acquire data relating to a vibration of the component. The sensor can be part of the train and alternatively or additionally arranged close to or on the component. For example, if the component is a wheel of the train, the sensor can be arranged close to the wheel. The sensor can be configured to acquire structure-borne sound. For example, the sensor can be a piezoelectric sensor or an electromagnetic sensor. Reading in the data relating to the vibration of the component from the sensor can be carried out by reading in the data from the sensor.Alternatively or additionally, the reading of data on the vibration of the component from the sensor can be carried out from a further computing device, wherein the data on the vibration of the component have previously been sent from the sensor to the further computing device.

[0009] The method comprises reading in data relating to a vibration of the component from a mobile device. The mobile device can be a smartphone or a tablet, for example. The mobile device can be configured to acquire data relating to the vibration of the component. Data relating to the vibration of the component can be read in from the mobile device, for example directly from the mobile device. Alternatively or additionally, the data relating to the vibration of the component can be read in from another computing device, wherein the data relating to the vibration of the component have previously been sent from the mobile device to the computing device.

[0010] When reading in data about the vibration of the component from the mobile device, data about an identical or a different vibration of the component can be read in as when reading in data about the vibration of the component from the sensor. For example, when reading in data about the vibration of the component from the mobile device, data about a first vibration can be read in. When reading in data about a vibration of the component from the sensor, data about a second vibration can be read in, whereby the first and the second vibration can be identical, partially match, or completely do not match. Two vibrations can be partially or completely match if at least parts of the frequencies of the vibrations match or if all frequencies of the vibrations match.The steps of reading component vibration data from the mobile device and reading component vibration data from the sensor can be performed independently of each other. The reading steps can be performed at different times or, alternatively, simultaneously.

[0011] The method includes determining a condition of the component based on the data read from the sensor and the mobile device. For example, when determining the condition of the component, bearing damage on the wheel or, alternatively, damage to a wheel tread can be identified. The condition of the component can be determined based on relationships between specific frequencies of the read vibrations and known conditions of the component. The condition of the component can be determined, for example, using a table, where the table can link information about the vibrations and information about the condition of the component.

[0012] Such a method makes it possible to determine the condition of a component in order to determine a maintenance plan for the train depending on the determined condition. The condition of the component is determined not only based on data on a vibration of the component read in from a sensor, but also and in combination with data on the vibration of the component read in from a mobile device. Thus, data on a vibration of the component is recorded by two different devices and used to determine the condition of the component. This can improve the hit rate for damage detection with regard to the condition of the component, for example, compared to methods that only use the condition based on data on a vibration from a sensor. This allows the condition of the component to be determined more precisely and thus, for example, damage detection can be improved.For example, bearing damage on a train wheel can be more easily detected. This can improve the determination of the maintenance schedule based on the specific condition, as a more precisely determined condition also allows the maintenance schedule to be determined more precisely. For example, in cases where the condition of a component shows only slight wear, the maintenance schedule can be determined so that the train and component do not yet require maintenance, thus avoiding unnecessary inspection. Alternatively, the method can determine that the component already requires maintenance, thus preventing a potential failure. This allows for more efficient fleet management and maintenance planning.

[0013] Damage predictions regarding the condition of the component are therefore more accurate and unnecessary maintenance of the train and the component can be prevented, and necessary maintenance can be identified more quickly using the method. Furthermore, the method uses a mobile device. The method can therefore also be applied by retrofitting a train that has already been delivered, as the mobile device can be easily combined with the train's existing sensors and used to carry out the method. The mobile device as a second data source can therefore improve the hit rate and probability of correctly predicting the condition of the component and thus, for example, damage to the train's wheel. The method can be used to determine the conditions and maintenance plans of one or more components of the train, whereby the condition of a respective component can be determined based on data on vibrations of the respective component.

[0014] According to a further embodiment, the method can be characterized in that data relating to a vibration can be acquired using a microphone of the mobile terminal. Airborne sound can be acquired using the microphone of the mobile terminal. The airborne sound can be induced by structure-borne sound from the component. Thus, the mobile terminal can be configured to acquire airborne sound initiated by structure-borne sound using the microphone. The acquisition can be carried out using one or more microphones of the mobile terminal. The acquisition can also be carried out using multiple microphones of multiple mobile terminals. The acquisition of data relating to a vibration using the microphone can be carried out with respect to a different vibration or the same vibration as the acquisition of data using the sensor.The acquisition of vibration data with the microphone can be performed independently of the acquisition of vibration data with the sensor, both spatially and temporally. Furthermore, the reading of data from the mobile device can be performed using the acquired data. The vibration data can be acquired first with the mobile device's microphone, and then the acquired data can be read from the mobile device.

