Flush air data sensing system real-time solving design and verification for supersonic vehicle

In view of the head shape of a typical supersonic aircraft, the array pressure data of supersonic aircrafts is calculated by computational fluid dynamics method. The real-time solution scheme for algorithm model of the flush air data sensing system is designed for the flight Mach umber range 2.0~4.5 based on BP neural network technique and FPGA+DSP architecture digital signal processing. The design results of the algorithm model are analyzed. The influence of the total error on the algorithm model is investigated by the Monte Carlo method, and the total error of the measurement system is obtained. The algorithm needs no more than 1 millisecond for one time solution in solving machine, which can meet the requirements of the flush air data sensing system for algorithm real-time solving. The test of real-time calculation of the principled sample machine is in 1.2 m×1.2 m supersonic wind tunnel. Test results are in good agreement with the measurement results of the wind tunnel system. The system is working without any fatal error in the whole test. It can reflect the variations of the flow parameters, and has good robustness and agility. The relative error of static pressure measurement is ≤ 6.9%, the Mach number measurement < 0.1, the angle of attack and the side slip angle measurement < 1°. The influence factors of measurement error are analyzed, and improvement method of the experiment is presented at last.