Abstract

Optimization of reinforcing bar (rebar) design represents a preponderant factor in reducing material usage and wastes for reinforced concrete (RC) structures. The assessment of constructability of such rebar designs is crucial to improve their practicality and reduce construction costs, which makes the problem multi-objective (MO). However, when applying optimization methods for the design of rebar in RC structures, little attention has been paid on columns, in comparison to beams and slabs. Meta-heuristic algorithms (MA) have been the ones mostly deployed for these types of elements, which have proven to be of high computational cost. Additionally, an existing gap in the literature as to how to relate the design and construction stage of rebar in RC structures through constructability analysis is evident. In this regard, research has been focused mainly at the building level but not at the element level. This works presents a novel algorithmic framework using Machine Learning (ML)-enhanced meta-heuristics for the optimal design of rebar on rectangular RC columns. To assess the constructability of the resulting rebar layouts a Buildability Score (BS) model at the element level is proposed. The complexity analysis of rebar design under the constructability restrictions, through combinatorial optimization (CO), is used to assess the global time efficiency of the framework. The Non-Sorting Genetic Algorithm II (NSGA-II) was deployed for showcase and five different ML algorithms were used to enhance it, namely the k-NN classifier, SVM regression, ANN, Gauss Process (GP) regression, and Tree Ensembles (TE), where the latter three showed the best performance.

Abstract

The present work exposes the design of optimization procedures both with the “Particle Swarm Optimization” (PSO) algorithm and the “Genetic Algorithm” (GA) for the design of reinforced concrete frames, making comparisons in cost, weight of the structure and predicted damage. The optimization procedures are built up using the “Idealized Smeared Reinforcement” (ISR) analogy for each element of the structural model frames considered for this work. Two different numerical structural plane-frame models were created for the application and comparison of the performance of the optimization design procedures hereby proposed. The optimization procedures were mono-objective with a cost-objective function, taking on account steel reinforcement and concrete for the cost computation. Two different design approaches were carried on for this work, one proposing asymmetrical reinforcement for columns and the other with symmetrical reinforcement. In order to compute the damage indices considered for this study a non-linear Pushover structural analysis is performed. Results show that asymmetrical reinforcement in columns may reduce concrete volumes, although such reduction in material might not be quite proportional with construction cost, given that asymmetric reinforcement in columns is more expensive than symmetrical, per unit-cost. The bigger the structure, the more likely is to obtain lighter structures by using asymmetrical reinforcement. Regarding damage of the structure, results show that when using asymmetrical reinforcement in columns, it is more likely to obtain smaller values for the expected damage with no great difference on the estimated collapse Safety Factors for the seismic loads. In general, the proposed methodology hereby proposed enhances quite good optima results, requiring only a few adjustments of clash-free and slap reinforcement after the optimization procedure terminates. When designing reinforced concrete frames with asymmetric reinforcement in columns, an increase in construction costs of as much as $25\%$ as that obtained for symmetric reinforcement could be enhanced. In general, with the proposed methodology to optimally design reinforced concrete frames, savings of as much as $20\%$ in construction costs from an initial structural proposal can be reached.

Abstract

It is presented the development and implementation of a new computed aided learning MatLab Toolbox for the design of reinforced concrete structures named as CALRECOD for their abbreviation \textit{Computer Aided Learning of Reinforced Concrete Design}. Such development emerges as the result of a series of research works in the Autonomous University of Queretaro with the main purpose of improving the way in which the design of reinforced concrete structures is taught in high education institutions. CALRECOD uses optimization methods and algorithms to aid students in their design interaction learning so that they are able to compare their own designs and what commercial software delivers with optimal ones given certain load conditions on the elements or structures. The software consists almost entirely of MatLab functions (.m files) and the ACI 318-19 code is taken as their main design reference to make it internationally useful, although in some cases the Mexican code NTC-17 specifications are used. Besides MatLab functions, the software consists as well of ANSYS SpaceClaim script functions (.scscript files) as an additional tool for the aid in the visualization of design results in a 3D space in the software ANSYS SpaceClaim. CALRECOD has proven to be versatile, flexible and of easy use with a huge potential to increase learning outcomes for students in high education programs related with the design of reinforced concrete structures as well as to enhance the creation of efficient interactive environments for researchers and academics focused on the development of new design and analysis methods for such structures. With their optimization design functions, a solid comparison platform of designs' performance could be laid out and with its extended function design packages for structural systems, reinforced concrete design courses could be enhanced in a great deal regarding their program content's scope. The software can be found at:{https://github.com/calrecod/CALRECOD