[0015] This means that standard mobile devices such as smartphones or tablets with built-in microphones can be used to perform the procedure. Special mobile devices are therefore not necessary. This makes the procedure particularly simple and can be performed using conventional mobile devices.

[0016] According to a further embodiment, the method can be characterized in that the recording can be carried out during operation of the train. The recording of data relating to the vibration of the component using the mobile terminal can be carried out during operation of the train. Alternatively or additionally, the recording of data relating to the vibration of the component can be carried out using one or more sensors of the train during operation of the train. During operation of the train, the train can be in motion. The operation of the train can be defined by the movement of the train. Alternatively or additionally, the operation of the train can be defined by the operation of a drive motor of the train, such as an electric motor or a diesel engine. Thus, the train is in operation when the train is in motion and alternatively or additionally when the drive motor is switched on.

[0017] This ensures that the train is in operation when collecting data on the component's vibration, allowing all vibrations of the component that occur during train operation to be recorded. This allows the condition of the component to be determined more precisely, as various vibrations of the component can be recorded.

[0018] According to a further embodiment, the method can be characterized in that an application can be executed on the mobile device for capturing data with the mobile device. When executing the application, a user of the mobile device can be instructed to capture data using a microphone of the mobile device. For example, the application can instruct the user at which points and positions in the train the vibration data should be captured using the microphone of the mobile device. Alternatively or additionally, the application can instruct the user at which times the data should be captured using the mobile device.

[0019] This allows a user of the mobile device, for example a technician with a smartphone, to be instructed on how, where, and when to collect vibration data from a component using the mobile device. This allows vibration data to be collected particularly quickly and efficiently using the mobile device without the user needing to be instructed elsewhere on how to collect the data. Unnecessary measurements and unnecessary steps in collecting vibration data using the mobile device can be avoided by providing the user with precise instructions via the application on which measurements and which collection steps should be carried out, where, when, and how, and which should not.

[0020] According to a further embodiment, the method can be characterized in that, based on the data read in from the sensor, a determination can be made as to how the data should be acquired using a microphone of the mobile terminal. For example, after data on the vibration of the component has been read in from the sensor, a determination can be made as to how the data should be acquired using the microphone of the mobile terminal. For this purpose, the data read in on the vibration of the component from the sensor can, for example, be classified and categorized. For example, it can be determined that, for a first component of the train, vibrations of a first frequency with a particularly high first amplitude occur and that, for a second component of the train, vibrations of a second frequency that is different from the first frequency occur.In this case, a second amplitude of the second frequency can be detected, which can be lower than the first amplitude. The first frequency with the first amplitude can be specific for a state of wear of the first component, whereby the second frequency with the second amplitude can be non-specific for a state of wear of the second component. It can then be determined that the recording of data with the microphone of the mobile device should only be for the first component, which exhibits vibrations at the first frequency due to the vibration behavior determined with the sensor. Recording of data with the microphone of the mobile device on the second component can be dispensable because no state of wear of the second component has been determined based on the data read in by the sensor.Alternatively or in addition to determining how data should be collected using the microphone of the mobile device, it is also possible to determine when data should be collected using the microphone of the mobile device. For example, it can be determined that data should be collected using the microphone of the mobile device at a certain speed of the train or on a certain route, such as cornering or traveling over uneven tracks. If, in an alternative embodiment, no determination is made as to how data should be collected using the microphone of the mobile device based on the data read in from the sensor, data can be collected using the microphone and the sensor at the same time or independently of one another in terms of time.This means that the steps of reading the data from the sensor and the mobile device can be carried out simultaneously or independently of each other.

