System for performing work on a track

Abstract

The invention relates to a system for working on a track (2) using a track maintenance machine (1), comprising a machine controller (4) and a working unit (3) actuated by the machine controller, wherein sensors (6) are provided for monitoring the working unit (3). The sensors (6) are coupled to a data acquisition module (7) for individually acquiring sensor data (S D ). The data acquisition module (7) is connected to a computing unit (8) in which a first algorithm (P1) is provided for calculating result data (E D ) from the sensor data (S D ). The system thus comprises additional structural components for processing the sensor signals (S S ). By means of the data acquisition module (7) and the computing unit (8), a differentiated evaluation of the working operation can be carried out independently of the existing monitoring function.

Description

Technical Field

[0001] This invention relates to a system for working on tracks using a track maintenance machine. The system includes a machine controller and a work unit controlled by the controller, wherein sensors are provided for monitoring work parameters. Furthermore, this invention relates to a method for operating said system. Background Technology

[0002] Patent AT 520698 A1 discloses a general system for monitoring the load on tamping units while they are operating on a track. To achieve this, sensors are included that record measurement data over a period of time and forward this data to an evaluation device. Based on this measurement data, the load timeline of the tamping unit's cyclical operation sequence can be determined. From this, conclusions can be drawn regarding the load status of the tamping unit, and maintenance measures or intervals can be determined based on these conclusions. Summary of the Invention

[0003] The object of this invention is to expand the advantages of sensors applicable to the types of systems described above. Furthermore, this document also provides a correspondingly improved method for operating such a system.

[0004] According to the present invention, a sensor is coupled to a data acquisition module for separately recording sensor data. This data acquisition module is connected to a computing unit, in which a first algorithm is configured to calculate result data based on the sensor data. Thus, the system includes additional structural components for processing sensor signals. By means of the data acquisition module and the computing unit in which the application-specific algorithm is configured, differentiated evaluation of operating modes can be performed independently of existing monitoring functions. A specific advantage of the present invention lies in the flexible configuration of sensor data recording and the selective adjustment of the calculation of result data.

[0005] In another improvement, the computing unit is configured to calculate at least one parameter based on sensor data recorded in the work sequence, wherein the computing unit is specifically coupled to the machine controller to automatically specify optimized work parameters. This enables continuous improvement of the work sequence performed using the work unit. The calculated parameters are adjusted according to the work unit in use and characterize the quality of the corresponding work sequence. Based on the above improvements, it is beneficial to establish a higher-level closed-loop control system at the level of a distributed control system.

[0006] Advantageously, the data acquisition module is configured for multi-channel data recording and coupled as a slave device to the computing unit, which is designed as an active device. This system architecture enables the efficient connection of several sensors to a subsystem consisting of the data acquisition module and the computing unit.

[0007] In another improvement, a monitoring device is provided for monitoring the work unit, which records sensor data at a lower sampling rate (e.g., 1 Hz) than the data acquisition module (e.g., its sampling rate is in the kHz range). This facilitates simple yet sufficient data processing for monitoring. Furthermore, for additional sensor evaluations performed by the computing unit, a dataset with high temporal resolution can be used.

[0008] According to a beneficial extension of the system, the computing unit is coupled to a database via a communication device to receive program data for modifying the first algorithm or for setting the second algorithm. This allows for simple modification of the evaluation performed by the computing unit. The system enables new analyses of job sequences without altering their structure. Furthermore, the system can be used to test new evaluation algorithms before adjustments to the job sequences are derived.

[0009] In this scenario, it is advantageous to include a VPN router in the communication setup. Thus, all devices connected to the VPN router can use a secure VPN tunnel. This involves computing units and other system components that exchange data with the database. A system-integrated VPN router can create more possibilities for the secure transmission of various types of data.

