Oxygen concentration variation in ullage influenced by dissolved oxygen evolution

To determine the oxygen concentration variation in ullage that results from dissolved oxygen evolution in an inert aircraft fuel tank,the CFD method with a mass transfer source is applied in the present study.An experimental system is also designed to evaluate the accuracy of the CFD simulations.The dissolved oxygen evolution is simulated under different conditions of fuel load and initial oxygen concentration in ullage of an inert fuel tank with stimulations of heating and pressure decrease.The increase in the oxygen concentration in ullage ranges from 0.82%to 5.92%upon stimulation of heating and from 0.735%to 12.36%upon stimulation of a pressure decrease for an inert ullage in the simulations.The heating accelerates the release of the dissolved oxygen from the fuel by increasing the mass transfer rate in the mass transfer source and decreasing the pressure,thereby accelerating the dissolved oxygen evolution by increasing the concentration difference between the gas and the fuel.The time constant that represents the oxygen evolution rate is independent of the initial oxygen concentration in ullage of an inert tank but depends closely on the fuel load,temperature and pressure.The time constant can be fitted using a polynomial equation relating the fuel load to temperature in the heating stimulation with an accuracy of 4.77%.Upon stimulation of a pressure decrease,the time constant can be expressed in terms of the fuel load and the pressure,with an accuracy of 5.02%.

Optimal model averaging estimator for multinomial logit models

In this paper,we study optimal model averaging estimators of regression coefficients in a multinomial logit model,which is commonly used in many scientific fields.A Kullback-Leibler(KL)loss-based weight choice criterion is developed to determine averaging weights.Under some regularity conditions,we prove that the resulting model averaging estimators are asymptotically optimal.When the true model is one of the candidate models,the averaged estimators are consistent.Simulation studies suggest the superiority of the proposed method over commonly used model selection criterions,model averaging methods,as well as some other related methods in terms of the KL loss and mean squared forecast error.Finally,the website phishing data is used to illustrate the proposed method.