Method and system for monitoring structural health of railway track

Abstract

Disclosed is a method (100) for monitoring structural health of a railway track (900), the method (100) comprising: measuring mechanical strain values of a train undercarriage (302) when the train (400) is on the railway track; measuring location coordinates of the train when the train is on the railway track; synchronizing the measured mechanical strain values with the measured location coordinates of the train; comparing the synchronized mechanical strain values with a reference dataset, where the reference dataset comprises a lower threshold limit and an upper threshold limit of mechanical strain for a plurality of different location coordinates of the train; generating a signal if the synchronized mechanical strain values exceed either the lower threshold limit or the upper threshold limit of mechanical strain and providing the generated signal to an operator (216) of the train for monitoring the structural health of the railway track.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to a method for monitoring structural health of a railway track. Moreover, the present disclosure relates to a system for monitoring structural health of a railway track.BACKGROUND

[0002] Nowadays, the analysis of railroad conditions relies heavily on specialized trains, which are purposefully-built for analyzing the railroad conditions.A Typically, only one such specialized train is employed per country due to its exorbitant cost (e.g., 30 to 40 million Euros per unit). The railway tracks can develop various defects over the time, such as corrugation, skip spots, and the like. If such kind of defects go undetected, it may lead to severe consequences, such as derailment. Conventionally, few methods have been developed for detecting the defects in the railway tracks, such as measuring carts, optical and laser-based methods and vibration sensors. Each of aforementioned methods have certain limitations, such as the measuring carts are expensive, slow and cumbersome because of consisting of an entire train cart. The optical methods can primarily detect debris and large obstacles around the track. Moreover, the vibration sensors are unable to detect certain faults, such as sinkage of a part of the railway track and skid spots.

[0003] Thus, there exists a technical problem of how to efficiently and reliably detect the defects in railroad tracks and provide a cost-effective solution for an enhanced safety while using the railway tracks.

[0004] Therefore, in light of the foregoing discussion, there exists a need to overcome the aforementioned drawbacks associated with the conventional ways of monitoring the condition of railway tracks as well as of the train.SUMMARY

[0005] The aim of the present disclosure is to provide a method and a system for monitoring structural health of a railway track. The aim of the present disclosure is achieved by a method and a system for monitoring structural health of a railway track that provides an efficient, more reliable and cost-effective solution for monitoring the condition of railway tracks. The disclosed method and system are based on monitoring fatigue of a mechanical structure, specifically emphasizing on the measurement of mechanical strain values that indicate changes, such as jumps, inclination and shaking of the railway tracks as defined in the appended independent claims to which reference is made to. Advantageous features are set out in the appended dependent claims.

[0006] Embodiments of the present disclosure substantially eliminate or at least partially address the aforementioned problems in the prior art, and enable an efficient and reliable monitoring of the railway tracks by measuring mechanical strain values of a train undercarriage as well as location coordinates of the train and providing a signal if the measured mechanical strain values start drifting from predetermined reference values of mechanical strain.

[0007] Throughout the description and claims of this specification, the words "comprise", "include", "have", and "contain" and variations of these words, for example "comprising" and "comprises", mean "including but not limited to", and do not exclude other components, items, integers or steps not explicitly disclosed also to be present. Moreover, the singular encompasses the plural unless the context otherwise requires. In particular, where the indefinite article is used, the specification is to be understood as contemplating plurality as well as singularity, unless the context requires otherwise.BRIEF DESCRIPTION OF THE DRAWINGS

[0008] FIG. 1 is a flowchart of a method for monitoring structural health of a railway track, in accordance with an embodiment of the present disclosure; FIG. 2 is a block diagram that illustrates various exemplary components of a system for monitoring structural health of a railway track, in accordance with an embodiment of the present disclosure; FIG. 3 illustrates position of a mechanical strain-gauge sensor module located on a train undercarriage, in accordance with an embodiment of the present disclosure; FIG. 4 illustrates installation of one or more mechanical strain-gauge sensor modules in different sections of a train, in accordance with an embodiment of the present disclosure; FIG. 5 illustrates a graphical representation of data compression, in accordance with an embodiment of the present disclosure; FIG. 6 illustrates a graphical representation of generating a signal if synchronized mechanical strain values exceed either a lower threshold limit or an upper threshold limit of mechanical strain at a given location coordinates of a train, in accordance with an embodiment of the present disclosure; FIG. 7A illustrates a window for measured mechanical strain values and threshold level limits, in accordance with an embodiment of the present disclosure; FIG. 7B illustrates a window for measured mechanical strain values and threshold level limits, in accordance with another embodiment of the present disclosure; FIG. 8 illustrates an exemplary user interface that represents data points on a railroad map, in accordance with another embodiment of the present disclosure; FIG. 9 illustrates a railway track having different types of soil in different areas, in accordance with an embodiment of the present disclosure; and FIG. 10 illustrates a graphical representation that represents generation of an alert signal, in accordance with another embodiment of the present disclosure. DETAILED DESCRIPTION OF EMBODIMENTS

[0009] The following detailed description illustrates embodiments of the present disclosure and ways in which they can be implemented. Although some modes of carrying out the present disclosure have been disclosed, those skilled in the art would recognize that other embodiments for carrying out or practising the present disclosure are also possible.

[0010] In a first aspect, the present disclosure provides a method for monitoring structural health of a railway track, the method comprising: measuring mechanical strain values of a train undercarriage by using at least one strain gauge configured for measuring mechanical strain when the train is on the railway track; measuring location coordinates of the train when the train is on the railway track; synchronizing the measured mechanical strain values with the measured location coordinates of the train; comparing the synchronized mechanical strain values with a reference dataset, wherein the reference dataset comprises a lower threshold limit and an upper threshold limit of mechanical strain for a plurality of different location coordinates of the train; generating a signal if the synchronized mechanical strain values exceed either the lower threshold limit or the upper threshold limit of mechanical strain at a given location coordinates of the train; and providing the generated signal to an operator of the train for monitoring the structural health of the railway track

[0011] In a second aspect, the present disclosure provides a system for monitoring structural health of a railway track, the system comprises: a mechanical strain-gauge sensor module comprising one or more mechanical strain-gauge sensors configured to measure mechanical strain values of a train undercarriage; a Global Positioning System receiver configured to measure location coordinates of the train on the railway track; and a computing device configured to: synchronize the measured mechanical strain values with the measured location coordinates of the train; compare the synchronized mechanical strain values with a reference dataset, wherein the reference dataset comprises a lower threshold limit and an upper threshold limit of mechanical strain for a plurality of different location coordinates of the train; generate a signal if the synchronized mechanical strain values exceed either the lower threshold limit or the upper threshold limit of mechanical strain at a given location coordinates of the train; and provide the generated signal to an operator of the train for monitoring the structural health of the railway track.

[0012] The present disclosure provides the aforementioned method and the aforementioned system for an efficient and reliable structural monitoring of the railway track. Also, the aforementioned method and the aforementioned system provide a cost-effective solution for the structural monitoring of the railway track. The aforementioned method and the aforementioned system explicitly disclose that the GPS receiver and the mechanical strain-gauge sensor module are mounted either on the same train or in different bogies of the same train or in different trains. Furthermore, the aforementioned method and the aforementioned system include synchronizing the measured mechanical strain values with the measured location coordinates of the train before storing the measured data into a memory of the aforementioned system. That's why the aforementioned method and the aforementioned system enable a reliable monitoring of the railway track. In contrast to the conventional methods of monitoring the structural health of the railway track, the aforementioned method and the aforementioned system generate the signal when the synchronized mechanical strain values exceed either the lower threshold limit or the upper threshold limit of mechanical strain at the given location coordinates of the train and provide the generated signal either to the operator of the train or a railway maintenance organization for a corresponding action required for maintaining the structural health of the railway track. Thus, the aforementioned method and the aforementioned system provides more efficient, reliable and fast monitoring of structural health of the railway track and ensures an enhanced safety when the train moves on the railway track.

[0013] Throughout the present disclosure, the term "monitoring structural health of a railway track" refers to a process or a technique that involves the continuous assessment and analysis of various conditions and integrity of infrastructure of the railway track. The structural health monitoring of the railway track includes the monitoring of key parameters, such as mechanical strain, vibrations and deformations in real time or near real time. The monitored parameters are used to detect and identify potential defects or anomalies that may compromise the safety and operational efficiency of the railway track.

[0014] Throughout the present disclosure, the term "system" refers to a specialized equipment that is configured for monitoring structural health of a railway track. It will be appreciated that the term "system" comprises a mechanical strain-gauge sensor module, a GPS receiver and a computing device communicably coupled to each of the mechanical strain-gauge sensor module and the GPS receiver. The system comprising the mechanical strain-gauge sensor module, the GPS receiver and the computing device is installed on a train. The term "mechanical strain-gauge sensor module" refers to a device designed to measure mechanical stress and deformations in structures, such as the railway tracks. The mechanical strain-gauge sensor module comprises one or more mechanical strain-gauge sensors, which are sensitive to changes in strain and therefore, detect alterations in shape or mechanical loading of the railway track affecting to mechanical strains of the train undercarriage. The alterations in shape or mechanical loading of the railway track may happen slowly due to an erosion or a frost of the soil below the railway track, or fast due to accidents or if e.g. a railway engine hums in a one place. Thus, the mechanical strain-gauge sensor module contributes in assessment of the structural health of the railway track and supports the identification of potential defects or irregularities of the railway track.

[0015] The term "GPS receiver" refers to a device that uses signals from satellites to determine an accurate and real-time geographical location on earth.

[0016] The term "computing device" refers to a device that performs computations on the measured mechanical strain values and the measured location coordinates of the train. Examples of implementation of the computing device may include but are not limited to, a central processing unit (CPU), an edge computing device, a processor, a coprocessor, a microprocessor, a microcontroller, a complex instruction set computing (CISC) processor, an application-specific integrated circuit (ASIC) processor, a reduced instruction set (RISC) processor, an integrated circuit, a very long instruction word (VLIW) processor, a state machine, a data processing unit, and other processors or circuits. Moreover, the computing device, may refer to one or more individual processors, processing devices, a processing unit that is part of a machine.

[0017] In operation, the method comprises measuring mechanical strain values of a train undercarriage by using at least one strain gauge configured for measuring mechanical strain when the train is on the railway track. The mechanical strain values of the train undercarriage are measured using various sensors, such as mechanical strain gauge sensors. The use of the mechanical strain gauge sensors quantifies the deformation or stress experienced by various components of the train's undercarriage (that is a supporting frame under the body of the train for example, a wheeled structure). Alternatively stated, any stress or deformation experienced by the wheeled structure of the train is measured when the train is on the railway track. Alternatively, the method comprises a loop: measure mechanical strain, save measurement results, calibrate, measure mechanical strain, etc. when measuring mechanical strain values. The measurement of the mechanical strain values of the train undercarriage supports in monitoring and assessment of structural integrity of the railway track. Optionally the mechanical strain values are measured on the frame of the train or the cart.