[0021] The method can therefore particularly efficiently capture data using the microphone of the mobile device. For example, data can only be captured using the microphone of the mobile device at those locations and times when it has also been determined that, based on the data from the sensor, data capture using the mobile device is necessary to precisely determine the condition of the component. For example, based on the data read in from the sensor regarding the vibration of the component, it can be determined that data should be captured using the microphone of the mobile device at a first bogie and a second wheel of the train. At other locations on the train, data capture using the microphone of the mobile device can be omitted. This provides a particularly efficient method for determining the condition of the component for determining the train's maintenance plan.For example, data can be recorded using the mobile device's microphone, either wheel-specific or additionally axle-specific. The user can then be guided to specific locations, such as axles or wheels of the train, via the application's instructions, where they can record data using the mobile device's microphone. Alternatively or additionally, different steps for recording data can be carried out at different locations using the mobile device's microphone. At a first location, data can be recorded using the mobile device's microphone for 2 seconds, and at a second location, different from the first location, data can be recorded using the mobile device's microphone for 5 seconds. This means that data can be recorded using the microphone on a location-specific basis.On the one hand, this can minimize time, memory, and computing effort, and on the other hand, the state can be determined more precisely because data is collected using the mobile device's microphone for a sufficiently long time at the points where it is necessary. This means that particularly detailed data on the component's vibration can be recorded at these points. This can depend on which data on the component's vibration was previously read in by the sensor. Based on the data on the component's vibration read in by the sensor, a previous state can be determined, and depending on the previous state, it can be determined how the data should be collected using the mobile device's microphone.For this purpose, the previous state can be improved by capturing data with the microphone of the mobile device in order to be able to determine the state or final state of the component based on the data read in from both the sensor and the mobile device. For example, if damage is suspected, data can be captured using the microphone based on the data on the vibration of the component read in from the sensor. In other cases, reading in the data on the vibration of the component from the sensor may be sufficient to determine the state of the component. This avoids unnecessary steps of capturing data with the microphone of the mobile device, since data should only be captured with the microphone when necessary. The predetermined state can be improved using the vibration data from the mobile device.

[0022] According to a further embodiment, the method can be characterized in that data relating to a vibration can be sent from the mobile terminal to a server. For example, the data relating to the vibration can be sent from the mobile terminal to the server when the data has been completely captured using the microphone of the mobile terminal. Alternatively, the data can be sent from the mobile terminal to the server when at least part of the data has been captured using the microphone of the mobile terminal, for example when data relating to vibrations of components of a train carriage has been captured. During the transmission, the data relating to the vibration of the component captured by the mobile terminal can be sent. The server can be implemented via a cloud and as a backend.For example, all data relating to the train's components acquired by the mobile device can be sent to the server in the sending step. Furthermore, vibration data can be sent from the sensor to the server. The data acquired by the sensor can be sent to the server. The transmission can be performed via a gateway to the transmitter. During the reading step, the sent data can be read into the server. Thus, the server can read in data relating to the component's vibration from the sensor and data relating to the component's vibration from the mobile device, so that the acquired and sent data relating to the component's vibration is read into the server.

[0023] The server can provide a particularly high level of computing power. This means that the step of determining the state using the read-in and sent data on the component's vibration can be carried out in the server, for example, web-based. The mobile device and the train's sensor therefore do not have to be trained and configured to carry out the step of determining the state. The data can be sent and thus communication between the sensor, mobile device, and server can take place via a gateway on the train and wireless or wired communication. For example, the data can be sent via cellular or Bluetooth. A first send can be made via Bluetooth from the sensor to the gateway, and a second send of the recorded data can be made from the gateway to the backend via cellular. A virtual private network can be used to send the data to ensure secure data transmission.Thus, even low-performance sensors and mobile devices can be used to execute the method, since computationally intensive steps, such as determining the state of the component, can be performed in the external server.

[0024] According to a further embodiment, the method can be characterized in that the determination of a state of the component can be carried out using a trained neural network. The trained neural network can, for example, have been trained using training data. The training data can contain data on the vibration of the component as well as specific states of the component. The neural network can be trained using supervised learning. The training and, alternatively or additionally, the determination of the state using the trained neural network can take place in the server. A neural network trained in this way can be used to determine the state of the component based on the data read in from the sensor and the mobile device.

[0025] Determining the condition using a trained neural network can enable a more precise determination of the component's condition. This can be done based on previously recorded and imported data on component vibrations from the sensor and mobile device, as well as determined and verified component conditions. This information can be obtained from previous condition determination steps based on imported data on component vibrations and comparison with reality via a user, such as a technician. Thus, fleet intelligence can be trained using the neural network, with information and data from various trains in a fleet being reflected in the trained neural network, enabling a particularly precise determination of the component's condition.

