International Journal of Optimization in Civil Engineering
Volume:12 Issue: 3, Summer 2022

Presently, the introduction of intelligent models to optimize structural problems has become an important issue in civil engineering and almost all other fields of engineering. Optimization models in artificial intelligence have enabled us to provide powerful and practical solutions to structural optimization problems. In this study, a novel method for optimizing structures as well as solving structure-related problems is presented. The main purpose of this paper is to present an algorithm that addresses the major drawbacks of commonly-used algorithms including the Grey Wolf Optimization Algorithm (GWO), the Gravitational Search Algorithm (GSA), and the Particle Swarm Optimization Algorithm (PSO), and at the same time benefits from a high convergence rate. Also, another advantage of the proposed CGPGC algorithm is its considerable flexibility to solve a variety of optimization problems. To this end, we were inspired by the GSA law of gravity, the GWO's top three search factors, the PSO algorithm in calculating speed, and the cellular machine theory in the realm of population segmentation. The use of cellular neighborhood reduces the likelihood of getting caught in the local optimal trap and increases the rate of convergence to the global optimal point. Achieving reasonable results in mathematical functions (CEC 2005) and spatial structures (with a large number of variables) in comparison with those from GWO, GSA, PSO, and some other common heuristic algorithms shows an enhancement in the performance of the introduced method compared to the other ones.

The wind loads considerably influence lightweight spatial structures. An example of spatial structures is scallop domes that contain various configurations and forms and the wind impact on a scallop dome is more complex due to its additional curvature. In our work, the wind pressure coefficient (Cp ) on the scallop dome surface is studied numerically and experimentally. Firstly, the programming language Formian-K is used for generating the scallop dome configuration. Then, the scallop dome scale model is designed using a CAD/CAM system, and it is constructed in fiberglass. Afterward, the wind tunnel of the atmospheric boundary layer is presented, and the scale model is applied for performing the tests so that the Cp  is obtained. The scallop dome scale model was taken into account in numerical investigation. For simulation of the turbulent flow, Large Eddy Simulation (LES), Reynolds Stress Turbulence Model (RSM), the k-ε RNG, and k-omega Shear Stress Transport (k-ω SST) approaches were used. Lastly, we compared the wind pressure coefficients obtained by Computational Fluid Dynamics (CFD) with the results of the experimental investigation. As indicated by the results, the LES method, particularly, RSM model, can be applied because of lower computational costs for the analysis of other scallop dome configurations for obtaining Cp .

Design optimization of structures with discrete and continuous search spaces is a complex optimization problem with lots of local optima. Metaheuristic optimization algorithms, due to not requiring gradient information of the objective function, are efficient tools for solving these problems at a reasonable computational time. In this paper, the Doppler Effect-Mean Euclidian Distance Threshold (DE-MEDT) metaheuristic algorithm is applied to solve the discrete and continuous optimization problems of the truss structures subject to multiple loading conditions and design constraints. DE-MEDT algorithm is a recently proposed metaheuristic developed based on a physical phenomenon called Doppler Effect (DE) with some idealized rules and a mechanism called Mean Euclidian Distance Threshold (MEDT). The efficiency of the DE-MEDT algorithm is evaluated by optimizing five large-scale truss structures with continuous and discrete variables. Comparing the results found by the DE-MEDT algorithm with those of other existing metaheuristics reveals that the DE-MEDT optimizer is a suitable optimization technique for discrete and continuous design optimization of large-scale truss structures.

The main purpose of this paper was to use a combination of Energy-based design method and whale algorithm (WOA), hereinafter referred to as E-WOA, to optimize steel moment frames and improve the seismic performance. In E-WOA, by properly estimating the seismic input energy and determining the optimal mechanism for the structure, steel frames are designed based on the energy balance method; according to the results, in a suitable search space, optimization is performed using the WOA algorithm. The objective function of the WOA algorithm, in addition to the frame weight, is meant to improve the behavior of the structure based on the performance level criteria of the ASCE41-17 standard and the uniformity of the drift distribution at the frame height. The results show that the initial design of the Energy method reduces the computational volume of the WOA algorithm to achieve the optimal solution and the plastic hinge pattern in frame is more favorable in the E-WOA method than in the design done by the Energy method.

