ANALYSIS DEVICE, PARAMETER OPTIMIZATION METHOD, AND PARAMETER OPTIMIZATION PROGRAM

To efficiently optimize parameters of a model analyzing structures.SOLUTION: Provided is an analysis device for analyzing optimization subject modeling, including: a low cost analysis section which sets a parameter to at least one low cost model representing an analysis model having shorter analysis...

Full description

Saved in:
Bibliographic Details
Main Authors TAKEUCHI MIZUHO, TATEISHI HIROKI
Format Patent
LanguageEnglish
Japanese
Published 26.07.2022
Subjects
Online AccessGet full text

Cover

Loading…
More Information
Summary:To efficiently optimize parameters of a model analyzing structures.SOLUTION: Provided is an analysis device for analyzing optimization subject modeling, including: a low cost analysis section which sets a parameter to at least one low cost model representing an analysis model having shorter analysis time than an optimization subject model, runs an analysis thereto in the low cost model, evaluates the parameter based on an analysis result, sets a next parameter based on an evaluation result, runs the analysis thereto in the low cost model for accumulating relations between the parameters and the analysis results; and an optimization analysis section which sets a parameter for running the analysis in the optimization subject model, runs an evaluation based on a relation between a calculated analysis result and the parameter and the information accumulated at the low cost analysis section, and repeats processing to search an optimal parameter based on an evaluation result for identifying the optimal parameter.SELECTED DRAWING: Figure 1 【課題】構造を解析するモデルのパラメータを効率よく最適化できる。【解決手段】最適化対象モデル化の解析を行う解析装置であって、最適化対象モデルよりも解析時間が短い解析モデルである少なくとも1つの低コストモデルに対して、パラメータを設定し、低コストモデルで解析を実行し、解析結果に基づいてパラメータを評価し、評価結果に基づいて次のパラメータを設定し、低コストモデルで解析を実行することで、パラメータと解析結果との関係を蓄積する低コスト解析部と、パラメータを設定して最適化対象モデルで解析を実行し、算出した解析結果とパラメータとの関係と、低コスト解析部で蓄積した情報とに基づいて、評価を実行し、評価結果に基づいて、最適パラメータを探索する処理を繰り返し、最適パラメータを特定する最適化解析部と、を含む。【選択図】図1
Bibliography:Application Number: JP20210003627