[0026] According to a further embodiment, the method can be characterized in that information relating to the specific condition can be sent to a display device. The sending can take place from the backend to the display device. The display device can, for example, be a display device of the mobile terminal, such as a cell phone of a technician on-site and on the train. Alternatively or additionally, the display device can be another display, such as a display statically installed in the train, which the technician can access. Based on information relating to the specific condition displayed in the display device, a user, such as a technician, can perform further steps, such as determining a maintenance plan by defining a maintenance interval for the component.The transmission of information relating to the specific state can then, for example, only take place once the determination of the state of the component has been fully carried out. In one embodiment, information relating to a predetermined state is first sent to the display device, and the predetermined state is displayed. The user can thus see the predetermined state on the display device based solely on data relating to the vibration of the component read in from the sensor. Data relating to the vibration of the component is then acquired and read in from the mobile device, and the state is determined based on data read in from the sensor and the mobile device. Information relating to the specific state is then sent, and this transmitted and received information relating to the specific state is displayed on the display device.Thus, the predetermined state displayed on the display device is updated and actualized by the final determined state.

[0027] A second aspect of the present invention relates to a method for determining a maintenance plan for a train depending on a condition of a component of the train determined using a method according to an embodiment of the first aspect of the present invention. The maintenance plan can be determined by the server and alternatively or additionally by a user, such as a technician. If the determination is made by a technician, this can be done by sending information about the determined condition to the display device and displaying the information about the determined condition on the display device. The user can recognize from the display that certain wheels exhibit a certain level of wear, and as a result, the user can determine a maintenance plan for these wheels, for example via an input.Alternatively or additionally, the maintenance plan can be determined automatically and by the server, based, for example, on previously determined maintenance plans and conditions and tables. Alternatively, a neural network can also be used to determine a maintenance plan. Thus, information about the determined condition is used to determine the maintenance plan. A third aspect of the present invention relates to a system comprising a sensor that can be mounted on a component of a train, a mobile terminal, and a server. The component can be a wheel or a wheel suspension. Communication between the sensor, the terminal, and the server can take place via a gateway and wireless or wired communication. Furthermore, the communication can take place via further transceiver devices. The sensor is configured to acquire data relating to a vibration of the component.The sensor can, for example, be a piezoelectric sensor and detect vibrations of a wheel. The mobile device can be configured to detect data relating to a vibration of the component. For example, the mobile device can be embodied as a smartphone with a microphone, and this smartphone can be configured to detect airborne sound induced by structure-borne sound, wherein the structure-borne sound can correspond to the vibration of the component, in this case the wheel. Furthermore, the sensor and the mobile device can be configured to send the detected data to the server. The sensor can be configured to send the detected data to the server via a gateway of the train. The server is configured to read in the detected data and to determine a state of the component based on the read-in data.For example, the server is configured to determine the wear level of a wheel based on the data on the vibration of the component recorded by the sensor and the mobile device. The server is further configured to determine a maintenance plan for the train depending on a specific condition. For example, it may have been previously determined that certain wheels exhibit a certain level of wear. Depending on this, the server can determine that certain wheels need to be serviced and replaced in the future, for example. Furthermore, the server can be configured to send information on the specific condition to a display device. The display device can be configured to display the information on the specific condition, whereby a user, such as a technician, can have this information displayed.According to one embodiment, the maintenance plan can be determined based on inputs from the user, whereby the user can make the inputs based on the displayed information about the specific condition. In one embodiment, the maintenance plan is also determined solely by the user, and the server only determines the condition of the component but not the maintenance plan. The system can also comprise the display device, whereby the display device can, for example, be part of the mobile device or part of the train or another device.

[0028] According to a further embodiment, the system may be characterized in that the system may be configured to carry out steps of methods according to an embodiment of the first aspect or the second aspect of the present invention.

[0029] Short description of the characters

[0030] Figure 1 schematically shows steps of a method for determining a condition of a component and for determining a maintenance plan of the train depending on the determined condition.

[0031] Figure 2 schematically shows a system for carrying out steps of the method shown in Figure 1.

