COMPUTER BASED DESIGN METHOD AND DESIGN SYSTEM

摘要

For a multiplicity of design variants (KV) of a technical product (TP), a training structural data set (SDT) specifying the particular design variant (KV) and a training quality value (QT1, QT2, QT3) quantifying a predefined design criterion (K1, K2, K3) are read in in each case as training data (TD). The training data (TD) are taken as a basis for training a Bayesian neural network (BNN) to determine an associated quality value (Q), together with an associated uncertainty comment (UC), on the basis of a structural data set (SD, SSD). Furthermore, a multiplicity of synthetic structural data sets (SSD) are generated and are fed into the trained Bayesian neural network (BNN) which generates a quality value (Q) with an associated uncertainty comment (UC) for each of the synthetic structural data sets (SSD). The uncertainty comments (UC) generated are compared with a predefined reliability comment (REL) and one of the synthetic structural data sets (SD) is selected on the basis thereof. The selected structural data set (SD) is then output for the purpose of producing the technical product (TP).