Journal of Soft Computing in Civil EngineeringJournal of Soft Computing in Civil Engineering
http://www.jsoftcivil.com/
Tue, 19 Nov 2019 21:50:30 +0100FeedCreatorJournal of Soft Computing in Civil Engineering
http://www.jsoftcivil.com/
Feed provided by Journal of Soft Computing in Civil Engineering. Click to visit.Neural Network based model to Estimate Dynamic Modulus E* for mixtures in Costa Rica
http://www.jsoftcivil.com/article_89875_9647.html
Various dynamic modulus (E*) predictive models have been developed to estimate E* as an alternative to laboratory testing. The most widely used model is the 1999 I-37A Witczak predictive equation based on North American mixtures laboratory results. The differences in material properties, traffic information, and environmental conditions for Latin American countries make it necessary to calibrate these models using local conditions. Consequently, the National Laboratory of Materials and Structural Models at the University of Costa Rica (in Spanish, LanammeUCR) has previously performed a local calibration of this model based on E* values for different types of Costa Rican mixtures. However, further research has shown that there is still room for improvement in the accuracy of the calibrated model (Witczak-Lanamme model) based on advanced regression techniques such as neural networks (NN). The objective of this study was to develop an improved and more effective dynamic modulus E* predictive regression model for mixtures in Costa Rica by means of NN based models. A comparison of the predicted E* values among the Witczak model, Witczak-Lanamme model and the new and improved model based on artificial neural networks (E* NN- model) indicated that the former not only met the model adequacy checking criteria but also exhibited the best goodness of fit parameters and the lowest overall bias. The findings of this study also supported the use of more advanced regression techniques that can become a more attractive alternative to local calibration of the Witczak I-37A equation.Sun, 31 Mar 2019 19:30:00 +0100Adaptive Neuro-Fuzzy and Simple Adaptive Control Methods for Alleviating the Seismic Responses ...
http://www.jsoftcivil.com/article_95796_0.html
This paper describes two adaptive control methods for mitigating the seismic responses of two connected buildings with MR dampers at different levels. First method developed in this study is the adaptive neuro-fuzzy controller which consists of a fuzzy logic controller provided with learning algorithm based on adaptive neural networks. The learning algorithm is implemented to modify the parameters of the fuzzy logic controller such that its outputs track the behavior of predetermined training data. Second method is the simple adaptive controller which falls into the category of model-following adaptive strategies. In this method, a plant is commanded to follow a well-designed reference-model with desirable trajectories through a closed loop action. The coupled system consists of two adjacent buildings having different heights in order to separate the model shapes of the individual buildings. Different types of feedbacks such as displacement, velocity, and acceleration are employed to identify their impacts on the performance of the developed adaptive controllers. Numerical analyses are carried out for the complex system assuming no change in the nominal design parameters and then for the system where a change in these parameters is introduced. The results reveal that using the adaptive controllers developed in this study to regulate the MR dampers connecting the two adjacent buildings can successfully alleviate the seismic responses under various types and intensities of earthquakes.Fri, 25 Oct 2019 20:30:00 +0100Evolutionary algorithm performance evaluation in project time-cost optimization
http://www.jsoftcivil.com/article_89544_9647.html
The time-cost trade-off problem pertains to the assessment of the best method of activity construction so that a project is completed within a given deadline and at least cost. Although several evolutionary-type of algorithms have been reported over the last two decades to solve this NP-hard combinatorial problem, there are not many comparative studies independently evaluating several methods. Such studies can provide support to project managers regarding the selection of the appropriate method. The objective of this work is to comparatively evaluate the performance potential of a number of evolutionary algorithms, each one with its own variations, for the time-cost trade-off problem. The evaluation is based on two measures of effectiveness, the solution quality (accuracy) and the processing time to obtain the solution. The solution is sought via a general purpose commercial optimization software without much interference in algorithm parameter setting and fine-tuning in an attempt to follow the anticipated project manager approach. The investigation has been based on case studies from the literature with varying project size and characteristics. Results indicate that certain structures of genetic algorithms, particle swarm optimization, and differential evolution method present the best performance.Sun, 31 Mar 2019 19:30:00 +0100Application of random forest regression in the Prediction of ultimate bearing capacity of strip ...
