V. R. Sobol - Post-graduate Student, Moscow Aviation Institute (National Research University). E-mail: vitsobol@mail.ru

For purposes of tracking space objects in optical band it is possible to use optico-mechanical devices (OMD), placed inside a fuselage of an aircraft. We consider that device is installed on the platform inside the fuselage and special cowling with a bay protects it from the direct airflow. Due to the airflows, wicking through the bay, turbulent flows emerge inside the cowling, causing pressure variation and vibrations of the device. In some cases these vibrations can cause degradation of accuracy of the device, or even destroy it. This article reports on the method of simulation modelling of processes under investigation, considering the results of full scale tests. For description of influence of turbulent flows on OMD stochastic and gas-dynamic models are used. Using gas-dynamic model it is possible to model variation of basic airflow characteristics at any point inside the cowling. But numerical integration of gas-dynamic equations consumes a lot of computational power and time, so this approach is not flexible and cannot provide immediate results. Nevertheless, gas-dynamic model can be used at the stage of data preparation for stochastic model identification. Parameters of gas-dynamic model should be calibrated considering the results of full-scale tests which provide us with information about pressure variation and vibration from sensors' locations. For effective use of this data in future, it is necessary to find optimal sensors layout, providing most citing data. To provide recommendation for sensors layout we use stochastic model instead of gas-dynamic, so the influence of an airflow on the device will be described using stochastic processes. This approach has lowest computational complexity but requires results validation. The way to check whether stochastic model results are valid is simulating the influence airflow has on arbitrary point inside the cowling using the data provided by sensors. All the result are compared to those provided with gas-dynamic model. At the points, where it is impossible to obtain results corresponding with the results of gas-dynamic modeling, sensors should be installed before the next full-scale test. To solve the problem of selection of sensors layout we use stochastic model, calibrating parameters and comparing the results to those ones provided by gas-dynamic model simulation.