Cálculo da dispersão de pontos de impacto de foguetes não controlados

Abstract

The main goal of this work is to compare the effectiveness between Monte Carlo and RSS (Root Sum Squared) methods about the generation of a dispersion area of disturbed impact points of uncontrolled rockets, meaning, impact points that differ from nominal by influence of factors such as: trawl variation, buoyancy, mass and deviation of the buoyant vector, wind, inaccuracy of the launcher among others. The RSS method considers each disturbance factor in isolation, that is, a deviation value is specified for each of them in each simulation, while the Monte Carlo method considers a set of factors disturbed simultaneously, being varied from random use of values for the variation of the parameters. In rocket launches, one of the mission's requirements is the delimitation of a dispersion area, as it influences the safety and payback costs of the payload. In the flight trajectory simulation process the two methods generated similar dispersion ellipses. However, the number of simulations required in the Monte Carlo method is considerably higher. This work also presents the development of a computational tool, called UROCKS, that determines the behavior of the rocket in flight via the numerical system of nonlinear ordinary differential equations that represent the dynamical and kinematic mathematical models of the rocket, what makes the main result of the simulation to be the point of nominal impact, that is, without deviation, resulting from the trajectory of the uncontrolled rocket. The validation of the data obtained is performed by comparison with another software, recognized for such function, called ROSI. The results of the flight parameters, such as altitude, angular and linear velocity, Mach number and dynamic pressure obtained by the use of the aforementioned two tools were very close, with small variations due to the difference of environmental and propulsive models implemented. In general, the dispersion results are satisfactory, since they show that both methods, computationally implemented in the software DAMCURS and DAIPRSS respectively, for the Monte Carlo and RSS methods, can be used to calculate the impact point dispersion, even considering that the Monte Carlo method is a stochastic tool while RSS, is deterministic.