http://www.jsoftcivil.com/article_89975_0.html
The paper presents the prediction of the ultimate bearing capacity of the strip footing resting on dense sand overlying loose sand deposit using random forest regression. In this study, 181 data collected from literature were used. 71 % of the total data was randomly selected for training the model and the rest of the data were utilized for the testing purpose. The various input parameters were friction angle of the dense sand layer (1), friction angle of the loose sand layer (2), unit weight of the dense sand layer (1), unit weight of the loose sand layer (), ratio of the thickness of the dense sand layer below base of the footing to the width of footing (H/B), ratio of the depth of the footing to the width of the footing (D/B) and (H+D)/B. Ultimate bearing capacity was the output in this study. Performance measures were used in order to make the comparison with the artificial neural network and M5P model tree. The result of this study reveals that the performance of the random forest regression was superior to the other soft computing techniques. The results of the sensitivity analysis reveals that the unit weight and the friction angle of the loose sand layer were the most important parameters affecting the output ultimate bearing capacity of the strip footing resting on the layered sand deposit.Thu, 04 Jul 2019 19:30:00 +0100Application of Soft Computing Techniques in Predicting the Ultimate Bearing Capacity of Strip ...
http://www.jsoftcivil.com/article_89749_9647.html
The present study attempts to predict the ultimate bearing capacity (UBC) of the strip footing resting on sand and subjected to inclined load having eccentricity with respect to the vertical using three different soft computing techniques such as support vector mechanism with radial basis function (SVM RBF kernel), M5P model tree (M5P) and random forest regression (RFR). The UBC was computed in the form of reduction factor and this reduction factor was assumed to be dependent on the ultimate bearing capacity (qu) of the strip footing subjected to vertical load, eccentricity ratio (e/B), inclination ratio (α/ϕ) and the embedment ratio (Df/B). The performance of each model was analyzed by comparing the statistical performance measure parameters. The outcome of present study suggests that SVM RBF kernel predicts the reduction factor with least error followed by M5P and RFR. All the model predictions further outperformed those based on semi-empirical approach available in literature. Finally, sensitivity analysis performed for the SVM RBF kernel model which suggests that the inclination ratio (α/ϕ) and eccentricity ratio (e/B) was an important parameter, in comparison to other parameters, considered for predicting the reduction factor.Sun, 31 Mar 2019 19:30:00 +0100Forecasting of Wind- Wave Height by using Adaptive Neuro-Fuzzy Inference System and Decision Tree
http://www.jsoftcivil.com/article_95949_0.html
Wind- induced waves are considered to be the most important waves in the sea due to their high energy and frequency. Among the characteristics of the waves, height is one of the most important parameters that are used in most equations related to marine engineering designs. Since the application of soft computing methods in marine engineering has been developed in recent years, in this study, an adaptive neuro-fuzzy inference system and a decision tree have been used to predict the wind-induced wave height in Bushehr port. In order to identify the effective parameters, implementing different models from different inputs. By considering the accuracy of the models, the effective parameters in wave height were identified using statistical measures correlation coefficient (r), Mean Square Error (mse). The results of this study indicated that in the prediction of wind-induced wave height, compared to the decision tree, the accuracy of the model of the neural-fuzzy system for 3, 6 and 9 hours was higher. Also, the results showed that the use of wind shear velocity instead of wind speed at 10 meters above the water level had a higher accuracy in forecasting of the significant wave height. The results also indicated that among the presented models, the combined model of the significant wave height, shear velocity, and the difference between the direction and wind speed as well as the length of the fetch has the highest accuracy.Sat, 02 Nov 2019 20:30:00 +0100MODELLING OF CONCRETE COMPRESSIVE STRENGTH ADMIXED WITH GGBFS USING GENE EXPRESSION PROGRAMMING
http://www.jsoftcivil.com/article_91665_9647.html
Several studies have established that strength development in concrete is not only determined by the water/binder ratio, but it is also affected by the presence of other ingredients. With the increase in the number of concrete ingredients from the conventional four materials by addition of various types of admixtures (agricultural wastes, chemical, mineral and biological) to achieve a desired property, modelling its behavior has become more complex and challenging. Presented in this work is the possibility of adopting the Gene Expression Programming (GEP) algorithm to predict the compressive strength of concrete admixed with Ground Granulated Blast Furnace Slag (GGBFS) as Supplementary Cementitious Materials (SCMs). A set of data with satisfactory experimental results were obtained from literatures for the study. Result from the GEP algorithm was compared with that from stepwise regression analysis in order to appreciate the accuracy of GEP algorithm as compared to other data analysis program. With R-Square value and MSE of -0.94 and 5.15 respectively, The GEP algorithm proves to be more accurate in the modelling of concrete compressive strength.Sun, 31 Mar 2019 19:30:00 +0100Machine Learning Method for predicting the depth of shallow lakes Using Multi-Band Remote ...