[0010] According to another improvement, the computing unit is connected to a storage device for storing sensor data and / or result data. Advantageously, the storage device is designed to store all result data, as well as all sensor data, if necessary, until the end of a specified readout interval. For example, this readout interval corresponds to the service interval of the monitored work unit. Furthermore, the data stored on the storage device can be accessed remotely at any time, preferably via a VPN tunnel. In particular, transmitting result data via remote access is useful. On the other hand, large amounts of sensor data are backed up in the storage device and read out when the system is modified.

[0011] To make the resulting data and, if necessary, sensor data centrally available, it is advantageous for the computing unit to be coupled to a computer network (cloud) via a modem for data transmission. This allows the data to be accessed at any time via online applications (web applications).

[0012] Advantageous embodiments of the system include tamping and / or stabilizing units as work units. These work units include vibratory tools for introducing vibrations onto the ballast track on which work has already been performed. Sensors mounted on the work units allow conclusions to be drawn regarding the quality of the track ballast bed and the quality of track ballast compaction. Therefore, the system provides not only information about the condition and function of the work units themselves, but also information about the condition of the track and the work being performed.

[0013] Advantageously, a motion sensor is provided as a sensor for recording the vibration cycle. In both the vibratory compaction unit and the stabilization unit, parameters of the compaction process can be obtained using the motion patterns and force curves during the vibration cycle.

[0014] In the method for operating the system according to the invention, sensor signals for monitoring work units are generated by sensors, wherein the sensor signals are provided to a data acquisition module for separately recording sensor data, and wherein result data is calculated based on the sensor data using a first algorithm set in a computing unit. According to this process sequence, result data is derived from the sensor data while monitoring the work units. Initially, the focus is not on the characteristics or quality of the result data, but rather on using freely definable algorithms through system components specifically provided for this purpose, namely the data acquisition module and the computing unit.

[0015] According to another advantageous improvement of this method, the parameters of the job sequence are calculated as the resulting data and transmitted to the machine controller. In this practical application of the system, the control loop can automatically improve the job sequence executed by the job unit.

[0016] The method described above is improved by adjusting an easily executable algorithm, wherein program data is transmitted to a computing unit for modifying the first algorithm or for setting a second algorithm. This is accomplished either via a VPN tunnel to a computer on which the program data is provided or via a direct connection to the computer on which the program data is provided.

[0017] In this scenario, it is advantageous to load the new program data into the memory of the computing unit in the first step, and to activate the new program data in the second step after restarting the computing unit. This two-step update process ensures that any erroneous program data will not cause system failure. Since the new program is only activated after a restart, the computing unit (processor) is always in a constrained state.

[0018] It is useful to transfer the resulting data from the computing unit to an external computer via a VPN tunnel or an offline connection. Therefore, this data can be used for further processing, either centrally or distributedly, and can be further used and archived in a variety of ways. Attached Figure Description

[0019] The invention will now be illustrated by way of example with reference to the accompanying drawings. The drawings schematically show:

[0020] Figure 1 The track maintenance machine is shown;

[0021] Figure 2 A block diagram of the system is shown;

[0022] Figure 3 The processing of program data is shown; and

[0023] Figure 4 The processing of sensor data and result data is illustrated. Detailed Implementation

[0024] The system includes, for example, a tamping machine, serving as a track maintenance machine 1, for operation on track 2. This track maintenance machine 1 has a tamping unit and a track lifting / shifting unit as operating units 3. Additionally, a stabilizing unit can also be provided as operating unit 3. Operating units 3 are controlled by a machine controller 4. Furthermore, the track maintenance machine 1 includes a measuring system 5 for recording the actual geometry of track 2.

[0025] Sensor 6 is configured to monitor the working unit 3, which is designed as a tamping unit. An exemplary sensor 6 is disclosed in the applicant's Austrian patent application A 290 / 2018. Sensor 6, mounted on the tamping unit or other working unit 3, measures acceleration and / or force acting on individual working unit components. Temperature measurement is also useful for monitoring the condition of the working unit 3.