[0018] Optionally, the mechanical strain is measured when the train is moving. In an implementation, the mechanical strain values of the train undercarriage are measured when the train is moving on the railway track.

[0019] The method further comprises measuring location coordinates of the train when the train is on the railway track and synchronizing the measured mechanical strain values with the measured location coordinates of the train. In addition to the mechanical strain values of the train undercarriage, the location coordinates (i.e., geographical coordinates) of the train are also measured. The measurement of the location coordinates includes measurement of latitude and longitude coordinates which, are used to determine the horizontal position of the train on Earth's surface. Furthermore, the measured mechanical strain values of the train undercarriage and the measured location coordinates of the train are synchronized before storing at a storage space (e.g., a cloud sever) in order to collect statistical force data (i.e., the statistical data of mechanical strain values for a provided location).

[0020] Optionally, synchronizing the measured mechanical strain values with the measured location coordinates of the train comprises matching time stamps of the measured mechanical strain values with the measured location coordinates of the train. The measured mechanical strain values of the train undercarriage are synchronized with the measured location coordinates of the train by matching the time stamps of the measured mechanical strain values with the time stamps of the measured location coordinates of the train.

[0021] The method further comprises comparing the synchronized mechanical strain values with a reference dataset, wherein the reference dataset comprises a lower threshold limit and an upper threshold limit of mechanical strain for a plurality of different location coordinates of the train. Furthermore, the synchronized mechanical strain values of the train undercarriage and the location coordinates of the train are compared with the reference dataset. The reference dataset comprises the lower threshold limit and the upper threshold limit of the mechanical strain for different location coordinates of the train.

[0022] Optionally, the method further comprises storing the synchronized mechanical strain values to collect a dataset of mechanical strain values, computing the lower threshold limit and the upper threshold limit of mechanical strain for the plurality of different location coordinates of the train based on the dataset of mechanical strain values and storing the lower threshold limit and the upper threshold limit of mechanical strain for the plurality of different location coordinates of the train. The synchronized mechanical strain values of the train undercarriage and the location coordinates of the train are stored in a memory to collect the dataset of mechanical strain values. The collected dataset may also be referred to as a raw dataset and it differs from the reference dataset. The collected dataset can be a dataset having either dynamical values of mechanical strain or statistical values of mechanical strain or historical values of mechanical strain or a combination of dynamical values, statistical values and historical values of mechanical strain. Thereafter, the collected dataset is used to compute the lower threshold limit and the upper threshold limit of mechanical strain for different location coordinates of the train. The computed lower threshold limit and the upper threshold limit of mechanical strain for different location coordinates is stored, for example, either in a memory of the disclosed system or in a cloud server wirelessly connected to the disclosed system.

[0023] Optionally, the reference dataset is updated based on measurements. The reference dataset that is the lower threshold limit and the upper threshold limit of mechanical strain are updated time-to-time (e.g., over a time period of 6 months or 1 year and may be more) based on the mechanical strain values of the train undercarriage and the location coordinates of the train measured over a specific time period. Moreover, the collected dataset (i.e., the raw dataset) is also used to update the reference dataset (i.e., the lower threshold limit and the upper threshold limit of mechanical strain for different location coordinates of the train). Optionally, the collected dataset (i.e., the raw dataset) is used to update the reference dataset only when measured datapoints are within the lower threshold limit and the upper threshold limit i.e., when there is a fault situation on the railway track the measured data is not used to update the reference dataset.

[0024] Optionally, the method further comprises storing the synchronized mechanical strain values in a cloud server after a pre-processing of the synchronized mechanical strain values, where the stored synchronized mechanical strain values are used for subsequent analysis. The synchronized mechanical strain values and location coordinates of the train may be stored in the cloud server for further analysis of the synchronized mechanical strain values.

[0025] Throughout the present disclosure, the term "cloud server" refers to a device used to store the synchronized mechanical strain values and the location coordinates of the train. Examples of the cloud server may include, but are not limited to, an application server, a storage server, or a combination thereof. Moreover, the cloud server may either be a single hardware server or a plurality of hardware servers operating in a parallel or distributed architecture to store the synchronized mechanical strain values and the location coordinates of the train and perform the subsequent analysis on the synchronized data.

[0026] At the cloud server, the synchronized mechanical strain values may be further processed and analyzed and may be displayed on a plurality of electronic devices, such as mobile phone, laptop, desktop computer, smart phone, and the like, by use of a web application. Before storing the synchronized mechanical strain values in the cloud server, the synchronized mechanical strain values are subjected to the pre-processing. The pre-processing includes filtration of noise (i.e., any noise present in the synchronized mechanical strain values), decimation of the synchronized mechanical strain values, and the like, while decimation can be used for example depending on the route, speed of train, or any other variables. The decimation refers to a process of reducing sampling rate of the synchronized mechanical strain values by removing or discarding few strain values from the synchronized mechanical strain values, while preserving the information required for an intended application.

[0027] Optionally, the pre-processing comprises applying data compression on the synchronized mechanical strain values to obtain a dataset of mechanical strain values with a comparatively smaller dimension. The synchronized mechanical strain values are simplified by use of the pre-processing. The pre-processing may include applying the data compression on the synchronized mechanical strain values to obtain the dataset of comparatively smaller dimensions, which is further stored in the cloud server. This is advantageous to obtain the dataset of smaller dimensions because sending the dataset of smaller dimensions to the cloud server results in bandwidth saving. The bandwidth saving can be enhanced by sending the dataset of smaller dimensions less frequently to the cloud server, for example, at each railway station. In a case, if there is no network connection to the cloud server then in such case, the dataset can be temporarily saved and shared with the cloud server after some time, for example, at next railway station where network connection is present. Furthermore, the data compression can be applied on the synchronized mechanical strain values by using any one of: a peak valley algorithm, a digital low pass filter, an audio compression algorithm, and skipping some of data points from the synchronized mechanical strain values. The peak valley algorithm identifies points in the sequence where the value is significantly higher (a peak) or lower (a valley) than its neighboring values. Minimum and maximum turning points are also identified making it possible to pair peaks and valleys together making it possible various analysis of the measured mechanical strain values including e.g. calculating the amplitudes of peaks and valleys, determining the frequency of occurrence of peaks and valleys, measuring the distance between peak-valley pairs, etc. The audio compression algorithm may include lossy and lossless compression algorithms, such as MPEG-1 Audio Layer III (MP3), Advanced Audio Coding (AAC), Free Lossless Audio Codec (FLAC), OGG Vorbis, WAVEPACK algorithm, Apple Lossless Audio Codec (ALAC), Monkey's Audio, Opus, Sub-Band Coding (SBC), Musical Instrument Digital Interface (MIDI), Adaptive Differential Pulse-Code Modulation (ADPCM), Linear Predictive Coding (LPC), and the like. Some data points may be skipped in an event of long stretches (i.e., extended or prolonged periods of time) where no new insights can be made from the measured data. For example, if the strain data hovers for 10,000 consecutive measurements, then approximately 10 values can be sent to the cloud server for storage and rest values can be skipped or discarded, resulting in a significant bandwidth saving.

[0028] Optionally, computing the reference dataset comprises measuring a first set of mechanical strain values on a first portion of the railway track of length L x by running a number of laps of the train through the first portion of the railway track until consistent boundary values are obtained, wherein a state of the first portion of the railway track is known to be in a healthy state, storing the first set of mechanical strain values measured on the first portion of the railway track of length L x and utilizing the first set of mechanical strain values to determine the lower threshold limit and the upper threshold limit of mechanical strain for the first portion of the railway track to determine a state of the track based on a behavior of the train undercarriage when the train is on the track at the first portion of the railway track of length L x . The first set of mechanical strain values is measured on the first portion of the railway track of Length L x (e.g., 1 to 1000 km) by running X (e.g., 50, 100 or more) number of laps through the first portion until the consistent boundary values are obtained. The first portion of the railway track is known to be in the healthy state. Alternatively stated, the state of the first portion of the railway track is already verified either through inspection (e.g., manual inspection) or by other means. The measured first set of mechanical strain values is used to determine the lower threshold limit and the upper threshold limit of mechanical strain for the first portion of the railway track of length L x . Thereafter, the determined lower threshold limit and the upper threshold limit of mechanical strain are used to determine the state of the track based on the behavior of the train undercarriage when the train moves on the track at the first portion of the railway track of length L x . The state of the track is determined as the healthy state. The data similar to the first set of mechanical strain values is collected for each track separately in a country or such area by first running a known number of laps through a portion of each track until the consistent boundary values can be obtained. The same data can be used to calibrate other trains running the same track. This way, the reference data comprises the lower threshold limit and the upper threshold limit of mechanical strain for the railway tracks which are considered in the healthy condition.

[0029] Optionally, computing the reference dataset comprises measuring a second set of mechanical strain values on a second portion of the railway track of length L x+y by running a number of laps of the train through the second portion of the railway track, wherein a state of the second portion of the railway track is unknown, storing the second set of mechanical strain values measured on the second portion of the railway track and utilizing the second set of mechanical strain values to determine the lower threshold limit and the upper threshold limit of mechanical strain for the second portion of the railway track. The second set of mechanical strain values is measured on the second portion of the railway track of Length L x+y (e.g., 1000 to 2000 km) by running multiple number of laps through the second portion of the railway track. The state of the second portion of the railway track is unknown which means the second portion of the railway track may not be in a useful condition. The second set of measured mechanical strain values is stored and used to determine the lower threshold limit and the upper threshold limit of mechanical strain for the second portion of the railway track. Thereafter, the determined lower threshold limit and the upper threshold limit of mechanical strain are used to determine the state of the track based on the behavior of the train undercarriage when the train moves on the track at the second portion of the railway track of length L x+y . This way, the reference data comprises the lower threshold limit and the upper threshold limit of mechanical strain for the railway tracks whose state is unknown but can be determined by use of the determined lower threshold limit and the upper threshold limit of mechanical strain for the second portion of the railway track.

[0030] Optionally, measuring the second set of mechanical strain values on the second portion of the railway track comprises measuring a first subset of mechanical strain values from at least two different locations having a first type of soil and a second subset of mechanical strain values from a location having a second type of soil. For example, the first subset of mechanical strain values is measured from the soil having stones (i.e., the first type of soil) and the second subset of mechanical strain values is measured from the soil having sand (i.e., the second type of soil). This means that the soil is different in terms of technical characteristics at different locations and does affect the measurement such that the first subset of mechanical strain values measured from the location having the first type of soil (i.e., stone soil) is evidently different from the second subset of mechanical strain values measured from the location having the second type of soil (i.e., sand soil).

[0031] Optionally, the mechanical strain is measured. The measured points are added in a 180-degree phase-shift compared to reference data and when the data differs from 0-point, then a signal is sent. In normal track condition the sum-effect of the measured data and 180-degree phase-shifted reference data the difference should be near to zero (0). This means that when there is a difference the difference is only sent to cloud for further processing saving data capacity and processing power.

