A system and method for monitoring a railway network

Abstract

The present invention relates to a system for monitoring at least one railway network, the system comprising: at least one sensor component configured to measure sensor data relate to the at least one railway network, at least one processing component configured to (pre)process the sensor data, at least one analyzing component configured to analyze the sensor data, and at least one interface configured to access at least one server configured to be bidirectionally connected to the system, wherein the sensor data comprises at least one data recorded during an inspection activity. The invention also relates to a method for monitoring a railway network, the method comprising: recording at least one railway network related data, pre-processing the at least one railway network related data, and generating at least one pre-processed railway network related data, wherein the pre-processing comprises structuring the at least one railway network related data and wherein the at least one railway related network comprises at least one data recorded during at least one inspection activity and/or at least one data recorded during at least one failure correction activity.

1 . A system for monitoring at least one railway network, the system comprising
 at least one sensor component configured to measure sensor data relate to the   at least one railway network,   at least one processing component configured to (pre)process the sensor data,   at least one analyzing component configured to analyze the sensor data, and   at least one interface configured to access at least one server configured to be bidirectionally connected to the system, wherein the sensor data comprises at least one data recorded during an inspection activity.   
     
     
         2 . The system according to  claim 1 , wherein the system is configured to
 correct the sensor data, and   generate a corrected railway network related dataset.   
     
     
         3 . The system according to  claim 1 , wherein the system is configured to
 generate a text form dataset,   unify at least one technical word from a plurality of dataset,   analyze the at least one dataset, and   develop at least one language-agnostic framework.   
     
     
         4 . The system according to  claim 3 , wherein the system is configured to
 collect incoming maintenance data from at least one railway network,   process the incoming maintenance data from the at least one railway network to generate at least one processed incoming maintenance dataset, and   generate at least one developing model based on the at least one processed incoming maintenance dataset, wherein the at least one developing model comprises at least one develop model with a structured data.   
     
     
         5 . The system to according to  claim 4 , wherein the system is configured to
 model the incoming maintenance data from the at least one railway network based on the at least one processed incoming maintenance dataset, and   generate at least one hypothesis based on the at least one processed incoming maintenance dataset,   wherein the at least one hypothesis comprises at least one of a maintenance recommendation and/or an action.   
     
     
         6 . The system according to  claim 1 , wherein the system is configured to identify and analyze at least one of
 a number of effective inspections,   a malfunction event, and   a trigger of a maintenance action.   
     
     
         7 . The system according to  claim 1 , wherein the system is configured to provide
 analyzed data of at least a first railway network,   analyzed data of at least a second railway network, and   compared data of the at least first railway network and the at least second railway network, wherein the at least first railway network is different from the at least second railway network, and wherein the system is configured to identify at least one data comprising similar parameters between the at least first railway network and the at least second railway network.   
     
     
         8 . A method for monitoring a railway network, the method comprising
 recording at least one railway network related data,   pre-processing the at least one railway network related data, and   generating at least one pre-processed railway network related data,   
       wherein the pre-processing comprises structuring the at least one railway network related data and wherein the at least one railway related network comprises at least one data recorded during at least one inspection activity and/or at least one data recorded during at least one failure correction activity. 
     
     
         9 . The method according to  claim 8 , wherein the method comprises
 correcting the at least one railway network related data   generating a corrected railway network related dataset,   generating a text form dataset,   unifying at least one technical word from a plurality of dataset,   analyzing the at least one dataset, and   developing at least one language-agnostic framework.   
     
     
         10 . The method according to  claim 8 , wherein the method comprises
 collecting incoming maintenance data from at least one railway network, and   processing the incoming maintenance data from the at least one railway network to generate at least one processed incoming maintenance dataset.   
     
     
         11 . The method according to  claim 10 , wherein the method comprises
 generating at least one developing model based on the at least one processed incoming maintenance dataset,   modelling the incoming maintenance data from the at least one railway network based on the at least one processed incoming maintenance dataset, and   generating at least one hypothesis based on the modelling,   
       wherein the at least one hypothesis comprises at least one of a maintenance recommendation and/or an action. 
     
     
         12 . The method according to  claim 11 , wherein the at least one hypothesis comprises at least one of
 a system recommendation,   a system question answering,   an inspection results related to at least one particular asset, and   clustering maintenance of activities to at least one particular group of events.   
     
     
         13 . The method according to  claim 8 , wherein the method comprises generating at least one report comprising at least one of
 a frequency of occurred failures,   a cause of track infrastructure deterioration,   a type of track infrastructure deterioration, and   a temporal evolution of failures comprising a defined time range.   
     
     
         14 . The method according to  claim 9 , wherein developing the framework comprises
 building at least one ingestion pipeline,   ingesting at least one data,   generating at least one ingested dataset,   processing the at least one ingested dataset,   generating at least one processed ingested dataset,   modeling data based on at least one of
 the at least one ingested dataset, and 
 the at least one processed ingested dataset, 
   storing the at least one processed ingested dataset in a database, and   retrieving data from the at least one database.   
     
     
         15 . The method according to  claim 8 , wherein the method comprises
 training at least one model to yield at least one pre-trained embedding,   applying the at least one pre-trained embedding to the at least one processed ingested dataset,   classifying the at least one processed ingested dataset into at least one category,   predicting the at least one category for the at least one processed ingested dataset based on at least one event, and   translating at least one category comprised by the at least one processed ingested dataset into at least one language, wherein the at least one language is selectively defined, and generating at least one translated dataset.