Although the H∞ control method is powerful tool to design robust controller, it is difficult to attain the performances in time domain. A new design method of H∞ state feedback controller is proposed which guarantees the robust stability and assigns eigenstructure in the desired regions. The time responses can be shaped with assigning appropriate eigenstructure. Conditions of robust stability and closed-loop eigenstructure assignments are solved in LMI form with one nonlinear constraint. This method is validated with numerical example of flight control design.
Direct Arc Injection Method (DAIM), which is our original and takes the form of discharging dc arcs directly into the combustion chamber, has been investigated on its ignition capability in supersonic airflows. In this study, the ignition performance of the DAIM was evaluated by comparing experimentally with the plasma torch method, one of the most common forced ignition methods for supersonic combustion, for both the upstream and downstream fuel injection cases. The results indicated that DAIM was superior to the plasma torch method on ignition: in downstream fuel injection case, DAIM ignited methane with about half electrical energy input as much as plasma torch needed possibly because DAIM maintained the heat and radicals active longer near downstream side of the ignitor.
The present paper studies behaviors of blade-stiffened CFRP panels under axial compression by using a finite element method. Effect of a debonding between skin panel and stiffener flange on compressive behavior is analyzed and discussed. The debonding is supposed an impact damage. Linear buckling analysis and non-linear post-buckling analysis are conducted. At the debonded area, contact condition is approximately solved by an introduction of a spring element which has a resistant force only in the compressive direction. The some pre-buckling deflection increased proportional to the compression load before buckling owing to the asymmetric lamination at the flange portion, while the predicted linear buckling load agrees well with that obtained from the nonlinear analysis. The results well explain the experimental findings including the little reduction of compressive performance of the stiffened panel due to impact damage. The effect of the partial debonding on the compressive behavior becomes significant when the debonding area reaches the size comparable to that of the wave length of the buckling mode.
A winged space reentry vehicle is under development in Japan. The vehicle will be launched atop of a rocket and fly back to the Earth. Flutter characteristics of this vehicle should be considered differently between launching and free-flight. Two kinds of supporting system were newly developed to put each configuration into practice for wind tunnel tests. They possess a function to control flutter during the tests. Flutter experiments were conducted by using these supporting systems in TWT. As the results, the anticipated flutter occurred under each configuration and the effectiveness of the supporting system has been demonstrated. Moreover, the flutter characteristics could be explained clearly by the numerical analyses with the non-planar DPM (Doublet-Point Method).
Failures of acceleration in thermally chocked ram accelerator are investigated using visualization by shadowgraph pictures. It is made clear in the present paper that not only the support from thermal choking pressure but also the flame holding in the boundary layer on the projectile surface has very important role for holding the combustion on the projectile afterbody. In the lower projectile velocity case, combustion behind the projectile is not so strong and falls off despite the good surface flame holding. Besides lower velocity also makes thermal choking easily at the projectile shoulder and wave unstart occurs. When the projectile velocity goes up to the 90% of CJ detonation velocity of the mixture, wave unstart occurs because combustion becomes too strong and thermal choking occurs at the projectile shoulder. If the heat release of the mixture is made lower to avoid this wave unstart, combustion region tends to falls off.
Induced velocity fields around a rotor which is subjected to uniform velocity with arbitrary attitude angles are estimated using measured blade flapping motion and system identification technique. The blade flapping motions are detected by blade mounted accelerometers and potentiometer so that blade flexibility effects on its aero-mechanical behavior could be captured. The measured data reconstructed by the sensor equation are fed into the kinematic observer for flapping rate estimation and then parameters which describe induced velocity distribution are identified by the recursive least-square method. Induced velocity distributions for a tilting rotor are modeled by interpolating these identified parameters. Comparisons between calculated and measured blade flapping responses for a tilting rotor are made. It is understood that identification processes are quite smooth and usefulness of the proposed identification procedure are ascertained.