Today, due to the complexity of engineering problems and at the same time the advancement of computer science, the use of machine learning (ML) methods and soft computing methods in solving engineering problems has been considered by many researchers. These methods can be used to find accurate estimates for problems in various scientific fields. This paper investigates the effectiveness of the Adaptive Network-Based Fuzzy Inference System (ANFIS) hybridized with Teaching Learning Based Optimization Algorithm (TLBO), to predict the ultimate strength of columns with square and rectangular cross-sections, confide with various fiber-reinforced polymer (FRP) sheets. In previous studies by many researchers, several experiments have been conducted on concrete columns confined by FRP sheets. The results indicate that FRP sheets effectively increase the compressive strength of concrete columns. Comparing the results of ANFIS-TLBO with the experimental findings, which were agreeably consistent, demonstrated the ability of ANFIS-TLBO to estimate the compressive strength of concrete confined by FRP. Also, the comparison of RMSE, SD, and R2 for ANFIS-TLBO and the studies of different researchers show that the ANFIS-TLBO approach has a good performance in estimating compressive strength. For example, the value of R2 in the proposed method was 0.92, while this parameter was 0.87 at best among the previous studies. Also, the obtained error in the prediction of the proposed model is much lower than the obtained error in the previous studies. Hence, the proposed model is more efficient and works better than other techniques.

In present study, the effects of optimization on seismic energy spectra including input energy, damping energy and yielding hysteretic energy are parametrically discussed. To this end, 12 generic steel moment-resisting frames having fundamental periods ranging from 0.3 to 3s are optimized by using uniform damage and deformation approaches subjected to a series of 40 non-pule strong ground motions. In order to obtain the optimum distribution of structural properties, an iterative optimization procedure has been adopted. In this approach, the structural properties are modified so that inefficient material is gradually shifted from strong to weak areas of a structure. This process is continued until a state of uniform damage is achieved. Then, the maximum energy demand parameters are computed for different structures designed by optimum load pattern as well as code-based pattern, and the mean energy spectra, energy-based reduction factor and the dispersion of the results are compared and discussed. Results indicate that optimum seismic load pattern can significantly affect the energy demands spectra especially in inelastic range of response. In addition, using energy-based reduction factors of optimum structures in short-period and long-period regions will result in respectively overestimation and underestimation of the required input energy demands for code-based structures, reflecting the difference dose exists in reality between the conventional forced-based methodology and energy-based seismic design approach that can more realistically incorporate the frequency content and duration of earthquake ground motions.

The lateral-torsional buckling (LTB) strength of cellular steel girders that were subjected to web distortion was rarely examined. Since no formulation has been presented for predicting the capacity of such beams, in the current paper an extensive numerical investigation containing 660 specimens was modeled using finite element analysis (FEA) to consider the ultimate lateral-distortional buckling (LDB) strength of such members. Then, a reliable algorithm based on the artificial neural networks (ANNs) was developed and the most accurate model was chosen to derive an efficient formula to evaluate the LDB capacity of steel cellular beams. The input and target data required in the ANN models were provided using the ANN analyzes. An attempt was made to include the proposed formula in all the variables affecting the LDB of cellular steel beams. In the next step, the validity of the proposed formula was proved by several statistical criteria, and also the most influential input variable was discussed. eventually, a comparison study was executed between the results provided by the ANN-based equation and the AS4100, EC3, and AISC codes. It was revealed that the presented equation is accurate enough and can be used by practical engineers.

Metaheuristic algorithms are preferred by the many researchers to reach the reliability based design optimization (RBDO) of truss structures. The cross-sectional area of the elements of a truss is considered as design variables for the size optimization under frequency constraints. The design of dome truss structures are optimized based on reliability by a popular metaheuristic optimization technique named Enhanced Vibrating Particle System (EVPS). Finite element analyses of structures and optimization process are coded in MATLAB. Large-scale dome truss of 600-bar, 1180-bar and 1410-bar are investigated in this paper and are compared with the previous studies. Also, a comparison is made between the reliability indexes of Deterministic Design Optimization (DDO) for large dome trusses and Reliability-Based Design Optimization (RBDO).