[0026] The corresponding sensor 6 generates sensor signal S S The sensor signal S S Data is recorded via the Data Acquisition Module (DAQ) 7 and further processed into sensor data S. D To achieve this, the data acquisition module 7 is connected to the computing unit 8. The computing unit 8 contains a first algorithm P1 (program) used to process sensor data S. D Calculate the result data E D The result data E D Used to evaluate the work sequence performed by work unit 3 or to evaluate the condition of track 2 on which work is performed. To achieve this, result data E... D Includes the corresponding parameters.

[0027] Advantageously, the computing unit 8 and the data acquisition module 7 are interconnected in a master-slave structure. The data acquisition module 7 includes, for example, several DAQ units with 12 to 16 channels, wherein each channel is assigned a sensor signal S. S The data acquisition module 7 records the sensor signal S at a high sampling rate in the kilohertz range. S To generate sensor data S with high temporal resolution. D , for subsequent processing.

[0028] However, for purely monitoring functions, low-resolution sensor data S D That's sufficient. Typically, only a small amount of sensor data S is needed per time unit. D (For example, a sampling rate of 1Hz) to track the wear progress of components in the work unit and assess possible maintenance measures. Therefore, for monitoring functions, it is useful to perform separate data processing via a dedicated data acquisition unit 9. The monitoring device 10 includes other components, such as a microprocessor 11 and a modem 12, for transmitting the monitoring data U D Transmitted to a computing network (cloud) 13. The applicant's patent AT 520 698 A1 discloses such a monitoring device 10.

[0029] In addition, the result data E generated by the computing unit 8 is transmitted using the modem 12 of the monitoring device 10 or a separate modem. D It is useful. Therefore, the resulting data E can be centrally used in computer network 13. D And sensor data S, which will also be transmitted when necessary. D For example, data S can be displayed and further processed (network access) via a secure online application (network application) on a computer 14 that is connected to a network. D and E D .

[0030] The track maintenance unit 1 includes, for example, a high-performance Linux server as a computing unit 8. This enables real-time processing of signal data S recorded at a high sampling rate. D In any case, it is useful to adjust the sampling rate of the data acquisition module 7 to match the processing capability of the computing unit 8 to ensure that the result data E is calculated in real time. D Therefore, various characteristic parameters of the work sequence can be determined directly on the track maintenance machine 1.

[0031] Furthermore, it is advantageous that the computing unit 8 is designed such that the CPU's capabilities can also be used to process advanced mathematical algorithms. These mathematical algorithms are models and computational algorithms used to evaluate the state of machine parts and adjust operating parameters. All algorithms set in computing unit 8 are used as tasks T1, T2, T... n (Process) execution. Specifically, the main application M runs on computing unit 8, and computing unit 8 initiates and launches various tasks T1, T2, T3 in a coordinated manner. n ( Figure 3 ).

[0032] Additionally or alternatively, besides transmitting sensor data S D and the results data E DThis data can also be transmitted outside of computer network 13. D E D The data is stored in a storage device 15 connected to the computing unit 8. For example, a dedicated processor (server) is implemented in the computing unit 8, which combines various system variables and processes the requested data S. D E D The data is stored in a large-capacity memory in storage device 15. Alternatively, for example, during calibration of the track maintenance machine 1, the stored data S can be transferred via data interface 16. D E D Transmitted to computer 14.

[0033] exist Figure 2 In the illustrated design, the system includes a communication device 17 for comparing program data with database 18. For example, to achieve this, a VPN router is provided, which is connected to computing unit 8. Thus, sensor data S can also be transmitted via VPN tunnel 19. D and the results data E D .

[0034] Advantageously, VPN tunnel 19 can also be used for software updates of computing unit 8. Figure 3 Therefore, task T was initiated. n Check if a new algorithm is available in database 18. For example, compare it with the current versions of running tasks T1 and T2. If necessary, load the modified algorithm P1 or the new algorithm P2 through VPN tunnel 19, compile the modified algorithm P1 or the new algorithm P2 and append it to task list T. Then, start and process the new task by restarting computing unit 8.