[0032] Optionally, a ground frost of the soil is used to define at least one of the following: soil modulus of elasticity, soil hardness. The ground frost data is collected to define the soil modulus of elasticity. For example, in winters, a sand soil may look like a stone. The ground frost data can be obtained from public data sources or performing multiple measurements with different soil conditions (i.e., measuring during different seasons). Temperatures and how wet the soil is can give information about how much the soil characteristics varies. From this it might be possible to estimate ground frost (depth). The ground frost data can also be used in calibrating the mechanical strain values because there are different soil modulus of elasticity in different ground frost areas. Based on the measured mechanical strain values over the ground frost of the soil, the change of the ground frost can be estimated too. When the ground frost is at the most (e.g., after a long cold winter), the reference data can be collected in a much easier way because the ground soil "error" can be eliminated. Optionally the reference data is collected separately based on a time of the year, month, a ground frost, a vehicle model.

[0033] Optionally, the mechanical strain is measured when the train is moving at two different speeds (v1, v2) at the first portion of the railway track of length L x or at the second portion of the railway track of length L x+y . For example, the first set of mechanical strain values includes the mechanical strain values measured when the train is moving at speed v1 and also, the mechanical strain values measured when the train is moving at speed v2 at the first portion of the railway track of length L x . Similarly, the second set of mechanical strain values includes the mechanical strain values measured when the train is moving at speed v1 and also, the mechanical strain values measured when the train is moving at speed v2 at the second portion of the railway track of length L x+y .

[0034] Optionally, the mechanical strain measurements performed at the two different speeds (v1, v2) of the moving train are compared to each other. For example, the mechanical strain values measured when the train is moving at speed v1 and the mechanical strain values measured when the train is moving at speed v2, are compared with each other.

[0035] The method further comprises generating a signal if the synchronized mechanical strain values exceed either the lower threshold limit or the upper threshold limit of mechanical strain at a given location coordinates of the train. For example, if the measured mechanical strain values are outside of the lower threshold limit, the signal is generated to slow down the speed of the train. If the measured mechanical strain values are outside of the upper threshold limit, the signal is generated to stop the train. If the measured mechanical strain values are within the lower threshold limit and the upper threshold limit, the signal is generated either to speed up the train or maintain the speed of the train.

[0036] Optionally, generating the signal comprises generating a first signal if the first subset of mechanical strain values measured from the at least two different locations having the first type of soil on the second portion of the railway track exceeds the lower threshold limit of mechanical strain; and generating a second signal if the second subset of mechanical strain values measured from the location having the second type of soil on the second portion of the railway track exceeds the upper threshold limit of mechanical strain. For example, the first signal can be generated like a warning signal to reduce the speed of the train when the first subset of mechanical strain values exceeds the lower threshold limit of mechanical strain. And the second signal can be generated like an emergency signal to stop the train when the second subset of mechanical strain values exceeds the upper threshold limit of mechanical strain.

[0037] The method further comprises providing the generated signal to an operator of the train for monitoring the structural health of the railway track. The generated signal can be displayed to the operator of the train by use of a color map having a number of data points in different colors (for example, green, red and yellow color). For example, the generated signal can be in the form of a data point having a red color on the color map which, refers to an alert situation on a rail track map and used to alarm the operator of the train in a case when a rail track condition is getting worse or operation of the train is unsafe. The operator can be either a train driver or an operator outside of the train for example, for monitoring the condition of the railway track. Mostly, the operator is outside of the train for monitoring the condition of the railway track.

[0038] Optionally, a corresponding action for maintaining structural health of the railway track is performed by utilizing the generated signal. Based on the generated signal, the corresponding action (e.g., remanufacture or repair) can be performed to maintain the structural health of the railway track.

[0039] The present disclosure also relates to the system as described above. Various embodiments and variants disclosed above, with respect to the aforementioned method, apply mutatis mutandis to the system.

[0040] Optionally, the one or more mechanical strain-gauge sensors of the mechanical strain-gauge sensor module are installed in different sections of the train. In an implementation, the one or more strain-gauge sensors can be installed on one tandem axle, for example, sensor a on left side and sensor b on right side, to measure strain values on both sides of the railway track, individually. This is advantageous to install the one or more mechanical strain-gauge sensors in different sections of the train to measure data (i.e., the mechanical strain values) with more precision and to eliminate the false measurement results in a simple and very fast way. The installation of the one or more mechanical strain-gauge sensors in different sections of the train lead to multiplication of the measured data with the number of strain-gauge sensors resulting in faster data collection. For example, in one case, three strain-gauge sensors can be installed in three different bogies of same train. For instance, a first mechanical strain-gauge sensor can be installed in a bogie lying at the beginning of the train, a second mechanical strain-gauge sensor can be installed in a bogie lying at the middle of the train and a third mechanical strain-gauge sensor can be installed in a bogie lying at the end of the train. In another case, the three strain-gauge sensors can be installed in three different trains.

[0041] Optionally, the one or more mechanical strain-gauge sensors of the mechanical strain-gauge sensor module and the GPS receiver are calibrated to a pre-determined starting point of the train before starting measurement of the mechanical strain values and the location coordinates of the train, respectively. The one or more mechanical strain-gauge sensors of the mechanical strain-gauge sensor module are calibrated to a certain and the pre-determined starting point before starting the measurement of the mechanical strain values. The pre-determined starting point is then used as a reference point for all the mechanical strain values collected over the railway track. The GPS receiver is also calibrated to ensure that that GPS receiver displays the location correctly before the measurement of data begins. Moreover, the system may have an automatic check within a software that accounts for moving of a neutral area (may also be referred as a 0-line) and re-evaluates the different zones according to new position of the neutral area. This is done by checking the data over a long period of time (e.g., 200 km long part of track) and adjusting the neutral area if the data is shifted from the expected neutral area.

[0042] Optionally, the system has a wireless connection to a cloud server configured to store the synchronized mechanical strain values after a pre-processing, wherein the stored synchronized mechanical strain values are used for subsequent analysis. For example, the system may be connected to the cloud server through a Local Area Network (LAN) connection.

[0043] Optionally, the pre-processing comprises applying data compression on the synchronized mechanical strain values to obtain a dataset of mechanical strain values with a comparatively smaller dimension than the reference dataset.

[0044] Optionally, the system further comprises an inclinometer sensor to detect a difference between elevation of different sides of the railway track. For example, left rail of the railway track may be higher than right rail of the railway track. In addition to the inclinometer sensor, other sensors, such as Inertial Measurement Unit (IMU), gyroscope or accelerometer may also be used in the system. The system may include an Artificial Intelligence (AI) model to estimate mechanical strain values using previously measured mechanical strain values for different location coordinates of the train. The AI model may also be used to estimate the mechanical strain values for the railway track having different soil types based on the mechanical strain values measured for similar soil types.

[0045] Optionally, the system further works as follows: Strain gauges are placed in the undercarriage of the train together with a GPS tracker (and other devices) to measure the strains (and other possible data) occurring in the structures of the train and to combine it with the location data of where the strains occurred into a pinpoint the location of the possible defect on the track. The signal (measurement data) omitted from the strain gauges, typically through analogue to digital conversion unit, is then transmitted to a CPU and later sent to a cloud service or to an external user. Optionally the raw data can be sent within a predetermined time, e.g. within 5 seconds, to a cloud service or to an external user in a case of alert signal caused by a possible defect in the track is detected. These are used to store data as well as conduct different analyses to detect threshold passing values and to detect several different types of faults. After detecting a possible defect in the track an appropriate signal (alert) is produced to inform the correct authorities about the abnormality in the data.DETAILED DESCRIPTION OF THE DRAWINGS

[0046] Referring to FIG. 1, illustrates steps of a method 100 for monitoring structural health of a railway track, in accordance with an embodiment of the present disclosure. At step 102, the method 100 includes measuring mechanical strain values of a train undercarriage by using at least one strain gauge configured for measuring mechanical strain when the train is on the railway track. At step 104, the method 100 includes measuring location coordinates of the train when the train is on the railway track. At step 106, the method 100 includes synchronizing the measured mechanical strain values with the measured location coordinates of the train. At step 108, the method 100 includes comparing the synchronized mechanical strain values with a reference dataset, where the reference dataset comprises a lower threshold limit and an upper threshold limit of mechanical strain for a plurality of different location coordinates of the train. At step 110, the method 100 includes generating a signal if the synchronized mechanical strain values exceed either the lower threshold limit or the upper threshold limit of mechanical strain at a given location coordinates of the train. At step 112, the method 100 includes providing the generated signal to an operator of the train for monitoring the structural health of the railway track.

[0047] The aforementioned steps are only illustrative and other alternatives can also be provided where one or more steps are added, one or more steps are removed, or one or more steps are provided in a different sequence without departing from the scope of the claims herein.

[0048] Referring to FIG. 2, illustrates a block diagram of a system 200 installed in a train for monitoring structural health of a railway track, in accordance with an embodiment of the present disclosure. The system 200 includes a mechanical strain-gauge sensor module 202 comprising one or more mechanical strain-gauge sensors 202A, a GPS receiver 204 and a computing device 206. The computing device 206 is further connected to a cloud server 208 through a wireless connection, e.g., an internet connection (represented by dotted lines in FIG. 2) and to a display device 214 through a local wired connection (represented by a solid line in FIG. 2). The mechanical strain-gauge sensor module 202 comprising the one or more mechanical strain-gauge sensors 202A is configured to measure mechanical strain values of a train undercarriage. The GPS receiver 204 is configured to measure location coordinates of the train when the train is on the railway track. The computing device 206 is configured to receive the measured mechanical strain values of the train undercarriage from the mechanical strain-gauge sensor module 202 and the measured location coordinates of the train from the GPS receiver 204. The computing device 206 may include a processor, a program execution memory and a data memory (not shown for sake of brevity). The measured mechanical strain values and the measured location coordinates of the train are processed by a program stored in the program execution memory, which is executed by the microprocessor. The computing device 206 is configured to synchronize the measured mechanical strain values with the measured location coordinates of the train. The measured mechanical strain values and the measured location coordinates of the train are pre-processed up to the extent that the synchronized mechanical strain values which are stored at the cloud server 208 have smaller dimensions. The pre-processing of the synchronized mechanical strain values is required because the synchronized mechanical strain values, which are transferred to the cloud server 208 are used for further analysis and the objective is to avoid sending the redundant data for bandwidth saving. The computing device 206 is configured to compare the synchronized mechanical strain values with a reference dataset, where the reference dataset comprises a lower threshold limit and an upper threshold limit of mechanical strain for a plurality of different location coordinates of the train. The computing device 206 is configured to generate a signal if the synchronized mechanical strain values exceed either the lower threshold limit or the upper threshold limit of mechanical strain at a given location coordinates of the train. The computing device 206 is configured to provide the generated signal to an operator 216 of the train for monitoring the structural health of the railway track. The generated signal can be made visible to the operator 216 of the train by displaying on the display device 214 (e.g., a Liquid Crystal Display (LCD) display device, Light Emitting Diode (LED) display device, a display screen, and the like) connected to the computing device 206 through the local wired or wireless connection. Alternatively, the generated signal can be analyzed at the cloud server 208 and displayed on a plurality of electronic devices 212 (e.g., a smart phone, a laptop, a desktop computer, and the like) through a web application 210.

