Conceptual design and computational development of thrust vectoring system for UAV application

Abstract

Manoeuvrability is a desired and important attribute of an aircraft (both manned and unmanned) if it must fit into the multi-purpose demand of our present-day need. An aircraft which can fit into this must have the capability of taking off at any given /available space and thus quickly get into the air as soon as possible. This plays an important role in aircraft total performance and endurance. With the challenges of limited space for take-off, an aircraft with short take-off ability is highly desired. Several works had been done to achieve short take-off and landing (STOL) but mainly on jet engines with little on pusher propeller aircraft. This propulsion system is gaining more relevance in drone technology due to its operational advantage for short and medium-range purposes like security, surveillance, parcel delivery, medical emergency response, and search and rescue operations. Even though UAVs (Unmanned Aerial Vehicles) with this type of propulsion system can achieve the above-mentioned mission applications, they often fall short due to the need for long take-off distances and therefore long runways. To tackle this problem, and improve the operational capabilities of Guardian II UAV, a CPUT AMTL (Advance Manufacturing Laboratory Technology) demonstration platform; this research project designed and computationally investigated the means of improving the manoeuvrability and making it operational in the condition mentioned above. One of the ways in which to reduce the ground role (Sg), and to increase the rate of climb (ROC) of an aircraft is through the introduction of short take-off and landing configurations. These flight performance parameters are primarily governed by the dimensionless coefficients for lift, drag and pitching moment. Literature shows a variety of these devices and their effectiveness in reducing the ground role and increasing rate of climb. Using a design trade-off table with the design specification and requirement of integrating Thrust Vectoring Configuration (TVC) system without major changes in the airframe in view; a shrouded configuration with jet vanes to direct the airflow from a pusher propeller UAV was selected as best suited for this application. The iterative selection choice was governed by these selection criteria namely weight penalty, ease of integration, and suitability. An experimental investigation using a shroud was conducted to determine the velocity profile of air exiting a pusher propeller configuration as used in the Guardian II UAV. Velocity profiles were obtained for speeds from 2000rpm to 4000rpm, and subsequently, mass flow rates of 11.173kg/s to 23.329kg/s respectively were calculated. The characterization experimental set-up is a test bench equipped with Turnigy C6374 – 200 brushless outrunner electric motor driving a 22 x 10inch three-bladed propeller, speed control device, Pitot tube, manometer and a data logging device. SolidWorks® 2017 was used to design a shroud with incorporated jet vanes to direct airflow from the pusher propeller; while ANSYS Fluent 2019R was used in the selection and characterization of the jet vanes. Due to the computation resources available, the Reynolds-average Navier-Stokes (RANS) turbulent model was employed to predict the flow parameters of interest. Specifically, three vane configurations were investigated, and results showed that the NACA0012 airfoil was best suited as a jet vane for this application due to its performance showing zero lift at zero deflection angle as desired. FLUENT was then used to analyse the entire thrust vectoring configuration, i.e. airflows from the propeller through the shroud with vanes deflected through 00 to 50, and flap angles of 0o, 5o, 10o and 15o to determine the dimensionless coefficients for lift, drag, and pitching moment. The computational domain of length 4800 mm and diameter 495 mm respectively were used as the virtual wind tunnel. The inlet (upstream of the model), outlet (downstream) and sides (walls) boundaries were located at 5L, 10L and 5D of the model respectively. These boundaries represented the propeller freestream airflowing onto the vanes, while the airflows exiting the shroud was acted upon by the vanes and the surrounding respectively. The domain was discretized into 3.8 million elements and inflation layers created around the vanes to predict the flow parameters of interest. A good and acceptable mesh quality was obtained using the localized meshing method. The boundary conditions used for the system characterization to obtain the flow parameters of interest are mass flow inlet and pressure outlet. The mass flowrate obtained from the preliminary experiment was set as the inlet boundary conditions at the speed, and the vane deflection angles were analysed, while the pressure at the exit was predicted in the analysis. This CFD set-up approximated/predicted the coefficients of drag, lift, and pitching moment when the vanes were deflected at the angles considered. The approximated flight performance parameters were used to calculate the Sg and ROC. The result showed reductions of Sg and increase in the ROC for the different thrust vectoring configurations investigated. The stable performance of the system was obtained at propeller speed from 3000 to 4000 rpm and vane deflection angles of 3o to 5o. At 4o vane deflection angle, a minimum and maximum reduction in GII Sg obtained were 48.12% at 3000 rpm and flaps angle 0o, and 64.98% at 4000 rpm and flaps angle 15o respectively. For the ROC, at 3o and 5o vane deflection angle and flap angle of 5o, 31.28% at 3500 rpm and 31.33% at 4000 rpm respectively, the minimum and maximum improvement in the climbing flight parameter were obtained. The performance was determined by the reduction in power usage by a minimum of 62.37%, and a maximum of 85.53% across the stable and optimal configurations analysed. A minimum of 59.89% at conditions of 2000 rpm, 1o deflection angle, and 0o flaps angle had a corresponding maximum reduction value of 62.36% at the same operational conditions. An optimal power usage which increases with flaps angle was recorded at 4o deflection angle and from 2000 rpm to 4000 rpm. The result is evident; however, it should be noted that the percentage reduction in Sg, rate of climb and power is expected to decrease when the TVC system is built and integrated onto the airframe of the GII UAV. The thrust vectoring system decreased the Sg by a maximum of 65.19% and improved the ROC by a maximum of 31.33%. This proves that the set objectives for this research project were met.