Rail Track Gradient Modeling for Safe Guided Vehicle Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining track gradient values in railway networks are either expensive, inaccurate, or require conservative estimates, leading to decreased performance and safety of guided vehicles.
Innovation Solution
A system and method that utilizes multiple sources of gradient data (construction, GPS, Lidar, and on-site measurements) to split railway tracks into portions, calculate average gradient values, and compare them to determine accurate sub-portion gradient values, generating alerts or using restrictive values when necessary.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conservative gradient values are used for unmeasured track portions, then safety is improved, but exploitation performance decreases
Solution Approach 1:
The track is divided into multiple portions, each with its own gradient value determination. This allows different portions to have different levels of measurement detail, enabling safety-critical sections to use measured values while less critical sections can use conservative estimates, thus balancing safety and performance
Solution Approach 2:
Different gradient value determination methods are applied to different track portions based on local requirements. High-speed or high-risk sections use accurately measured gradient values, while other sections may use conservative estimates, optimizing the balance between safety and exploitation performance locally rather than uniformly across the entire network
2Measurement precision
If on-site gradient measurement tests are performed, then measurement precision is improved, but cost and time increase
Solution Approach 1:
Gradient measurements are performed in advance during track construction or maintenance activities, and the results are stored for later use. This preliminary action eliminates the need for repeated measurements, reducing both time and cost while maintaining high measurement precision when needed
Solution Approach 2:
Instead of performing physical on-site measurements repeatedly, the patent uses stored gradient measurement data from previous tests as a copy. This allows the system to access accurate gradient information without incurring the time and cost of new measurements, while still maintaining measurement precision through the use of previously captured accurate data
3Reliability
If conservative gradient values are used, then safety factor is improved, but authorized vehicle speed decreases
Solution Approach 1:
The system applies different gradient value strategies to different track portions based on local characteristics. Safety-critical sections use conservative gradient values to ensure adequate safety factors, while sections with accurately measured gradient values allow for optimized vehicle speeds that reflect the actual track conditions, thus maintaining safety factors where needed while preserving speed where possible
Data Source
Figure 1~3
Figure 4
AI summary
The invention relates to a system (100) and a method (400) for modeling a gradient of a track (111) of a railway network (110), said method (400) comprising: - splitting (401) the railway network track (111) in N portions P1, P2,...,PN of track, wherein each portion Pk of track starts at a position pk and ends at a position p'k that is located within a directly next portion of track PI ; - calculating (402) for each Pk, an average gradient value AG1,k obtained from gradient values of a first source S1, and an average gradient value AG2,k obtained from gradient values of a second source S2; - for each Pk, comparing (404) AG1,k to AG2,k, and if a difference between their values is smaller than a predefined reference value, then assigning to a sub-portion of said track the value of AG1,k or of AG2,k or a mean value obtained from AG1,k and AG2,k; otherwise if said difference is greater than said predefined reference value, then using a third source S3 of gradient values for calculating for Pk an average gradient value AG3,k, and comparing the latter to at least one of the values AG1,k and AG2,k in order to determine if their difference is smaller than said predefined reference value, and, in the affirmative, assigning to said sub-portion of track a gradient value that is the average gradient value of said at least one of the values AG1,k and AG2,k, or the value of AG3,k, or a mean value obtained from their combination; in the negative, comparing AG3,k to the other one of the values AG1,k and AG2,k in order to determine if their difference is smaller than said predefined reference value, and, in the affirmative, assigning to said sub-portion a gradient value that is the average gradient value of said other one of the values AG1,k and AG2,k, or AG3,k, or a mean value obtained from their combination; and if both the difference between AG3,k and AG1,k and the difference between AG3,k and AG2,k are greater than said predefined reference value, then generating an alert and/or using a restrictive gradient value for said sub-portion.