[0049] Referring to FIG. 3, illustrates position of the mechanical strain-gauge sensor module 202 located on a train undercarriage 302, in accordance with an embodiment of the present disclosure.

[0050] Referring to FIG. 4, illustrates installation of one or more mechanical strain-gauge sensor module in different sections of a train 400, in accordance with an embodiment of the present disclosure. A first mechanical strain-gauge sensor module 402A is installed in a bogie lying at the beginning of the train 400, a second mechanical strain-gauge sensor module 402B is installed in a bogie lying at the middle of the train 400, and a third mechanical strain-gauge sensor module 402C is installed in a bogie lying at the end of the train 400. Each of the first mechanical strain-gauge sensor module 402A, the second mechanical strain-gauge sensor module 402B and the third mechanical strain-gauge sensor module 402C comprises one or more mechanical strain-gauge sensors to measure mechanical strain values of the train 400 undercarriage. In FIG. 4, the first mechanical strain-gauge sensor module 402A, the second mechanical strain-gauge sensor module 402B and the third mechanical strain-gauge sensor module 402C is installed in different bogies of the same train (i.e., the train 400 ). In another implementation, the first mechanical strain-gauge sensor module 402A, the second mechanical strain-gauge sensor module 402B and the third mechanical strain-gauge sensor module 402C can be installed in different trains. Only, three mechanical strain-gauge sensor modules are shown in FIG. 4 which, are installed in the train 400 (for sake of brevity). Although, more than three mechanical strain-gauge sensor modules can be installed in the train 400.

[0051] Referring to FIG. 5, illustrates a graphical representation 500 of data compression, in accordance with an embodiment of the present disclosure. With reference to the graphical representation 500, there is shown a X-axis 502 that represents distance travelled by a train on a railway track and a Y-axis 504 that represents mechanical strain values. There is further shown a first horizontal line 506 that represents a reference line (shown as a dotted line in the graphical representation 500). A plurality of second horizontal lines 508 (i.e., above and below the reference line, shown as solid lines) represents a first threshold level (i.e., lower threshold limit of mechanical strain) to notify to change the driving style. A plurality of third horizontal lines 510 represents a second threshold level (i.e., upper threshold limit of mechanical strain) to notify to send an alert to check the condition of the railway track. A fourth line 512 represents larger number of data points (i.e., the mechanical strain values) which, are not sent in order to compress the data. Only the data points (e.g., compressed mechanical strain values) marked as circles in the graphical representation 500, are sent to the user. For example, if the mechanical strain values hover around the neutral line for 10,000 measurements, then approximately, 10 out of the 10,000 values can be sent to be stored and shown to a user.

[0052] Referring to FIG. 6, illustrates a graphical representation 600 of generating a signal if synchronized mechanical strain values exceed either a lower threshold limit or an upper threshold limit of mechanical strain at a given location coordinates of a train, in accordance with an embodiment of the present disclosure. With reference to the graphical representation 600, there is shown a X-axis 602 that represents time and a Y-axis 604 that represents mechanical strain values. There is further shown a curve 606 that represents a neutral zone, which is determined by the reference data's highest and lowest points. Furthermore, there is shown a first horizontal line 608 that represents a reference line (shown as a dotted line in the graphical representation 600 ). A plurality of second horizontal lines 610 (i.e., above and below the reference line, shown as solid lines) represents a first threshold level (i.e., the lower threshold limit of mechanical strain) to notify to change the driving style. A plurality of third horizontal lines 612 represents a second threshold level (i.e., the upper threshold limit of mechanical strain) to notify to send an alert to check the condition of the railway track. A first signal 614 (e.g., a warning signal of a yellow color) is generated if the measured mechanical strain values exceed the lower threshold limit of mechanical strain. The first signal 614 may be used to reduce the speed of the train. A second signal 616 (e.g., a red color signal) is generated if the measured mechanical strain values exceed the upper threshold limit of mechanical strain. The second signal 616 may be used to stop the train. The boundary between the first signal 614 and the second signal 616 is determined by adding a predetermined load factor (1, 2x-1, 5x measured with normal behavior) to the measured data points (i.e., the measured mechanical strain values).

[0053] Referring to FIG. 7A, illustrates a window for measured mechanical strain values and threshold level limits, in accordance with an embodiment of the present disclosure. There is shown a plurality of first curves 702 that represents threshold level limits of mechanical strain and a second curve 704 that represents an average data (or average mechanical strain values) which should be within the threshold level limits. The average data lies outside the threshold level limits at point A on the second curve 704, in response to, a signal is generated either to reduce the speed of the train or stop the train.

[0054] Referring to FIG. 7B, illustrates a window for measured mechanical strain values and threshold level limits, in accordance with another embodiment of the present disclosure. There is shown a plurality of first lines 706 that represents threshold level limits of mechanical strain and a curve 708 that represents an average data (or average mechanical strain values) which should be within the threshold level limits. The average data lies outside the threshold level limits at point B on the curve 708 in response to, a signal is generated either to reduce the speed of the train or stop the train.

[0055] Referring to FIG. 8, illustrates an exemplary user interface 800 that represents data points on a railroad map, in accordance with an embodiment of the present disclosure. The data points are coloured to represent the current state of the track at the specific data point. A data point 802 represents a yellow signal (not so critical) on the railroad map. The yellow signal may be an indication either to reduce the speed of the train or maintain the speed of the train within specific limits (e.g., 50-60 km / h). Another data point 804 represents a red signal (i.e., critical one) on the railroad map. The red signal may be an indication to stop the train. There is further shown a plurality of data points 806 that represents a green signal on the railroad map. The individual data point contains the measured data from part of the railway track that it represents, which is then provided to the operator of the train for monitoring the structural health of the railway track.

[0056] Referring to FIG. 9, illustrates a railway track 900 having different types of soil in different areas, in accordance with an embodiment of the present disclosure. The structural health of the railway track 900 is unknown. The railway track 900 is divided into four areas, for example, a first area 902 between points A1 to A2, a second area 904 between points B1 to B2, a third area 906 between points C1 to C2 and a fourth area 908 between points D1 to D2. The first area 902, the second area 904, the third area 906 and the fourth area 908 has different type of soil. For example, the soil of the first area 902 has stones, the soil of the second area 904 has sand, the soil of the third area 906 has bog and the soil of the fourth area 908 has stone. The starting and end point of each of the first area 902, the second area 904, the third area 906 and the fourth area 908 is determined by use of a land register (not shown here). The different types of soil affect the measurement of the mechanical strain values such that when the strain values are measured at the first area 902 (i.e., A1-A2, stone area) or at the second area 904 (i.e., B1-B2, sand area), the outcome is evidently different because the second area 904 is softer ground than the first area 902, which means that the railway track moves more when the train is driving over the second area 904. Moreover, when similar soil type (e.g., stone type) is used several times during a train trip between the point A1 to D2 then, the mechanical strain values measured within the first area 902 (i.e., A1-A2) can be used to teach an AI model to collect the data (i.e., the mechanical strain values) for the fourth area 908 (i.e., D1-D2). Although, the location is different but the soil type is same. This way, the data collection can be done in a faster way. Typically, the data is collected once per train trip from point A1 to D2.

[0057] Referring to FIG. 10, illustrates a graphical representation 1000 that represents generation of an alert signal, in accordance with another embodiment of the present disclosure. The graphical representation 1000 includes a time axis 1002, a measured strain values axis 1003, a reference line 1004 and threshold level limits 1006. The reference line 1004 represents a healthy railroad. The data is measured based on several train trips and it is observed that the data is measured between points 1 and 2 lies above the reference line 1004, the data measured at point 3 lies at the reference line 1004 and the data measured between points 4 to 8 lies below the reference line 1004. This is determined that the data measured at the point 8 exceeds the threshold level limits 1006 and therefore, causes an alert signal. Moreover, the data measured at points 4, 5 and 6 lies approximately in a straight line therefore, the data at the point 8 can be estimated. By calculating event time between the data measured at the points 4, 5 and 6, it can be estimated when the data at the point 8 would happen.

Examples

Embodiment Construction

[0009]The following detailed description illustrates embodiments of the present disclosure and ways in which they can be implemented. Although some modes of carrying out the present disclosure have been disclosed, those skilled in the art would recognize that other embodiments for carrying out or practising the present disclosure are also possible.

[0010]In a first aspect, the present disclosure provides a method for monitoring structural health of a railway track, the method comprising:

measuring mechanical strain values of a train undercarriage by using at least one strain gauge configured for measuring mechanical strain when the train is on the railway track; measuring location coordinates of the train when the train is on the railway track; synchronizing the measured mechanical strain values with the measured location coordinates of the train; comparing the synchronized mechanical strain values with a reference dataset, wherein the reference dataset comprises a lower threshold l...

TECHNICAL FIELD

[0001] The present disclosure relates to a method for monitoring structural health of a railway track. Moreover, the present disclosure relates to a system for monitoring structural health of a railway track.BACKGROUND

[0002] Nowadays, the analysis of railroad conditions relies heavily on specialized trains, which are purposefully-built for analyzing the railroad conditions.A Typically, only one such specialized train is employed per country due to its exorbitant cost (e.g., 30 to 40 million Euros per unit). The railway tracks can develop various defects over the time, such as corrugation, skip spots, and the like. If such kind of defects go undetected, it may lead to severe consequences, such as derailment. Conventionally, few methods have been developed for detecting the defects in the railway tracks, such as measuring carts, optical and laser-based methods and vibration sensors. Each of aforementioned methods have certain limitations, such as the measuring carts are expensive, slow and cumbersome because of consisting of an entire train cart. The optical methods can primarily detect debris and large obstacles around the track. Moreover, the vibration sensors are unable to detect certain faults, such as sinkage of a part of the railway track and skid spots.

[0003] Thus, there exists a technical problem of how to efficiently and reliably detect the defects in railroad tracks and provide a cost-effective solution for an enhanced safety while using the railway tracks.

[0004] Therefore, in light of the foregoing discussion, there exists a need to overcome the aforementioned drawbacks associated with the conventional ways of monitoring the condition of railway tracks as well as of the train.SUMMARY

[0005] The aim of the present disclosure is to provide a method and a system for monitoring structural health of a railway track. The aim of the present disclosure is achieved by a method and a system for monitoring structural health of a railway track that provides an efficient, more reliable and cost-effective solution for monitoring the condition of railway tracks. The disclosed method and system are based on monitoring fatigue of a mechanical structure, specifically emphasizing on the measurement of mechanical strain values that indicate changes, such as jumps, inclination and shaking of the railway tracks as defined in the appended independent claims to which reference is made to. Advantageous features are set out in the appended dependent claims.

[0006] Embodiments of the present disclosure substantially eliminate or at least partially address the aforementioned problems in the prior art, and enable an efficient and reliable monitoring of the railway tracks by measuring mechanical strain values of a train undercarriage as well as location coordinates of the train and providing a signal if the measured mechanical strain values start drifting from predetermined reference values of mechanical strain.

