Graph-based reinforcement learning for discrete cross-section optimization of planar steel frames
نویسندگان
چکیده
A combined method of graph embedding (GE) and reinforcement learning (RL) is developed for discrete cross-section optimization planar steel frames, in which the section size each member selected from a prescribed list standard sections. The RL agent aims to minimize total structural volume under various practical constraints. GE extracting features data with irregular connectivity. While most existing methods aim at node features, an improved formulation edges associated members this study. Owing proposed operations, capable grasping property columns beams considering their connectivity frame arbitrary as feature vectors same size. Using vectors, trained estimate accurate return action take proper actions on reduce or increase using algorithm. applicability versatile because frames different numbers nodes can be used both training application phases. In numerical examples, agents outperform particle swarm benchmark terms computational cost design quality cross-sectional changes; successfully assign reasonable cross-sections geometry, connectivity, support load conditions frames.
منابع مشابه
Bat inspired algorithm for discrete size optimization of steel frames
Bat inspired (BI) algorithm is a recently developed metaheuristic optimization technique inspired by echolocation behavior of bats. In this study, the BI algorithm is examined in the context of discrete size optimization of steel frames designed for minimum weight. In the optimum design problem frame members are selected from available set of steel sections for producing practically acceptable ...
متن کاملPERFORMANCE-BASED DESIGN OPTIMIZATION OF STEEL MOMENT FRAMES
This study deals with performance-based design optimization (PBDO) of steel moment frames employing four different metaheuristics consisting of genetic algorithm (GA), ant colony optimization (ACO), harmony search (HS), and particle swarm optimization (PSO). In order to evaluate the seismic capacity of the structures, nonlinear pushover analysis is conducted (PBDO). This method is an iterative ...
متن کاملlayout optimization for x-bracing of planar steel frames using ant system
in this paper, the problem of layout optimization for x-bracing of steel frames is studied using the ant system (as). a new design method is employed to share the gravity and the lateral loads between the main frame and the bracings according to the requirements of the ibc2006 code. an algorithm is developed which is called optimum steel designer (osd). an optimization method based on an approx...
متن کاملA Discrete Hybrid Teaching-Learning-Based Optimization algorithm for optimization of space trusses
In this study, to enhance the optimization process, especially in the structural engineering field two well-known algorithms are merged together in order to achieve an improved hybrid algorithm. These two algorithms are Teaching-Learning Based Optimization (TLBO) and Harmony Search (HS) which have been used by most researchers in varied fields of science. The hybridized algorithm is called A Di...
متن کاملMinlp Optimization of Steel Frames
ABSTRACT: The paper presents the discrete dimension optimization of unbraced rigid steel plane frames. The optimization of steel frames was carried out by the Mixed-Integer Non-linear Programming (MINLP) approach. The MINLP is a combined discrete-continuous optimization technique. It performs the discrete optimization of discrete decisions simultaneously with the continuous optimization of cont...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Advanced Engineering Informatics
سال: 2022
ISSN: ['1474-0346', '1873-5320']
DOI: https://doi.org/10.1016/j.aei.2021.101512