抄録
We focused on a decomposit type optimization method that makes it easier to solve individual optimization problems by dividing the optimization problem into smaller optimization problems in multiple design spaces. Then, based on the parametric relationship of the product system model, a hierarchical optimization method to divide the optimization problem into optimizable optimization problems that can definitely define the objective function was studied. As a result, we focused on the dependency relationship between the design variables of the optimization problem and the objective function, and constructed an algorithm that divides and hierarchizes the optimization problem into an optimizable problem, which can be definitely defined by the objective function. In addition, we have constructed an automatic division and hierarchization method that automatically realizes division and hierarchization based on division / hierarchical algorithm, and realized a practical automatic hierarchical optimization method without user intervention.