[0007] Throughout the description and claims of this specification, the words "comprise", "include", "have", and "contain" and variations of these words, for example "comprising" and "comprises", mean "including but not limited to", and do not exclude other components, items, integers or steps not explicitly disclosed also to be present. Moreover, the singular encompasses the plural unless the context otherwise requires. In particular, where the indefinite article is used, the specification is to be understood as contemplating plurality as well as singularity, unless the context requires otherwise.BRIEF DESCRIPTION OF THE DRAWINGS

[0008] FIG. 1 is a flowchart of a method for monitoring structural health of a railway track, in accordance with an embodiment of the present disclosure; FIG. 2 is a block diagram that illustrates various exemplary components of a system for monitoring structural health of a railway track, in accordance with an embodiment of the present disclosure; FIG. 3 illustrates position of a mechanical strain-gauge sensor module located on a train undercarriage, in accordance with an embodiment of the present disclosure; FIG. 4 illustrates installation of one or more mechanical strain-gauge sensor modules in different sections of a train, in accordance with an embodiment of the present disclosure; FIG. 5 illustrates a graphical representation of data compression, in accordance with an embodiment of the present disclosure; FIG. 6 illustrates a graphical representation of generating a signal if synchronized mechanical strain values exceed either a lower threshold limit or an upper threshold limit of mechanical strain at a given location coordinates of a train, in accordance with an embodiment of the present disclosure; FIG. 7A illustrates a window for measured mechanical strain values and threshold level limits, in accordance with an embodiment of the present disclosure; FIG. 7B illustrates a window for measured mechanical strain values and threshold level limits, in accordance with another embodiment of the present disclosure; FIG. 8 illustrates an exemplary user interface that represents data points on a railroad map, in accordance with another embodiment of the present disclosure; FIG. 9 illustrates a railway track having different types of soil in different areas, in accordance with an embodiment of the present disclosure; and FIG. 10 illustrates a graphical representation that represents generation of an alert signal, in accordance with another embodiment of the present disclosure. DETAILED DESCRIPTION OF EMBODIMENTS

[0009] The following detailed description illustrates embodiments of the present disclosure and ways in which they can be implemented. Although some modes of carrying out the present disclosure have been disclosed, those skilled in the art would recognize that other embodiments for carrying out or practising the present disclosure are also possible.

[0010] In a first aspect, the present disclosure provides a method for monitoring structural health of a railway track, the method comprising: measuring mechanical strain values of a train undercarriage by using at least one strain gauge configured for measuring mechanical strain when the train is on the railway track; measuring location coordinates of the train when the train is on the railway track; synchronizing the measured mechanical strain values with the measured location coordinates of the train; comparing the synchronized mechanical strain values with a reference dataset, wherein the reference dataset comprises a lower threshold limit and an upper threshold limit of mechanical strain for a plurality of different location coordinates of the train; generating a signal if the synchronized mechanical strain values exceed either the lower threshold limit or the upper threshold limit of mechanical strain at a given location coordinates of the train; and providing the generated signal to an operator of the train for monitoring the structural health of the railway track

[0011] In a second aspect, the present disclosure provides a system for monitoring structural health of a railway track, the system comprises: a mechanical strain-gauge sensor module comprising one or more mechanical strain-gauge sensors configured to measure mechanical strain values of a train undercarriage; a Global Positioning System receiver configured to measure location coordinates of the train on the railway track; and a computing device configured to: synchronize the measured mechanical strain values with the measured location coordinates of the train; compare the synchronized mechanical strain values with a reference dataset, wherein the reference dataset comprises a lower threshold limit and an upper threshold limit of mechanical strain for a plurality of different location coordinates of the train; generate a signal if the synchronized mechanical strain values exceed either the lower threshold limit or the upper threshold limit of mechanical strain at a given location coordinates of the train; and provide the generated signal to an operator of the train for monitoring the structural health of the railway track.

[0012] The present disclosure provides the aforementioned method and the aforementioned system for an efficient and reliable structural monitoring of the railway track. Also, the aforementioned method and the aforementioned system provide a cost-effective solution for the structural monitoring of the railway track. The aforementioned method and the aforementioned system explicitly disclose that the GPS receiver and the mechanical strain-gauge sensor module are mounted either on the same train or in different bogies of the same train or in different trains. Furthermore, the aforementioned method and the aforementioned system include synchronizing the measured mechanical strain values with the measured location coordinates of the train before storing the measured data into a memory of the aforementioned system. That's why the aforementioned method and the aforementioned system enable a reliable monitoring of the railway track. In contrast to the conventional methods of monitoring the structural health of the railway track, the aforementioned method and the aforementioned system generate the signal when the synchronized mechanical strain values exceed either the lower threshold limit or the upper threshold limit of mechanical strain at the given location coordinates of the train and provide the generated signal either to the operator of the train or a railway maintenance organization for a corresponding action required for maintaining the structural health of the railway track. Thus, the aforementioned method and the aforementioned system provides more efficient, reliable and fast monitoring of structural health of the railway track and ensures an enhanced safety when the train moves on the railway track.

[0013] Throughout the present disclosure, the term "monitoring structural health of a railway track" refers to a process or a technique that involves the continuous assessment and analysis of various conditions and integrity of infrastructure of the railway track. The structural health monitoring of the railway track includes the monitoring of key parameters, such as mechanical strain, vibrations and deformations in real time or near real time. The monitored parameters are used to detect and identify potential defects or anomalies that may compromise the safety and operational efficiency of the railway track.

[0014] Throughout the present disclosure, the term "system" refers to a specialized equipment that is configured for monitoring structural health of a railway track. It will be appreciated that the term "system" comprises a mechanical strain-gauge sensor module, a GPS receiver and a computing device communicably coupled to each of the mechanical strain-gauge sensor module and the GPS receiver. The system comprising the mechanical strain-gauge sensor module, the GPS receiver and the computing device is installed on a train. The term "mechanical strain-gauge sensor module" refers to a device designed to measure mechanical stress and deformations in structures, such as the railway tracks. The mechanical strain-gauge sensor module comprises one or more mechanical strain-gauge sensors, which are sensitive to changes in strain and therefore, detect alterations in shape or mechanical loading of the railway track affecting to mechanical strains of the train undercarriage. The alterations in shape or mechanical loading of the railway track may happen slowly due to an erosion or a frost of the soil below the railway track, or fast due to accidents or if e.g. a railway engine hums in a one place. Thus, the mechanical strain-gauge sensor module contributes in assessment of the structural health of the railway track and supports the identification of potential defects or irregularities of the railway track.

[0015] The term "GPS receiver" refers to a device that uses signals from satellites to determine an accurate and real-time geographical location on earth.

[0016] The term "computing device" refers to a device that performs computations on the measured mechanical strain values and the measured location coordinates of the train. Examples of implementation of the computing device may include but are not limited to, a central processing unit (CPU), an edge computing device, a processor, a coprocessor, a microprocessor, a microcontroller, a complex instruction set computing (CISC) processor, an application-specific integrated circuit (ASIC) processor, a reduced instruction set (RISC) processor, an integrated circuit, a very long instruction word (VLIW) processor, a state machine, a data processing unit, and other processors or circuits. Moreover, the computing device, may refer to one or more individual processors, processing devices, a processing unit that is part of a machine.

[0017] In operation, the method comprises measuring mechanical strain values of a train undercarriage by using at least one strain gauge configured for measuring mechanical strain when the train is on the railway track. The mechanical strain values of the train undercarriage are measured using various sensors, such as mechanical strain gauge sensors. The use of the mechanical strain gauge sensors quantifies the deformation or stress experienced by various components of the train's undercarriage (that is a supporting frame under the body of the train for example, a wheeled structure). Alternatively stated, any stress or deformation experienced by the wheeled structure of the train is measured when the train is on the railway track. Alternatively, the method comprises a loop: measure mechanical strain, save measurement results, calibrate, measure mechanical strain, etc. when measuring mechanical strain values. The measurement of the mechanical strain values of the train undercarriage supports in monitoring and assessment of structural integrity of the railway track. Optionally the mechanical strain values are measured on the frame of the train or the cart.

[0018] Optionally, the mechanical strain is measured when the train is moving. In an implementation, the mechanical strain values of the train undercarriage are measured when the train is moving on the railway track.

[0019] The method further comprises measuring location coordinates of the train when the train is on the railway track and synchronizing the measured mechanical strain values with the measured location coordinates of the train. In addition to the mechanical strain values of the train undercarriage, the location coordinates (i.e., geographical coordinates) of the train are also measured. The measurement of the location coordinates includes measurement of latitude and longitude coordinates which, are used to determine the horizontal position of the train on Earth's surface. Furthermore, the measured mechanical strain values of the train undercarriage and the measured location coordinates of the train are synchronized before storing at a storage space (e.g., a cloud sever) in order to collect statistical force data (i.e., the statistical data of mechanical strain values for a provided location).

[0020] Optionally, synchronizing the measured mechanical strain values with the measured location coordinates of the train comprises matching time stamps of the measured mechanical strain values with the measured location coordinates of the train. The measured mechanical strain values of the train undercarriage are synchronized with the measured location coordinates of the train by matching the time stamps of the measured mechanical strain values with the time stamps of the measured location coordinates of the train.

[0021] The method further comprises comparing the synchronized mechanical strain values with a reference dataset, wherein the reference dataset comprises a lower threshold limit and an upper threshold limit of mechanical strain for a plurality of different location coordinates of the train. Furthermore, the synchronized mechanical strain values of the train undercarriage and the location coordinates of the train are compared with the reference dataset. The reference dataset comprises the lower threshold limit and the upper threshold limit of the mechanical strain for different location coordinates of the train.

[0022] Optionally, the method further comprises storing the synchronized mechanical strain values to collect a dataset of mechanical strain values, computing the lower threshold limit and the upper threshold limit of mechanical strain for the plurality of different location coordinates of the train based on the dataset of mechanical strain values and storing the lower threshold limit and the upper threshold limit of mechanical strain for the plurality of different location coordinates of the train. The synchronized mechanical strain values of the train undercarriage and the location coordinates of the train are stored in a memory to collect the dataset of mechanical strain values. The collected dataset may also be referred to as a raw dataset and it differs from the reference dataset. The collected dataset can be a dataset having either dynamical values of mechanical strain or statistical values of mechanical strain or historical values of mechanical strain or a combination of dynamical values, statistical values and historical values of mechanical strain. Thereafter, the collected dataset is used to compute the lower threshold limit and the upper threshold limit of mechanical strain for different location coordinates of the train. The computed lower threshold limit and the upper threshold limit of mechanical strain for different location coordinates is stored, for example, either in a memory of the disclosed system or in a cloud server wirelessly connected to the disclosed system.

