Design optimization of a space launch vehicle using a genetic algorithm

This paper describes an effort to optimize the design of an entire space launch vehicle to low Earth (circular) orbit, consisting of multiple stages using a genetic algorithm with the goal of minimizing vehicle weight and ultimately vehicle cost. The entire launch vehicle system is analyzed using various multistage configurations to reach low Earth orbit. Specifically, three- and four-stage solid propellant vehicles have been analyzed. The vehicle performance modeling requires that analysis from four separate disciplines be integrated into the design optimization process. The disciplines of propulsion characteristics, aerodynamics, mass properties, and flight dynamics have been integrated to produce a high-fidelity system model of the entire vehicle. In addition, the system model has been validated using the existing launch vehicle data. The cost model is mass based and uses extensive historical data to produce a cost estimating relationship for a solid propellant vehicle. For the design optimization, the goal is for the genetic algorithm to minimize the differences between the desired and actual orbital parameters. This ensures that the payload achieves the desired orbit. One final goal is to minimize the overall vehicle mass, thus minimizing the system cost per launch. This paper will represent the first effort of its kind to minimize the solid propellant launch vehicle cost at the preliminary design level using a genetic algorithm.