[0032] Detailed description of embodiments

[0033] Figure 1 schematically shows steps of a method for determining S6 a state of a component 4 of a train 2 and for determining S8 a maintenance plan for the train 2. Steps of the method are carried out by a system. The system comprises a sensor 6, which is mounted on the component 4 of the train 2, as well as a mobile terminal 8 and a server 10, here forming a backend. These elements of the system are shown schematically in Figure 2. Furthermore, the system comprises a display device 12, here a display. The mobile terminal 8, the sensor 6, the server 10 and the display device 12 are communicatively connected to one another. The sensor 6 is communicatively connected to the server 10 via a gateway (not shown). The terminal 8 is communicatively connected to the server 10. The server is communicatively connected to the display device 12.The server 10 acts as a central data interface between the sensor 6, the mobile terminal 8, and the display device 12. The sensor 6 is configured to acquire data SO relating to a vibration of the component 4. The sensor 6, embodied here as a piezoelectric sensor, measures a vibration of the component 4 embodied as a wheel. The sensor 6 is configured to transmit data S3 relating to a vibration from the sensor 6 to the server 10. This is done via the gateway. Data relating to the vibration of the component 4 is then read in S5 from the sensor 6 of the train 2, with the server 10 being configured to carry out the reading S5 of the acquired data. The server 10 is configured to predetermine a state of the component 4. This is done depending on the vibration data read in from the sensor 6.

[0034] The server 10 determines S9 how data collection S1 should be carried out using a microphone of the mobile terminal 8. This determination S9 is carried out based on the data read in from the sensor 6, wherein the determination S9 is carried out depending on the predetermined state of the component 4. During the determination S9, it is determined at which points in the train 2 and the component 4 and at which times the data collection S1 should be carried out using the microphone of the mobile terminal 8. Thus, a first state is predetermined for a first wheel, and a second state is predetermined for a second wheel. The first state indicates that the first wheel probably requires maintenance, wherein the second state indicates that the second wheel does not require maintenance.During determination S9, it is now determined that in order to improve a damage prediction of the first wheel, the acquisition S1 of data with the microphone of the mobile terminal 8 should only be carried out on the first wheel, but not on the second wheel.

[0035] The mobile terminal 8 is configured to collect S1 data relating to a vibration of component 4 using the microphone of the mobile terminal 8 as a function of the determination S9. This collection S1 is carried out during the operation of the train 2. For example, the train 2 travels from location A to location B and has a certain speed. Certain vibrations occur due to the travel and operation of the train 2. These vibrations are collected, among other things, in the collection step S1. For collection S1, an execution S1.1 of an application is carried out on the mobile terminal 8. During execution S1.1 of the application, a user of the mobile terminal 8, in this case a technician with a smartphone, is instructed to collect S1 data using the microphone of the mobile terminal 8.The application guides the user at which points of component 4 and train 2 and at which times the acquisition S1 of data should be carried out using the microphone of the mobile device 8.

[0036] Then, a transmission S2 of data relating to a vibration is carried out from the mobile terminal 8 to the server 10. The transmission takes place via a mobile radio network. For this purpose, the mobile terminal 8 is set up to transmit S2 data relating to the vibration from the mobile terminal 8. Then, a reading S4 of data relating to the vibration of component 4 from the mobile terminal 8 is carried out by the server 10. The reading S4 is carried out using the data acquired by the mobile terminal 8. The reading S4 of data relating to a vibration of component 4 from the mobile terminal 8 is carried out after the reading S5 of data relating to a vibration of component 4 from the sensor 6. The acquisition S1 of data relating to a vibration with the microphone of the mobile terminal 8 is carried out after the acquisition S0 of data relating to a vibration of component 4 with the sensor 6. During the reading S4, S5, the sent data are read into the server 10.

[0037] A determination S6 of a state of component 4 is carried out by server 10 based on the data read in from sensor 6 and mobile device 8. The determination S6 of the state of component 4 is carried out using a trained neural network. The neural network was previously trained with training data in server 10. The training data includes data on vibrations of a component (not shown here) from other trains (not shown here), as well as states of these components that belong to the vibrations. The neural network was trained using supervised training. During determination S6, the predetermined state of component 4 is improved, and a final state of component 4 is determined based on data on vibrations from sensor 6 and mobile device 8.