[0023] Optionally, the reference dataset is updated based on measurements. The reference dataset that is the lower threshold limit and the upper threshold limit of mechanical strain are updated time-to-time (e.g., over a time period of 6 months or 1 year and may be more) based on the mechanical strain values of the train undercarriage and the location coordinates of the train measured over a specific time period. Moreover, the collected dataset (i.e., the raw dataset) is also used to update the reference dataset (i.e., the lower threshold limit and the upper threshold limit of mechanical strain for different location coordinates of the train). Optionally, the collected dataset (i.e., the raw dataset) is used to update the reference dataset only when measured datapoints are within the lower threshold limit and the upper threshold limit i.e., when there is a fault situation on the railway track the measured data is not used to update the reference dataset.

[0024] Optionally, the method further comprises storing the synchronized mechanical strain values in a cloud server after a pre-processing of the synchronized mechanical strain values, where the stored synchronized mechanical strain values are used for subsequent analysis. The synchronized mechanical strain values and location coordinates of the train may be stored in the cloud server for further analysis of the synchronized mechanical strain values.

[0025] Throughout the present disclosure, the term "cloud server" refers to a device used to store the synchronized mechanical strain values and the location coordinates of the train. Examples of the cloud server may include, but are not limited to, an application server, a storage server, or a combination thereof. Moreover, the cloud server may either be a single hardware server or a plurality of hardware servers operating in a parallel or distributed architecture to store the synchronized mechanical strain values and the location coordinates of the train and perform the subsequent analysis on the synchronized data.

[0026] At the cloud server, the synchronized mechanical strain values may be further processed and analyzed and may be displayed on a plurality of electronic devices, such as mobile phone, laptop, desktop computer, smart phone, and the like, by use of a web application. Before storing the synchronized mechanical strain values in the cloud server, the synchronized mechanical strain values are subjected to the pre-processing. The pre-processing includes filtration of noise (i.e., any noise present in the synchronized mechanical strain values), decimation of the synchronized mechanical strain values, and the like, while decimation can be used for example depending on the route, speed of train, or any other variables. The decimation refers to a process of reducing sampling rate of the synchronized mechanical strain values by removing or discarding few strain values from the synchronized mechanical strain values, while preserving the information required for an intended application.

[0027] Optionally, the pre-processing comprises applying data compression on the synchronized mechanical strain values to obtain a dataset of mechanical strain values with a comparatively smaller dimension. The synchronized mechanical strain values are simplified by use of the pre-processing. The pre-processing may include applying the data compression on the synchronized mechanical strain values to obtain the dataset of comparatively smaller dimensions, which is further stored in the cloud server. This is advantageous to obtain the dataset of smaller dimensions because sending the dataset of smaller dimensions to the cloud server results in bandwidth saving. The bandwidth saving can be enhanced by sending the dataset of smaller dimensions less frequently to the cloud server, for example, at each railway station. In a case, if there is no network connection to the cloud server then in such case, the dataset can be temporarily saved and shared with the cloud server after some time, for example, at next railway station where network connection is present. Furthermore, the data compression can be applied on the synchronized mechanical strain values by using any one of: a peak valley algorithm, a digital low pass filter, an audio compression algorithm, and skipping some of data points from the synchronized mechanical strain values. The peak valley algorithm identifies points in the sequence where the value is significantly higher (a peak) or lower (a valley) than its neighboring values. Minimum and maximum turning points are also identified making it possible to pair peaks and valleys together making it possible various analysis of the measured mechanical strain values including e.g. calculating the amplitudes of peaks and valleys, determining the frequency of occurrence of peaks and valleys, measuring the distance between peak-valley pairs, etc. The audio compression algorithm may include lossy and lossless compression algorithms, such as MPEG-1 Audio Layer III (MP3), Advanced Audio Coding (AAC), Free Lossless Audio Codec (FLAC), OGG Vorbis, WAVEPACK algorithm, Apple Lossless Audio Codec (ALAC), Monkey's Audio, Opus, Sub-Band Coding (SBC), Musical Instrument Digital Interface (MIDI), Adaptive Differential Pulse-Code Modulation (ADPCM), Linear Predictive Coding (LPC), and the like. Some data points may be skipped in an event of long stretches (i.e., extended or prolonged periods of time) where no new insights can be made from the measured data. For example, if the strain data hovers for 10,000 consecutive measurements, then approximately 10 values can be sent to the cloud server for storage and rest values can be skipped or discarded, resulting in a significant bandwidth saving.

[0028] Optionally, computing the reference dataset comprises measuring a first set of mechanical strain values on a first portion of the railway track of length L x by running a number of laps of the train through the first portion of the railway track until consistent boundary values are obtained, wherein a state of the first portion of the railway track is known to be in a healthy state, storing the first set of mechanical strain values measured on the first portion of the railway track of length L x and utilizing the first set of mechanical strain values to determine the lower threshold limit and the upper threshold limit of mechanical strain for the first portion of the railway track to determine a state of the track based on a behavior of the train undercarriage when the train is on the track at the first portion of the railway track of length L x . The first set of mechanical strain values is measured on the first portion of the railway track of Length L x (e.g., 1 to 1000 km) by running X (e.g., 50, 100 or more) number of laps through the first portion until the consistent boundary values are obtained. The first portion of the railway track is known to be in the healthy state. Alternatively stated, the state of the first portion of the railway track is already verified either through inspection (e.g., manual inspection) or by other means. The measured first set of mechanical strain values is used to determine the lower threshold limit and the upper threshold limit of mechanical strain for the first portion of the railway track of length L x . Thereafter, the determined lower threshold limit and the upper threshold limit of mechanical strain are used to determine the state of the track based on the behavior of the train undercarriage when the train moves on the track at the first portion of the railway track of length L x . The state of the track is determined as the healthy state. The data similar to the first set of mechanical strain values is collected for each track separately in a country or such area by first running a known number of laps through a portion of each track until the consistent boundary values can be obtained. The same data can be used to calibrate other trains running the same track. This way, the reference data comprises the lower threshold limit and the upper threshold limit of mechanical strain for the railway tracks which are considered in the healthy condition.

[0029] Optionally, computing the reference dataset comprises measuring a second set of mechanical strain values on a second portion of the railway track of length L x+y by running a number of laps of the train through the second portion of the railway track, wherein a state of the second portion of the railway track is unknown, storing the second set of mechanical strain values measured on the second portion of the railway track and utilizing the second set of mechanical strain values to determine the lower threshold limit and the upper threshold limit of mechanical strain for the second portion of the railway track. The second set of mechanical strain values is measured on the second portion of the railway track of Length L x+y (e.g., 1000 to 2000 km) by running multiple number of laps through the second portion of the railway track. The state of the second portion of the railway track is unknown which means the second portion of the railway track may not be in a useful condition. The second set of measured mechanical strain values is stored and used to determine the lower threshold limit and the upper threshold limit of mechanical strain for the second portion of the railway track. Thereafter, the determined lower threshold limit and the upper threshold limit of mechanical strain are used to determine the state of the track based on the behavior of the train undercarriage when the train moves on the track at the second portion of the railway track of length L x+y . This way, the reference data comprises the lower threshold limit and the upper threshold limit of mechanical strain for the railway tracks whose state is unknown but can be determined by use of the determined lower threshold limit and the upper threshold limit of mechanical strain for the second portion of the railway track.

[0030] Optionally, measuring the second set of mechanical strain values on the second portion of the railway track comprises measuring a first subset of mechanical strain values from at least two different locations having a first type of soil and a second subset of mechanical strain values from a location having a second type of soil. For example, the first subset of mechanical strain values is measured from the soil having stones (i.e., the first type of soil) and the second subset of mechanical strain values is measured from the soil having sand (i.e., the second type of soil). This means that the soil is different in terms of technical characteristics at different locations and does affect the measurement such that the first subset of mechanical strain values measured from the location having the first type of soil (i.e., stone soil) is evidently different from the second subset of mechanical strain values measured from the location having the second type of soil (i.e., sand soil).

[0031] Optionally, the mechanical strain is measured. The measured points are added in a 180-degree phase-shift compared to reference data and when the data differs from 0-point, then a signal is sent. In normal track condition the sum-effect of the measured data and 180-degree phase-shifted reference data the difference should be near to zero (0). This means that when there is a difference the difference is only sent to cloud for further processing saving data capacity and processing power.

[0032] Optionally, a ground frost of the soil is used to define at least one of the following: soil modulus of elasticity, soil hardness. The ground frost data is collected to define the soil modulus of elasticity. For example, in winters, a sand soil may look like a stone. The ground frost data can be obtained from public data sources or performing multiple measurements with different soil conditions (i.e., measuring during different seasons). Temperatures and how wet the soil is can give information about how much the soil characteristics varies. From this it might be possible to estimate ground frost (depth). The ground frost data can also be used in calibrating the mechanical strain values because there are different soil modulus of elasticity in different ground frost areas. Based on the measured mechanical strain values over the ground frost of the soil, the change of the ground frost can be estimated too. When the ground frost is at the most (e.g., after a long cold winter), the reference data can be collected in a much easier way because the ground soil "error" can be eliminated. Optionally the reference data is collected separately based on a time of the year, month, a ground frost, a vehicle model.

[0033] Optionally, the mechanical strain is measured when the train is moving at two different speeds (v1, v2) at the first portion of the railway track of length L x or at the second portion of the railway track of length L x+y . For example, the first set of mechanical strain values includes the mechanical strain values measured when the train is moving at speed v1 and also, the mechanical strain values measured when the train is moving at speed v2 at the first portion of the railway track of length L x . Similarly, the second set of mechanical strain values includes the mechanical strain values measured when the train is moving at speed v1 and also, the mechanical strain values measured when the train is moving at speed v2 at the second portion of the railway track of length L x+y .

[0034] Optionally, the mechanical strain measurements performed at the two different speeds (v1, v2) of the moving train are compared to each other. For example, the mechanical strain values measured when the train is moving at speed v1 and the mechanical strain values measured when the train is moving at speed v2, are compared with each other.

[0035] The method further comprises generating a signal if the synchronized mechanical strain values exceed either the lower threshold limit or the upper threshold limit of mechanical strain at a given location coordinates of the train. For example, if the measured mechanical strain values are outside of the lower threshold limit, the signal is generated to slow down the speed of the train. If the measured mechanical strain values are outside of the upper threshold limit, the signal is generated to stop the train. If the measured mechanical strain values are within the lower threshold limit and the upper threshold limit, the signal is generated either to speed up the train or maintain the speed of the train.

[0036] Optionally, generating the signal comprises generating a first signal if the first subset of mechanical strain values measured from the at least two different locations having the first type of soil on the second portion of the railway track exceeds the lower threshold limit of mechanical strain; and generating a second signal if the second subset of mechanical strain values measured from the location having the second type of soil on the second portion of the railway track exceeds the upper threshold limit of mechanical strain. For example, the first signal can be generated like a warning signal to reduce the speed of the train when the first subset of mechanical strain values exceeds the lower threshold limit of mechanical strain. And the second signal can be generated like an emergency signal to stop the train when the second subset of mechanical strain values exceeds the upper threshold limit of mechanical strain.

