Railway Track & Structures - November 2007 - (Page 34)

Automated m/w planning Figure 3, above, is a summary of annual rail wear replacement, 2008-2012. Figure 4, right, is a Weibull fatigue life projection. TECH Associates, Inc., a Harsco Company. “This is particularly true for the planning of rail relay or replacement programs, which require accurate and timely information about the condition of the rail, to include wear- and fatiguerelated information. “With the growing use of rail wear- and profilemeasurement systems to collect wear data and the concurrent extensive use of ultrasonic test cars to collect fatigue data, the information needed to more-effectively plan rail relay programs is becoming more readily available,” he said. “With this capability, planning of 3-, 5-, and even 10-year rail programs becomes possible, with realistic predictions of rail requirements. “The planning tools that are now available for management of rail renewals rely on the availability of this type of information to make good decisions,” Zarembski said. “One recent application of this rail renewal planning approach, using current rail wear and fatigue data, was on a high-density region of a major international railroad, where the ZETA-TECH RailLife rail wear- and fatigue planningmodel was used to determine short- and medium-term rail relay requirements. This territory had regular inspection with a laser profile measurement system, as well as conventional ultrasonic testing for rail defects. Both wear and fatigue predictions were performed for a five-year planning horizon.” In order to implement such a planning program, the track is divided into discrete rail segments based on changes in the characteristics of the rail, track, traffic, and maintenance history. These segments are assumed to be homogeneous, meaning that the rail life, as well as certain key parameters, will be the same or similar for the entire segment. Thus, each segment contains enough engineering information to form the basis for proper rail replacement decisions. Using rail wear data, taken from laser-based profile measurement systems, allows for a statistically-based rail wear loss rate analysis and associated prediction of future rail wear and rail replacement points. ZETA-TECH’s RailLife wear analysis uses a multivariate regression analysis of the rail wear data, using all available rail wear inspection data. This is illustrated in Figure 1, which shows a good statistical regression analysis (R2 = 90 percent) allowing for an accurate projection of wear rates and corresponding wear life. Figure 2 shows the same type of statistical projection, for a highwear-rate environment such as a sharp curve with inadequate lubrication. The RailLife wear model supplements this statistical analysis with an engineering-based rail wear model. This 34 Railway Track & Structures November 2007 engineering model allows for the forecasting of rail replacement dates, even when the rail wear rates are not sufficiently large to permit good rail wear inspection measurement data. It also allows for the identification of high-wear-rate segments for further investigation as to the cause of the high wear, e.g., inoperative lubricators. As can be seen in Figure 3, as much as 60 percent of the actual segments require this engineering model supplement, because full measured wear data is not always available for all track segments. Table 1 presents a corresponding five-year forecast for rail segments due to be replaced due to wear, based on the RailLife wear model. In a similar manner, rail fatigue condition is analyzed using rail defect data, including ultrasonic test and service defects, as well as tonnage and related track information. The RailLife fatigue analysis utilizes a statistical Weibull analysis (Figure 4) that uses the history of reported rail defects. Thus, for the analysis presented here, 15 years of defect data was used to calculate the projected fatigue rate. That, combined with the railway defined fatigue limit and the current condition of the rail segment, is used to estimate the required rail replacement date for the fatigue mode. The RailLife algorithms allow for the optimization of track segment lengths so as to have available a sufficient data set for the Weibull analysis. In addition, the algorithms utilize a “hierarchy” of analysis approaches, directly related to the actual amount of data available. Thus, if a sufficient level of data is available by segment, then a full conventional Weibull analysis is performed. If a smaller amount of data is available, then a defect rate analysis is performed based on the rate of change (first derivative) of the Weibull function. If even less data are available, then alternate approaches are implemented which include default values for one or both of the Weibull parameters (a, ß), or using an engineering model approach based on limited data. Table 2 presents a corresponding five-year forecast for rail segments to be replaced due to fatigue, based on the RailLife fatigue model. “The result is a comprehensive multi-year rail replacement plan that encompasses both wear and fatiguebased degradation and replacement modes,” Zarembski said. “Thus, these new-generation tools allow for an accurate planning of future rail replacement needs, budget requirements, and associated support resources to allow a railroad to effectively manage its expensive track assets.” www.rtands.com http://www.rtands.com

Table of Contents Feed for the Digital Edition of Railway Track & Structures - November 2007

Railway Track & Structures - November 2007 Contents On Track Industry Today Supplier News AREMA News NRC News TTCI R&D Seattle Retrofits Downtown Transit Tunnel Switch Stand, Switch Machines Planning M/W with Modern, High-Tech Tools Products & Literature People Calendar Advertisers Index Sales Representatives Website Directory Professional Directory Classified Advertising Chicago Perspective

Railway Track & Structures - November 2007

For optimal viewing of this digital publication, please enable JavaScript and then refresh the page. If you would like to try to load the digital publication without using Flash Player detection, please click here.