Impact of Fractional Order Methods on Optimized Tilt Control for Rail Vehicles

Advances in the use of fractional order calculus in control theory increasingly make their way into control applications such as in the process industry, electrical machines, mechatronics/robotics, albeit at a slower rate into control applications in automotive and railway systems. We present work on advances in high-speed rail vehicle tilt control design enabled by use of fractional order methods. Analytical problems in rail tilt control still exist especially on simplified tilt using non-precedent sensor information (rather than use of the more complex precedence (or preview) schemes). Challenges arise due to suspension dynamic interactions (due to strong coupling between roll and lateral dynamic modes) and the sensor measurement. We explore optimized PID-based non-precedent tilt control via both direct fractional-order PID design and via fractional-order based loop shaping that reduces effect of lags in the design model. The impact of fractional order design methods on tilt performance (track curve following vs ride quality) trade off is particularly emphasized. Simulation results illustrate superior benefit by utilizing fractional order-based tilt control design.