[0037] The method further comprises providing the generated signal to an operator of the train for monitoring the structural health of the railway track. The generated signal can be displayed to the operator of the train by use of a color map having a number of data points in different colors (for example, green, red and yellow color). For example, the generated signal can be in the form of a data point having a red color on the color map which, refers to an alert situation on a rail track map and used to alarm the operator of the train in a case when a rail track condition is getting worse or operation of the train is unsafe. The operator can be either a train driver or an operator outside of the train for example, for monitoring the condition of the railway track. Mostly, the operator is outside of the train for monitoring the condition of the railway track.

[0038] Optionally, a corresponding action for maintaining structural health of the railway track is performed by utilizing the generated signal. Based on the generated signal, the corresponding action (e.g., remanufacture or repair) can be performed to maintain the structural health of the railway track.

[0039] The present disclosure also relates to the system as described above. Various embodiments and variants disclosed above, with respect to the aforementioned method, apply mutatis mutandis to the system.

[0040] Optionally, the one or more mechanical strain-gauge sensors of the mechanical strain-gauge sensor module are installed in different sections of the train. In an implementation, the one or more strain-gauge sensors can be installed on one tandem axle, for example, sensor a on left side and sensor b on right side, to measure strain values on both sides of the railway track, individually. This is advantageous to install the one or more mechanical strain-gauge sensors in different sections of the train to measure data (i.e., the mechanical strain values) with more precision and to eliminate the false measurement results in a simple and very fast way. The installation of the one or more mechanical strain-gauge sensors in different sections of the train lead to multiplication of the measured data with the number of strain-gauge sensors resulting in faster data collection. For example, in one case, three strain-gauge sensors can be installed in three different bogies of same train. For instance, a first mechanical strain-gauge sensor can be installed in a bogie lying at the beginning of the train, a second mechanical strain-gauge sensor can be installed in a bogie lying at the middle of the train and a third mechanical strain-gauge sensor can be installed in a bogie lying at the end of the train. In another case, the three strain-gauge sensors can be installed in three different trains.

[0041] Optionally, the one or more mechanical strain-gauge sensors of the mechanical strain-gauge sensor module and the GPS receiver are calibrated to a pre-determined starting point of the train before starting measurement of the mechanical strain values and the location coordinates of the train, respectively. The one or more mechanical strain-gauge sensors of the mechanical strain-gauge sensor module are calibrated to a certain and the pre-determined starting point before starting the measurement of the mechanical strain values. The pre-determined starting point is then used as a reference point for all the mechanical strain values collected over the railway track. The GPS receiver is also calibrated to ensure that that GPS receiver displays the location correctly before the measurement of data begins. Moreover, the system may have an automatic check within a software that accounts for moving of a neutral area (may also be referred as a 0-line) and re-evaluates the different zones according to new position of the neutral area. This is done by checking the data over a long period of time (e.g., 200 km long part of track) and adjusting the neutral area if the data is shifted from the expected neutral area.

[0042] Optionally, the system has a wireless connection to a cloud server configured to store the synchronized mechanical strain values after a pre-processing, wherein the stored synchronized mechanical strain values are used for subsequent analysis. For example, the system may be connected to the cloud server through a Local Area Network (LAN) connection.

[0043] Optionally, the pre-processing comprises applying data compression on the synchronized mechanical strain values to obtain a dataset of mechanical strain values with a comparatively smaller dimension than the reference dataset.

[0044] Optionally, the system further comprises an inclinometer sensor to detect a difference between elevation of different sides of the railway track. For example, left rail of the railway track may be higher than right rail of the railway track. In addition to the inclinometer sensor, other sensors, such as Inertial Measurement Unit (IMU), gyroscope or accelerometer may also be used in the system. The system may include an Artificial Intelligence (AI) model to estimate mechanical strain values using previously measured mechanical strain values for different location coordinates of the train. The AI model may also be used to estimate the mechanical strain values for the railway track having different soil types based on the mechanical strain values measured for similar soil types.

[0045] Optionally, the system further works as follows: Strain gauges are placed in the undercarriage of the train together with a GPS tracker (and other devices) to measure the strains (and other possible data) occurring in the structures of the train and to combine it with the location data of where the strains occurred into a pinpoint the location of the possible defect on the track. The signal (measurement data) omitted from the strain gauges, typically through analogue to digital conversion unit, is then transmitted to a CPU and later sent to a cloud service or to an external user. Optionally the raw data can be sent within a predetermined time, e.g. within 5 seconds, to a cloud service or to an external user in a case of alert signal caused by a possible defect in the track is detected. These are used to store data as well as conduct different analyses to detect threshold passing values and to detect several different types of faults. After detecting a possible defect in the track an appropriate signal (alert) is produced to inform the correct authorities about the abnormality in the data.DETAILED DESCRIPTION OF THE DRAWINGS

[0046] Referring to FIG. 1, illustrates steps of a method 100 for monitoring structural health of a railway track, in accordance with an embodiment of the present disclosure. At step 102, the method 100 includes measuring mechanical strain values of a train undercarriage by using at least one strain gauge configured for measuring mechanical strain when the train is on the railway track. At step 104, the method 100 includes measuring location coordinates of the train when the train is on the railway track. At step 106, the method 100 includes synchronizing the measured mechanical strain values with the measured location coordinates of the train. At step 108, the method 100 includes comparing the synchronized mechanical strain values with a reference dataset, where the reference dataset comprises a lower threshold limit and an upper threshold limit of mechanical strain for a plurality of different location coordinates of the train. At step 110, the method 100 includes generating a signal if the synchronized mechanical strain values exceed either the lower threshold limit or the upper threshold limit of mechanical strain at a given location coordinates of the train. At step 112, the method 100 includes providing the generated signal to an operator of the train for monitoring the structural health of the railway track.

[0047] The aforementioned steps are only illustrative and other alternatives can also be provided where one or more steps are added, one or more steps are removed, or one or more steps are provided in a different sequence without departing from the scope of the claims herein.

[0048] Referring to FIG. 2, illustrates a block diagram of a system 200 installed in a train for monitoring structural health of a railway track, in accordance with an embodiment of the present disclosure. The system 200 includes a mechanical strain-gauge sensor module 202 comprising one or more mechanical strain-gauge sensors 202A, a GPS receiver 204 and a computing device 206. The computing device 206 is further connected to a cloud server 208 through a wireless connection, e.g., an internet connection (represented by dotted lines in FIG. 2) and to a display device 214 through a local wired connection (represented by a solid line in FIG. 2). The mechanical strain-gauge sensor module 202 comprising the one or more mechanical strain-gauge sensors 202A is configured to measure mechanical strain values of a train undercarriage. The GPS receiver 204 is configured to measure location coordinates of the train when the train is on the railway track. The computing device 206 is configured to receive the measured mechanical strain values of the train undercarriage from the mechanical strain-gauge sensor module 202 and the measured location coordinates of the train from the GPS receiver 204. The computing device 206 may include a processor, a program execution memory and a data memory (not shown for sake of brevity). The measured mechanical strain values and the measured location coordinates of the train are processed by a program stored in the program execution memory, which is executed by the microprocessor. The computing device 206 is configured to synchronize the measured mechanical strain values with the measured location coordinates of the train. The measured mechanical strain values and the measured location coordinates of the train are pre-processed up to the extent that the synchronized mechanical strain values which are stored at the cloud server 208 have smaller dimensions. The pre-processing of the synchronized mechanical strain values is required because the synchronized mechanical strain values, which are transferred to the cloud server 208 are used for further analysis and the objective is to avoid sending the redundant data for bandwidth saving. The computing device 206 is configured to compare the synchronized mechanical strain values with a reference dataset, where the reference dataset comprises a lower threshold limit and an upper threshold limit of mechanical strain for a plurality of different location coordinates of the train. The computing device 206 is configured to generate a signal if the synchronized mechanical strain values exceed either the lower threshold limit or the upper threshold limit of mechanical strain at a given location coordinates of the train. The computing device 206 is configured to provide the generated signal to an operator 216 of the train for monitoring the structural health of the railway track. The generated signal can be made visible to the operator 216 of the train by displaying on the display device 214 (e.g., a Liquid Crystal Display (LCD) display device, Light Emitting Diode (LED) display device, a display screen, and the like) connected to the computing device 206 through the local wired or wireless connection. Alternatively, the generated signal can be analyzed at the cloud server 208 and displayed on a plurality of electronic devices 212 (e.g., a smart phone, a laptop, a desktop computer, and the like) through a web application 210.

[0049] Referring to FIG. 3, illustrates position of the mechanical strain-gauge sensor module 202 located on a train undercarriage 302, in accordance with an embodiment of the present disclosure.

[0050] Referring to FIG. 4, illustrates installation of one or more mechanical strain-gauge sensor module in different sections of a train 400, in accordance with an embodiment of the present disclosure. A first mechanical strain-gauge sensor module 402A is installed in a bogie lying at the beginning of the train 400, a second mechanical strain-gauge sensor module 402B is installed in a bogie lying at the middle of the train 400, and a third mechanical strain-gauge sensor module 402C is installed in a bogie lying at the end of the train 400. Each of the first mechanical strain-gauge sensor module 402A, the second mechanical strain-gauge sensor module 402B and the third mechanical strain-gauge sensor module 402C comprises one or more mechanical strain-gauge sensors to measure mechanical strain values of the train 400 undercarriage. In FIG. 4, the first mechanical strain-gauge sensor module 402A, the second mechanical strain-gauge sensor module 402B and the third mechanical strain-gauge sensor module 402C is installed in different bogies of the same train (i.e., the train 400 ). In another implementation, the first mechanical strain-gauge sensor module 402A, the second mechanical strain-gauge sensor module 402B and the third mechanical strain-gauge sensor module 402C can be installed in different trains. Only, three mechanical strain-gauge sensor modules are shown in FIG. 4 which, are installed in the train 400 (for sake of brevity). Although, more than three mechanical strain-gauge sensor modules can be installed in the train 400.

[0051] Referring to FIG. 5, illustrates a graphical representation 500 of data compression, in accordance with an embodiment of the present disclosure. With reference to the graphical representation 500, there is shown a X-axis 502 that represents distance travelled by a train on a railway track and a Y-axis 504 that represents mechanical strain values. There is further shown a first horizontal line 506 that represents a reference line (shown as a dotted line in the graphical representation 500). A plurality of second horizontal lines 508 (i.e., above and below the reference line, shown as solid lines) represents a first threshold level (i.e., lower threshold limit of mechanical strain) to notify to change the driving style. A plurality of third horizontal lines 510 represents a second threshold level (i.e., upper threshold limit of mechanical strain) to notify to send an alert to check the condition of the railway track. A fourth line 512 represents larger number of data points (i.e., the mechanical strain values) which, are not sent in order to compress the data. Only the data points (e.g., compressed mechanical strain values) marked as circles in the graphical representation 500, are sent to the user. For example, if the mechanical strain values hover around the neutral line for 10,000 measurements, then approximately, 10 out of the 10,000 values can be sent to be stored and shown to a user.