[0035] This update can also be used to analyze previously unnoticed sequences on track maintenance machine 1. First, a new algorithm P2 adapted to the problem definition to be analyzed is loaded into computing unit 8 and compiled. For example, if a specified event occurs, the corresponding task T2 will take sensor data S from some selected sensors 6. D The data is written to storage device 15. After a sufficient period of recording, the collected data S is... D E D Uploaded to computer network 13 and analyzed.

[0036] Figure 4 This demonstrates another improvement to the system. The work unit 3 is monitored by various sensors 6. These sensors 6, along with other sensors 6 arranged on the track maintenance machine 1 (inertial measurement unit, laser cutting sensor, hydraulic gauge, etc.), provide sensor data S to the computing unit 8 via the data acquisition module 7. D With the help of various algorithms P1, P2, Pn The control-related parameters are calculated as the result data E. D The relevant parameters are fed back to the machine controller 4, which then triggers active intervention in the operation process.

[0037] To achieve this purpose, the machine controller 4 (the control system of the track maintenance machine 1) includes a central controller 20, which coordinates several distributed subsystems 21. These subsystems 21 include, for example, a subsystem 21 for adjusting the speed of the vibration drive device that generates vibration, a subsystem 21 for the opening width of the tamping pick of the tamping unit, a subsystem 21 for the automatic penetration system of the tamping pick, and a subsystem 21 for positioning the work unit.

[0038] Therefore, the physical parameters of the affected job sequences are recorded and measured. The recorded parameters are fed as a data stream to computing unit 8, where all tasks T1, T2, and T... n Full access to the sensor data S D In tasks T1, T2, and T... n During execution, characteristic parameters of the task sequence are determined. These parameters are then fed back to the central controller 20 to preset optimized task parameters for the subsystem 21. Thus, a higher-level closed-loop system with observation-based controllers is established at the level of the distributed control system.

[0039] In another beneficial improvement, the optimized job parameters are calculated directly in the calculation unit 8. To achieve this, corresponding algorithms P1, P2, and P3 are set in the calculation unit 8. n The newly calculated operating parameters are specified for the central controller 20. Therefore, the machine controller 4 itself does not perform parameter calculations. Thus, the safety requirements applicable to the machine controller 4 are not affected.

[0040] The following section will explain the specification of the new operating parameters in more detail using the example of multiple tamping via tamping units. In multiple tamping, vibratory tamping picks are lowered into the ballast bed at the same location, and they are repeatedly compacted to improve the compaction of the ballast.

[0041] For parameter optimization, firstly, sensor data S is recorded over a relatively long observation period. D For example, record the pressure and stroke of the tamping unit's compression cylinder. For each recorded tamping cycle, calculate characteristic parameters and use them as basic data for the next step.

[0042] The prediction model can be trained offline using the basic data recorded by this system. Specifically, the recorded data and the corresponding target variable (the number of tamping insertions in each tamping cycle) are used as training data. The trained prediction model corresponds to the new algorithm P2, which is capable of predicting the target variable.

[0043] The new algorithm P2 can be further improved through testing and validation. The test data used is different from the previously used training data. The predictions for the target variable are adjusted to a specified target value to evaluate the quality of the prediction model. If necessary, new training steps are performed on algorithm P2 to improve the prediction quality.

[0044] Using the completed algorithm P2, the corresponding operational parameters (target variables) are directly specified in real time on the track maintenance machine 1. Once the tamping pick penetrates the ballast bed, sensor 6 provides meaningful sensor data S. D This is used to calculate parameters specific to the condition of the ballast bed. In any case, sufficient sensor data S is available at the end of the first tamping insertion. D It can be used to calculate reliable result data E D In this example, the result data E from machine controller 4 D Provide real-time feedback on whether further tamping insertion is necessary at the same location to achieve the desired compaction effect.

