Base pressure control through micro jets at supersonic mach numbers using experimental and machine learning approach

Sudden expansion phenomena are prevalent in defense and automotive applications, where flow separation at the blunt base of structures such as fuselages, missiles, and rockets leads to low-pressure recirculation zones, significantly reducing base pressure and increasing drag. This study presents active control methods using microjets to regulate base pressure, employing experimental and machine learning approaches. Experiments were conducted using duct diameters of 16 mm, 18 mm, 22 mm, and 25 mm, level of expansion, the Nozzle pressure ratio ranging from 3 to 11, Mach numbers (1.25, 1.3, 1.48, 1.6, 2.0, and 3.0), and length-to-diameter ratios (10–1) were varied to evaluate their impact on flow evolution and base pressure. Active control was achieved using micro-jets of 0.5 mm radius, positioned at 90° intervals along a pitch circle with a radius of 0.65 times the nozzle exit diameter. Micro-jets significantly increased base pressure under favorable pressure-gradient conditions for the Mach numbers 1.25, 1.3, 1.48, 1.6, and 2.0. At Mach M = 3, the control is ineffective as the NPRs are such that the flow from the nozzle remained over-expanded. Furthermore, machine learning (ML) algorithms were utilized to predict base pressure outcomes and optimize control strategies. These algorithms demonstrated high predictive accuracy, as evidenced by low error rates, indicating their reliability in high-speed flow-control applications. The findings reveal that base pressure is strongly influenced by nozzle pressure ratio, Mach number, L/D ratio, and duct area ratio. The study presents cost-effective, energy-efficient methods to enhance base pressure, offering critical insights into the aerodynamic optimization of high-speed systems. This comprehensive approach integrates experimental techniques and ML–based predictions to achieve optimal results in flow control.