[0052] Referring to FIG. 6, illustrates a graphical representation 600 of generating a signal if synchronized mechanical strain values exceed either a lower threshold limit or an upper threshold limit of mechanical strain at a given location coordinates of a train, in accordance with an embodiment of the present disclosure. With reference to the graphical representation 600, there is shown a X-axis 602 that represents time and a Y-axis 604 that represents mechanical strain values. There is further shown a curve 606 that represents a neutral zone, which is determined by the reference data's highest and lowest points. Furthermore, there is shown a first horizontal line 608 that represents a reference line (shown as a dotted line in the graphical representation 600 ). A plurality of second horizontal lines 610 (i.e., above and below the reference line, shown as solid lines) represents a first threshold level (i.e., the lower threshold limit of mechanical strain) to notify to change the driving style. A plurality of third horizontal lines 612 represents a second threshold level (i.e., the upper threshold limit of mechanical strain) to notify to send an alert to check the condition of the railway track. A first signal 614 (e.g., a warning signal of a yellow color) is generated if the measured mechanical strain values exceed the lower threshold limit of mechanical strain. The first signal 614 may be used to reduce the speed of the train. A second signal 616 (e.g., a red color signal) is generated if the measured mechanical strain values exceed the upper threshold limit of mechanical strain. The second signal 616 may be used to stop the train. The boundary between the first signal 614 and the second signal 616 is determined by adding a predetermined load factor (1, 2x-1, 5x measured with normal behavior) to the measured data points (i.e., the measured mechanical strain values).

[0053] Referring to FIG. 7A, illustrates a window for measured mechanical strain values and threshold level limits, in accordance with an embodiment of the present disclosure. There is shown a plurality of first curves 702 that represents threshold level limits of mechanical strain and a second curve 704 that represents an average data (or average mechanical strain values) which should be within the threshold level limits. The average data lies outside the threshold level limits at point A on the second curve 704, in response to, a signal is generated either to reduce the speed of the train or stop the train.

[0054] Referring to FIG. 7B, illustrates a window for measured mechanical strain values and threshold level limits, in accordance with another embodiment of the present disclosure. There is shown a plurality of first lines 706 that represents threshold level limits of mechanical strain and a curve 708 that represents an average data (or average mechanical strain values) which should be within the threshold level limits. The average data lies outside the threshold level limits at point B on the curve 708 in response to, a signal is generated either to reduce the speed of the train or stop the train.

[0055] Referring to FIG. 8, illustrates an exemplary user interface 800 that represents data points on a railroad map, in accordance with an embodiment of the present disclosure. The data points are coloured to represent the current state of the track at the specific data point. A data point 802 represents a yellow signal (not so critical) on the railroad map. The yellow signal may be an indication either to reduce the speed of the train or maintain the speed of the train within specific limits (e.g., 50-60 km / h). Another data point 804 represents a red signal (i.e., critical one) on the railroad map. The red signal may be an indication to stop the train. There is further shown a plurality of data points 806 that represents a green signal on the railroad map. The individual data point contains the measured data from part of the railway track that it represents, which is then provided to the operator of the train for monitoring the structural health of the railway track.

[0056] Referring to FIG. 9, illustrates a railway track 900 having different types of soil in different areas, in accordance with an embodiment of the present disclosure. The structural health of the railway track 900 is unknown. The railway track 900 is divided into four areas, for example, a first area 902 between points A1 to A2, a second area 904 between points B1 to B2, a third area 906 between points C1 to C2 and a fourth area 908 between points D1 to D2. The first area 902, the second area 904, the third area 906 and the fourth area 908 has different type of soil. For example, the soil of the first area 902 has stones, the soil of the second area 904 has sand, the soil of the third area 906 has bog and the soil of the fourth area 908 has stone. The starting and end point of each of the first area 902, the second area 904, the third area 906 and the fourth area 908 is determined by use of a land register (not shown here). The different types of soil affect the measurement of the mechanical strain values such that when the strain values are measured at the first area 902 (i.e., A1-A2, stone area) or at the second area 904 (i.e., B1-B2, sand area), the outcome is evidently different because the second area 904 is softer ground than the first area 902, which means that the railway track moves more when the train is driving over the second area 904. Moreover, when similar soil type (e.g., stone type) is used several times during a train trip between the point A1 to D2 then, the mechanical strain values measured within the first area 902 (i.e., A1-A2) can be used to teach an AI model to collect the data (i.e., the mechanical strain values) for the fourth area 908 (i.e., D1-D2). Although, the location is different but the soil type is same. This way, the data collection can be done in a faster way. Typically, the data is collected once per train trip from point A1 to D2.

[0057] Referring to FIG. 10, illustrates a graphical representation 1000 that represents generation of an alert signal, in accordance with another embodiment of the present disclosure. The graphical representation 1000 includes a time axis 1002, a measured strain values axis 1003, a reference line 1004 and threshold level limits 1006. The reference line 1004 represents a healthy railroad. The data is measured based on several train trips and it is observed that the data is measured between points 1 and 2 lies above the reference line 1004, the data measured at point 3 lies at the reference line 1004 and the data measured between points 4 to 8 lies below the reference line 1004. This is determined that the data measured at the point 8 exceeds the threshold level limits 1006 and therefore, causes an alert signal. Moreover, the data measured at points 4, 5 and 6 lies approximately in a straight line therefore, the data at the point 8 can be estimated. By calculating event time between the data measured at the points 4, 5 and 6, it can be estimated when the data at the point 8 would happen.

Examples

Embodiment Construction

[0009]The following detailed description illustrates embodiments of the present disclosure and ways in which they can be implemented. Although some modes of carrying out the present disclosure have been disclosed, those skilled in the art would recognize that other embodiments for carrying out or practising the present disclosure are also possible.

[0010]In a first aspect, the present disclosure provides a method for monitoring structural health of a railway track, the method comprising:

measuring mechanical strain values of a train undercarriage by using at least one strain gauge configured for measuring mechanical strain when the train is on the railway track; measuring location coordinates of the train when the train is on the railway track; synchronizing the measured mechanical strain values with the measured location coordinates of the train; comparing the synchronized mechanical strain values with a reference dataset, wherein the reference dataset comprises a lower threshold l...

Claims

1. A method (100) for monitoring structural health of a railway track (900), the method (100) comprising: measuring mechanical strain values of a train undercarriage (302) by using at least one strain gauge configured for measuring mechanical strain when the train (400) is on the railway track (900); measuring location coordinates of the train (400) when the train (400) is on the railway track (900); synchronizing the measured mechanical strain values with the measured location coordinates of the train (400); comparing the synchronized mechanical strain values with a reference dataset, wherein the reference dataset comprises a lower threshold limit and an upper threshold limit of mechanical strain for a plurality of different location coordinates of the train (400); generating a signal if the synchronized mechanical strain values exceed either the lower threshold limit or the upper threshold limit of mechanical strain at a given location coordinates of the train (400); and providing the generated signal to an operator (216) of the train (400) for monitoring the structural health of the railway track (900).

2. The method (100) of claim 1, further comprising: storing the synchronized mechanical strain values to collect a dataset of mechanical strain values; computing the lower threshold limit and the upper threshold limit of mechanical strain for the plurality of different location coordinates of the train (400) based on the dataset of mechanical strain values; and storing the lower threshold limit and the upper threshold limit of mechanical strain for the plurality of different location coordinates of the train (400).

3. The method (100) of claim 2, further comprising storing the synchronized mechanical strain values in a cloud server (208) after a pre-processing of the synchronized mechanical strain values, wherein the stored synchronized mechanical strain values are used for subsequent analysis.

4. The method (100) of claim 3, wherein the pre-processing comprises applying data compression on the synchronized mechanical strain values to obtain a dataset of mechanical strain values with a comparatively smaller dimension.

5. The method (100) of any of claims 2-4, wherein computing the reference dataset comprising: measuring a first set of mechanical strain values on a first portion of the railway track of length Lx by running a number of laps of the train (400) through the first portion of the railway track until consistent boundary values are obtained, wherein a state of the first portion of the railway track is known to be in a healthy state; storing the first set of mechanical strain values measured on the first portion of the railway track of length Lx; and utilizing the first set of mechanical strain values to determine the lower threshold limit and the upper threshold limit of mechanical strain for the first portion of the railway track to determine a state of the track based on a behavior of the train undercarriage (302) when the train (400) is on the track at the first portion of the railway track of length Lx.

6. The method (100) of any of claims 2-5, wherein computing the reference dataset comprising: measuring a second set of mechanical strain values on a second portion of the railway track of length Lx+y by running a number of laps of the train (400) through the second portion of the railway track, wherein a state of the second portion of the railway track is unknown; storing the second set of mechanical strain values measured on the second portion of the railway track; and utilizing the second set of mechanical strain values to determine the lower threshold limit and the upper threshold limit of mechanical strain for the second portion of the railway track.

7. The method (100) of claim 6, wherein measuring the second set of mechanical strain values on the second portion of the railway track comprises measuring a first subset of mechanical strain values from at least two different locations having a first type of soil and a second subset of mechanical strain values from a location having a second type of soil.

8. The method (100) of any of claims 1 or 7, wherein generating the signal comprises: generating a first signal (614) if the first subset of mechanical strain values measured from the at least two different locations having the first type of soil on the second portion of the railway track exceeds the lower threshold limit of mechanical strain; and generating a second signal (616) if the second subset of mechanical strain values measured from the location having the second type of soil on the second portion of the railway track exceeds the upper threshold limit of mechanical strain.

9. The method (100) of any of previous claims, wherein a ground frost of the soil is used to define at least one of the following: soil modulus of elasticity, soil hardness.

10. The method (100) of any of claims 5 or 6, wherein the reference dataset is updated based on measurements.

11. The method (100) of any of previous claims, wherein the mechanical strain is measured when the train (400) is moving.

12. The method (100) of claim 11, wherein the mechanical strain is measured when the train (400) is moving at two different speeds (v1, v2) at the first portion of the railway track of length Lx or at the second portion of the railway track of length Lx+y.

13. The method (100) of claim 11, wherein the mechanical strain measurements performed at the two different speeds (v1, v2) of the moving train are compared to each other.

14. A system (200) for monitoring structural health of a railway track (900), the system (200) comprises: a mechanical strain-gauge sensor module (202) comprising one or more mechanical strain-gauge sensors (202A) configured to measure mechanical strain values of a train undercarriage (302); a Global Positioning System (GPS) receiver (204) configured to measure location coordinates of the train (400) when the train (400) is on the railway track (900); and a computing device (206) configured to: synchronize the measured mechanical strain values with the measured location coordinates of the train (400); compare the synchronized mechanical strain values with a reference dataset, wherein the reference dataset comprises a lower threshold limit and an upper threshold limit of mechanical strain for a plurality of different location coordinates of the train (400); generate a signal if the synchronized mechanical strain values exceed either the lower threshold limit or the upper threshold limit of mechanical strain at a given location coordinates of the train (400); and provide the generated signal to an operator (216) of the train (400) for monitoring the structural health of the railway track (900).

15. The system (200) of claim 14, wherein the pre-processing comprises applying data compression on the synchronized mechanical strain values to obtain a dataset of mechanical strain values with a comparatively smaller dimension than the reference dataset.

Images