[0045] Another advantage of this system is its multi-sleeper tamping unit, in which several tamping units are arranged one after another. These tamping units are lowered together into the ballast bed to tampe several sleepers simultaneously. In this paper, sensor data S is recorded and processed in real time. D This is used for differentiated control of each tamping unit. Specifically, the ballast bed condition determined when the tamping pick penetrates the ballast bed is used to specify different compaction pressures. If necessary, different compaction times are specified for each tamping unit. When tamping several sleepers simultaneously, there is sometimes a problem where the ballast bed, in its initial state, has different compaction levels under each sleeper.

[0046] For each tamping unit, based on the assigned sensor data S D The calculated parameters indicate the corresponding degree of compaction at relevant locations on the ballast bed during the penetration process. Using the corresponding algorithm P2, appropriate compression pressures and, if necessary, compression times are specified for the respective sub-controllers. At locations where compaction has been increased, less tamping energy is introduced into the ballast bed by reducing the compression pressure and time. However, at penetration locations with lower compaction, compression occurs with increasing pressure and duration. Thus, uniform ballast compaction is achieved for ballast bed sections using multi-sleeper tamping units.

Claims

1. A system for operating on a track (2) using a track maintenance machine (1), the system comprising a machine controller (4) and a work unit (3) controlled by the machine controller (4), wherein sensors (6) are provided for monitoring the work unit (3), characterized in that, These sensors (6) are coupled to a device for individually recording sensor data (S D The data acquisition module (7) is connected to the computing unit (8). The computing unit (8) is equipped with a data acquisition module (7) for processing the sensor data (S). D ) Calculate the result data (E) D The first algorithm (P1) includes a monitoring device (10) for monitoring the work unit (3), the monitoring device (10) recording the sensor data (S) at a lower sampling rate than the data acquisition module (7). D ).

2. The system according to claim 1, characterized in that, The computing unit (8) is configured to calculate the sensor data (S) recorded in the job sequence based on the data. D The calculation unit (8) calculates at least one parameter, wherein the calculation unit (8) is specifically coupled to the machine controller (4) to automatically specify optimized operating parameters.

3. The system according to claim 1 or 2, characterized in that, The data acquisition module (7) is configured for multi-channel data recording and is coupled as a slave device to the computing unit (8) which is designed as an active device.

4. The system according to claim 1 or 2, characterized in that, The computing unit (8) is coupled to the database (18) via a communication device (17) to receive program data for modifying the first algorithm (P1) or for setting the second algorithm (P2).

5. The system according to claim 4, characterized in that, The communication device (17) includes a VPN router.

6. The system according to claim 1 or 2, characterized in that, The computing unit (8) is connected to the storage device (15), which is used to store sensor data (S). D ) and / or outcome data (E D ).

7. The system according to claim 1 or 2, characterized in that, The computing unit (8) is coupled to the computer network (13) via a modem (12) to transmit data.

8. The system according to claim 1 or 2, characterized in that, The system also includes a tamping unit and / or a stabilizing unit as a working unit (3).

9. The system according to claim 7, characterized in that, A motion sensor is provided as a sensor for recording the vibration period (6).

10. A method for operating the system according to any one of claims 1 to 9, wherein, Sensor signals (S) for monitoring the work unit (3) are generated by the sensor (6). S The characteristic of this invention is that the sensor signal (S) is... S The data is provided to the data acquisition module (7) for separately recording sensor data, and the data is processed by the first algorithm (P1) set in the computing unit (8) based on the sensor data (S). D ) Calculate the result data (E) D ).

11. The method according to claim 10, characterized in that, The parameters of the job sequence are calculated into the result data (E). D And transmit it to the machine controller (4).

12. The method according to claim 10 or 11, characterized in that, The program data is transmitted to the computing unit (8) for modifying the first algorithm (P1) or for setting the second algorithm (P2).

13. The method according to claim 12, characterized in that, In the first step, new program data is loaded into the memory of the computing unit (8), and in the second step, the new program data is activated after the computing unit (8) is restarted.

14. The method according to claim 10 or 11, characterized in that, The results data (E) can be transmitted via VPN tunnel or via offline connection. D The data is transmitted from the computing unit (8) to the external computer (14).

Images