---

A hybrid meta-heuristic optimization method is introduced to efficiently find the optimal shape of concrete gravity dams including dam-water-foundation rock interaction subjected to earthquake loading. The hybrid meta-heuristic optimization method is based on a hybrid of gravitational search algorithm (GSA) and particle swarm optimization (PSO), which is called GSA-PSO. The operation of GSA-PSO includes three phases. In the first phase, a preliminary optimization is accomplished using GSA as local search. In the second phase, an optimal initial swarm is produced using the optimum result of GSA. Finally, PSO is employed to find the optimum design using the optimal initial swarm. In order to reduce the computational cost of dam analysis subject to earthquake loading, weighted least squares support vector machine (WLS-SVM) is employed to accurately predict dynamic responses of gravity dams. Numerical results demonstrate the high performance of the hybrid meta-heuristic optimization for optimal shape design of concrete gravity dams. The solutions obtained by GSA-PSO are compared with those of GSA and PSO. It is revealed that GSA-PSO converges to a superior solution compared to GSA and PSO, and has a lower computation cost.

---

An excavation damage zone (EDZ) can be defined as a rock zone where the rock properties and conditions have been changed due to the processes related to an excavation. This zone affects the behavior of rock mass surrounding the construction that reduces the stability and safety factor and increase probability of failure of the structure. This paper presents an approach to build a model for the identification and classification of the EDZ. The Support vector machine (SVM) is a new machine learning method based on statistical learning theory, which can solve the classification problem with small sampling, non-linearity and high dimension. However, the practicability of the SVM is influenced by the difficulty of selecting appropriate SVM parameters. In this study, the proposed hybrid Harmony search (HS) with the SVM was applied for identification and classification of damaged zone, in which HS was used to determine the optimized free parameters of the SVM. For identification and classification of the EDZ, based upon the modulus of the deformation modulus and using the hybrid of HS with the SVM a model for the identification and classification of the EDZ was built. To illustrate the capability of the HS-SVM model defined, field data from a test gallery of the Gotvand dam, Iran were used. The results obtained indicate that the HS-SVM model can be used successfully for identification and classification of damaged zone around underground spaces.

---

The slope stability analysis is routinely performed by engineers to estimate the stability of river training works, road embankments, embankment dams, excavations and retaining walls. This paper presents a new approach to build a model for the prediction of slope stability state. The support vector machine (SVM) is a new machine learning method based on statistical learning theory, which can solve the classification problem with small sampling, non-linearity and high dimension. However, the practicability of the SVM is influenced by the difficulty of selecting appropriate SVM parameters. In this study, the proposed hybrid harmony search (HS) with SVM was applied for the prediction of slope stability state, in which HS was used to determine the optimized free parameters of the SVM. A dataset that includes 55 data points was applied in current study, while 45 data points (80%) were used for constructing the model and the remainder data points (10 data points) were used for assessment of degree of accuracy and robustness. The results obtained indicate that the SVM-HS model can be used successfully for the prediction of slope stability state for circular failure.

---

Concrete is the second most consumed material after water and the most widely used construction material in the world. The compressive strength of concrete is one of its most important mechanical properties, which highly depends on its mix design. The present study uses the intelligent methods with instance-based learning ability to predict the compressive strength of concrete. To achieve this objective, first, a set of data pertaining to concrete mix designs containing fly ash was collected. Then, mix design parameters were used as the inputs of the artificial neural network (ANN), support vector machine (SVM), and adaptive neuro-fuzzy inference system (ANFIS) developed for predicting the compressive strength. In all these models, prediction accuracy largely depends on the parameters of the learning model. Hence, the particle swarm optimization (PSO) algorithm, as a powerful population-based algorithm for solving continuous and discrete optimization problems, was used to determine the optimal values of algorithm parameters. The hybrid models were trained and tested with 426 experimental data and their results were compared by statistical criteria. Comparing the results of the developed models with the real values showed that the ANFIS-PSO hybrid model has the best performance and accuracy among the assessed methods.

---

Flow number of asphalt–aggregate mixtures as an explanatory factor has been proposed in order to assess the rutting potential of asphalt mixtures. This study proposes a multiple–kernel based support vector machine (MK–SVM) approach for modeling of flow number of asphalt mixtures. The MK–SVM approach consists of weighted least squares–support vector machine (WLS–SVM) integrating two kernel functions in order to improve the learning and generalization ability of WLS–SVM. In the proposed method, a linear convex combination of the radial basis function (RBF) and Morlet wavelet kernel functions is adopted, which are considered as the most popular kernel functions. To validate the efficiency of the proposed method, experiments are conducted on a database including 118 uniaxial dynamic creep test results. The results of the statistical criteria show a good agreement between the predicted and measured flow number values. Further, the simulation results demonstrate that the proposed MK–SVM approach has more superior performance than the single kernel based WLS–SVM and other methods found in the literature.

---

In this study, the support vector machine and Monte Carlo simulation are applied to predict natural frequencies of truss structures with uncertainties. Material and geometrical properties (e.g., elasticity modulus and cross-section area) of the structure are assumed to be random variables. Thus, the effects of multiple random variables on natural frequencies are investigated. Monte Carlo simulation is used for probabilistic eigenvalue analysis of the structure. In order to reduce the computational cost of Monte Carlo simulation, a support vector machine model is trained to predict the required natural frequencies of the structure computed in the simulations. The provided examples demonstrate the computational efficiency and accuracy of the proposed method compared to the direct Monte Carlo simulation in the computation of the natural frequencies for trusses with random parameters.