Track and Ride Condition Correlation for Optimised Rail Maintenance

通过测量列车车体和走行部的加速度数据，定期、经济有效地收集准确的轨道几何和状态信息，仍然是基于状态维修(CBM)方法进行轨道维修的关键障碍。采用CBM方法每年可节省铁路维修费用100m；轨道和车辆维修费用昂贵，依赖于经常造成延误的定期和反应性维修。在19/20年度，网络铁路(NR)每年花费13.95亿GB用于轨道维护；轨道故障限速导致2460,453分钟延误。MoniRail的解决方案直接应对这一挑战，是一种可改装的低成本传感器系统，将在8个月期间安装两列欧洲之星列车和从运行中的列车收集的数据，并使用数据分析来寻找数据中的相关性和/或趋势，以确定正在发生的轨道和车辆故障，并确定导致路况不佳的根本路线原因。预测算法的开发将为工程师提供故障的早期发现和预测，以实现更有效的维护计划。该解决方案可能会花费数百万美元的维护成本，并减少乘客的延误。

Regular, cost-effective collection of accurate track geometry and condition information by measurement of acceleration data from the train body and running-gear remains a key barrier to condition-based maintenance (CBM) approaches for rail-track maintenance. Adoption of CBM approaches could save \>£100M/yr in rail maintenance costs.Rail-track and vehicle maintenance is costly relying on scheduled and reactive maintenance often causing delays. In 19/20 Network Rail (NR) spent £1.395Bn/yr on track maintenance; and track fault speed-restrictions caused 2,460,453 delay minutes.NR use 7 inspection trains, each costing \>£8m to build and £5.5m/yr to operate, track is inspected every 4-26 weeks, insufficient frequency for CBM approaches.MoniRail's solution directly addresses the challenge and is a retrofittable, low-cost sensor systems will be fitted two Eurostar trains and data collected during an 8 month period from the trains running in-service and using data-analytics seek correlation and/or trends in the data to identify developing track and vehicle faults and identify underlying route causes for rough-rides. Development of predictive algorithms will provide engineers with early sight and prediction of failure to enable more effective planning of maintenance. The solution has the potential millions in maintenance costs and reduce delays for passengers.