[0038] Furthermore, the server 10 is configured to determine S8 a maintenance plan of the

[0039] Train 2 depending on the specific condition. For example, if it is determined that certain wheels have a certain level of wear, server 10 determines that these wheels require maintenance and, if necessary, replacement.

[0040] Furthermore, information on the specific condition is sent S7 to the display device 12. The user thus receives information on the specific condition via the display device 12. Using an input not described in detail here, the user thus inputs information to the server 10 on how to determine a maintenance plan. For example, based on the information on the specific condition displayed on the display device 12, the user recognizes that certain wheels of the train 2 need to be serviced and replaced. They input this information to the server 10. According to one embodiment, the server 10 is thus configured to determine S8 the maintenance plan depending on the specific condition and depending on the user input.

[0041] Reference symbol

[0042] 2 train

[0043] 4 Component

[0044] 6 Sensor

[0045] 8 Mobile device

[0046] 10 servers

[0047] 12 Display device

[0048] 50 Collecting data on a vibration of the component with the sensor

[0049] 51 Collecting data on a vibration of the component using a microphone of the mobile device

[0050] S1 .1 Running an application on the mobile device to collect data with the mobile device

[0051] 52 Sending data to a vibration from the mobile device

[0052] 53 Sending data about a vibration from the sensor

[0053] 54 Reading data on a vibration of the component from the mobile device

[0054] 55 Reading data on a vibration of the component from the sensor

[0055] 56 Determining a state of the component

[0056] 57 Sending information about a specific state

[0057] 58 Determining a maintenance plan

[0058] 59 Determine how data collection should be carried out using the mobile device's microphone

Claims

Patent claims 1 . Method for determining (S6) a state of a component (4) of a train (2) for determining (S8) a maintenance plan of the train (2), comprising the steps of: reading (S5) data relating to a vibration of the component (4) from a sensor (6) of the train (2); reading (S4) data relating to a vibration of the component (4) from a mobile terminal (8); and determining (S6) a state of the component (4) based on the read-in data from the sensor (6) and the mobile terminal (8).

2. Method according to claim 1, characterized in that a detection (S1) of data relating to a vibration is carried out using a microphone of the mobile terminal (8) and the reading (S4) of data from the mobile terminal (8) is carried out using the detected data.

3. Method according to the preceding claim, characterized in that the detection (S1) is carried out during operation of the train (2).

4. Method according to one of claims 2 or 3, characterized in that an execution (S1.1) of an application on the mobile terminal (8) for capturing (S1) data with the mobile terminal (8) is carried out, wherein Execution (S1.1) of the application instructs a user of the mobile terminal (8) to carry out the acquisition (S1) of data using a microphone of the mobile terminal (8).

5. Method according to one of claims 2 to 4, characterized in that based on the data read in from the sensor (6) a determination (S9) is carried out as to how the acquisition (S1) of data is to be carried out with a microphone of the mobile terminal (8).

6. Method according to one of the preceding claims, characterized in that a transmission (S2) of data relating to an oscillation is carried out from the mobile terminal (8) to a server (10), and a transmission (S3) of data to a Vibration is carried out from the sensor (6) to the server (10), wherein during reading (S4, S5) the sent data are read into the server (10).

7. Method according to one of the preceding claims, characterized in that the determination (S6) of a state of the component (4) is carried out by means of a trained neural network.

8. Method according to one of the preceding claims, characterized in that a transmission (S7) of information on the specific state to a display device (12) is carried out.

9. Method for determining (S8) a maintenance plan of a train (2) as a function of a state of a component (4) of the train (2) determined using a method according to one of the preceding claims.

10. System comprising a sensor (6) which can be mounted on a component (4) of a train (2), a mobile terminal (8) and a server (10), wherein the sensor (6) is configured to acquire (S0) data relating to a vibration of the component (4), wherein the mobile terminal (8) is configured to acquire (S1) data relating to a vibration of the component (4), wherein the sensor (6) and the mobile terminal (8) are configured to Sending (S2, S3) the acquired data to the server (10), wherein the server (10) is configured to read in (S4, S5) the acquired data and to determine (S6) a state of the component (4) based on the read-in data, and wherein the server (10) is further configured to determine (S8) a maintenance plan of the train (2) depending on the determined state. 1 1. System according to the preceding claim, characterized in that the system is arranged to carry out steps of methods according to one of claims 1 to 9.

Images