Abstract
A system optimization framework based on design of experiment (DOE) and simulation is proposed. Simulation is effective in the optimization of many complex systems, however, it is time consuming and tough. DOE can reduce the number of test cases drastically while ensuring input combination is of wise coverage of design space. A framework incorporating simulation and DOE is proposed, which provides a methodology to deal with optimization problem. It contains two key technologies: Latin hypercube design and surrogate model optimization, which are all detailed in paper. A problem of micro satellite system optimization is proposed to illustrate the correctness and effectiveness of the framework. The final results are listed. Also effect of each design variable on the system performance is analyzed.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
