Target UAV conceptual design optimization using multi-objective genetic algorithm

Abstract

In particular, target UAVs, which simulate enemy threats for training, detection, tracking, and weapon testing, modern military operations rely much on unmanned aerial vehicles (UAVs). Among the performance criteria these vehicles must meet are high maneuverability, exact speed thresholds, and unique radar cross sections. The thesis addresses the conceptual design optimization of target UAVs with the specific objective of maximizing their operational effectiveness by meeting important performance criteria, including a cruise speed of Mach 0.6, a minimum flight endurance of 30 minutes, the capacity to support 6 g sustained and 9 g instantaneous maneuvering loads, and reliable flight at altitudes at 15,000 feet. In the context of aircraft design, which inevitably involves multidisciplinary factors in the areas of aerodynamics, propulsion, structures, and control, the exact identification of geometric and performance parameters during the conceptual design phase is crucial. The design variables define the size and position of the wing, the horizontal tail, the vertical tail, and the elevator. AMT Nike gas turbine engine for the propulsion system; At full power, it generates roughly 784 N of thrust and has a lightweight construction that fits high-speed target UAV operations. Using historical data and similar systems to estimate the mass properties of the UAV allowed a realistic mass estimation to be maintained throughout the conceptual design stage. This enables the accurate prediction of stability metrics and aerodynamic performance, which are absolutely essential for the consequent optimization process. The aerodynamic parameters were computed using USAF Digital DATCOM software, which gave stability and performance data to guide the optimization process. Typical target UAVs have an appropriate operational envelope given the reliable accuracy of digital DATCOM in the subsonic flight range. Its simplicity of use and computational efficiency make it perfect for iterative optimization processes and parametric studies. The aerodynamic data obtained from DATCOM is judged to be sufficiently accurate for conceptual design applications, where effective design space exploration depends on fast evaluation of many configurations. MOGA leads the optimization framework. This evolutionary algorithm searches a large design space by simulating the natural genetic processes of cross-over, mutation, and selection. Its primary strength is its ability to manage multiple, often conflicting objectives, in this case, minimizing trim drag and maximizing sustainable normal load factor while minimizing the elevator hinge moment sustainable elevator control effectiveness. The method discovers the optimal trade-offs by building a Pareto front, whereby no design concurrently meets all of its objectives. This helps designers to base their decisions on performance criteria specific to the mission. The constraints are determined to guarantee the stability, performance, and feasibility of the candidate solutions. In addition, design variables include the geometric characteristics of the wings and control surfaces (span, aspect/taper ratios, positions). The study shows that in the conceptual design of UAVs, the use of MOGA in the context of MDO presents an efficient approach to achieve the balance between stability and performance. By combining aerodynamic modeling with optimization strategies, the method presented in this thesis provides a quick way to conceptual design of target UAVs. The suggested method allows designers to satisfy demanding mission-specific criteria and yet produce design solutions with aerodynamic stability. The work also highlights how well digital analysis tools and optimization algorithms manage the complex needs of conceptual aircraft design.