Method and system for monitoring a point system of a railway network, and point system of a railway network

Abstract

Method for monitoring at least one point system of a railway network comprising the steps of placing several sensors in correspondence of said at least one point system; acquiring from said sensors current and voltage signals of a point machine during a maneuver; and segmenting the signals of the maneuver according to different predetermined phases of movement. Then extracting predetermined features from each segment; comparing the extracted features with a set of predetermined values which represent a “healthy” maneuver, thus obtaining a global indicator representative of the conditions of the maneuver at the point system; and comparing said global indicator with a failure threshold, and if it exceeds said failure threshold, detecting a failure in the point system.

1 . Method for monitoring at least one point system of a railway network comprising the steps of:
 placing several sensors in correspondence of said at least one point system;   acquiring from said sensors current and voltage signals of a point machine during a maneuver;   segmenting the signals of the maneuver according to different predetermined phases of movement;   extracting predetermined features from each segment;   comparing the extracted features with a set of predetermined values which represent a “healthy” maneuver, thus obtaining a global indicator representative of the conditions of the maneuver at the point system; and   comparing said global indicator with a failure threshold, and if it exceeds said failure threshold, detecting a failure in the point system.   
     
     
         2 . The method according to  claim 1 , further comprising:
 determining library of vectors relative to degradation mechanisms, replicating multiple degradation patterns of the point system;   concatenating the values of the extracted features into a vector, said values representing the effect of a degradation of the point system; and   comparing the vector with the library of vectors to identify the degradation mechanism.   
     
     
         3 . The method according to  claim 1 , wherein the global indicator corresponds to an aggregation of values obtaining through the comparison of the extracted features with the set of predetermined values. 
     
     
         4 . The method according to  claim 1 , further comprising:
 evaluating the global indicator to determine different degradation patterns with different levels, comprising but not limited to misalignment, obstacle, excessive force.   
     
     
         5 . The method according to  claim 1 , further comprising the step of concatenating the values of the extracted features into a vector, said values representing the effect of a degradation of the point system. 
     
     
         6 . The method according to  claim 1 , wherein the step of comparing comprises the steps of:
 normalizing the extracted features with respect to a set of reference features, thus obtaining corresponding indicator values; and   combining said indicators to obtain the global indicator.   
     
     
         7 . The method of  claim 6 , wherein normalizing the extracted features comprises:
 subtracting an original value taken from the signal in the segment from a predetermined mean value and dividing the result by a predetermined standard deviation value.   
     
     
         8 . The method of  claim 6 , wherein combining said indicators comprises calculating the Mahalanobis distance of said indicators. 
     
     
         9 . The method of  claim 6 , wherein combining said indicators comprises calculating a distance such as a Mahalanobis distance between subgroup of said indicators or applying a principal component analysis algorithm to subgroup of said indicators or calculating minimum quantization error from subgroup of said indicators. 
     
     
         10 . The method according to  claim 1 , wherein the extracted features include the peak value of the signal, differentials of the signal, average of the signal, mean of the signal, maximum of the signal, the slope of the signal curve. 
     
     
         11 . The method according to  claim 1 , wherein the predetermined values which represent a “healthy” maneuver are values determined through a machine learning process based on data relative to previous maneuvers, from a point system of nominal status. 
     
     
         12 . The method according to  claim 4 , wherein the step of evaluating the global indicator comprising different levels of misalignment, obstacles, excessive force. 
     
     
         13 . A monitoring system for monitoring at least one point system of a railway network comprising a point system of a railway network, a plurality of sensors placed in correspondence with said point system and an elaboration unit connected to said sensors and to the point system and arranged to carry out the method according to  claim 1 . 
     
     
         14 . A point system of a railway network including a plurality of sensors arranged to be connected to an elaboration unit arranged to carry out the method according to  claim 1 .