An improved fast-converging collaborative optimization method of multidisciplinary design optimization

In order to reduce the computation cost of collaborative optimization (CO) method and increase its convergence speed, an improved fast-converging collaborative optimization (IFCCO) method of multidisciplinary design optimization (MDO) for complex engineering system was proposed. First, all variables of system were divided into public and private design variables. Constraints involving public design variables were added to system level model, therefore, the system level optimized the problem target by changing public design variables and the discipline level optimized its problem target by changing its private design variables. Then, the consistency constraints were constructed in the two levels to ensure the solving accuracy of coupling variables based on discipline state equations, and each discipline was endowed with full autonomy. Finally, two typical MDO examples were used to verify the computing characteristic of IFCCO. Results show that this method needs much less computation cost and has much faster convergence speed than CO method.