Optimizing Composite Rocket Motor Development Using Advanced Evolutionary Algorithms

It is proposed to develop an integrated analysis/optimization toolset for the design and optimization of composite solid rocket motors for tactical weapons applications. The proposed process takes advantage of recent advances in publicly availableintegration/optimization environments to provide the most capable, robust, and supportable/ maintainable system possible. A flexible optimization approach is used, in which various optimization techniques will be available, and can be selected by the userto allow tailoring to the specific problem at hand. The methods available will include Evolutionary Algorithms as well as gradient based methods, and facilities will be provided to combine methods in sequential or hierarchical combinations to obtain thebest optimization results possible with limited computational resources. Through development of a standardized Application Programming Interface (API) for each Module, the proposed process is inherently flexible (i.e. the optimization goals, constraints,and algorithms can easily be modified), extensible (i.e. additional modules can be easily added), and maintainable (i.e. modules can be updated or replaced with minimal impact). A variable fidelity approach to the analysis/optimization approach isproposed in which simplified models are used for preliminary design space screening to limit computational requirements, and higher fidelity models are used for location of the final optimum. The proposed process will directly result in higher performancepropulsion systems for tactical weapons systems, including reduced structural weight, increased range and payload, increased durability and reliability, and reduced acquisition cost. Through applying advanced analysis and optimization techniques andinter-disciplinary interactions earlier in the design cycle, the inevitable