http://www.jsoftcivil.com/article_95794_9647.html
Knowing the lake’s characteristics information such as depth is an essential requirement for the water managers; however, conducting a comprehensive bathymetric survey is considered as a difficult task. After the advent of remote sensing, and satellite imagery, it has been recognized that water depth can be estimated in some way over shallow water. There are many models that can evaluate relationships between multi-band images, and depth measurements; however, artificial computation methods can be used as an approximation tool for this issue. They are also considered as fairly simple and practical models to estimate depth in shallow waters. In this paper, different methods of artificial computation are used to calculate the depth of shallow lake, then these methods are compared. The results show that Artificial Neural Network (ANN), Adaptive Neuro Fuzzy Inference System (ANFIS), and regression learner are best methods for this issue with RMSE 0.8 , 1.47, and 0.96 respectively.Sun, 31 Mar 2019 19:30:00 +0100Hydrologic Input-output model of Mt. Isarog watershed: Estimating the channel cross section ...
http://www.jsoftcivil.com/article_95960_0.html
This study estimated the daily extreme runoff water on the watershed based from the gathered twenty four hour peak precipitation per month of year 2018 using modified soil conservation system (SCS-CN) method. This established fuzzy rule based system which is used to estimate the runoff that could pass the channel without overflow, and this described the event of overflow in the river channel. Rain gauge was used to collect the daily rainfall data. Pattern recognition method was used to compute watershed area through satellite images and in confirming areas with evidences of runoff overflow. The process was centered on the size of the cross sectional area of the river and the amount of river water discharge. The highest precipitation event that happened on the month of December has found the river channel cross section to be insufficient to transport the extreme daily watershed runoff. Traces of runoff overflow are visible on satellite images.Mon, 04 Nov 2019 20:30:00 +0100Prediction of Flexural strength of Concrete Produced by Using Pozzolanic Materials and Partly ...
http://www.jsoftcivil.com/article_95795_9647.html
The use of huge quantity of natural fine aggregate (NFA) and cement in civil construction work which have given rise to various ecological problems. The industrial waste like Blast furnace slag (GGBFS), fly ash, metakaolin, silica fume can be used as partly replacement for cement and manufactured sand obtained from crusher, was partly used as fine aggregate .In this work, MATLAB software model is developed using neural network toolbox to predict the flexural strength of concrete made by using pozzolanic materials and partly replacing natural fine aggregate (NFA) by Manufactured sand(MS). Flexural strength was experimentally calculated by casting beams specimens and results obtained from experiment were used to develop the artificial neural network (ANN) model. Total 131 results values were used to modeling formation and from that 30% data record was used for testing purpose and 70% data record was used for training purpose. 25 input materials properties were used to find the 28 days flexural strength of concrete obtained from partly replacing cement with pozzolans and partly replacing natural fine aggregate (NFA) by manufactured sand (MS). The results obtained from ANN model provides very strong accuracy to predict flexural strength of concrete obtained from partly replacing cement with pozzolans and natural fine aggregate (NFA) by manufactured sand.Sun, 31 Mar 2019 19:30:00 +0100Comparison of Genetic Algorithm(GA) and Particle Swarm Optimization Algorithm(PSO) for Discrete ...
http://www.jsoftcivil.com/article_95950_9647.html
Optimization of truss structures including topology, shape and size optimization were investigated by different researchers in the previous years. The aim of this study is discrete and continuous size optimization of two-dimensional truss structures with the fixed topology and the shape. For this purpose, the section area of the members are considered as the decision variables and the weight minimization as the objective function. The constraints are the member stresses and the node displacements which should be limited at the allowable ranges for each case.In this study, Genetic Algorithm and Particle Swarm Optimization algorithm are used for truss optimization. To analyse and determine the stresses and displacements, OpenSees software is used and linked with the codes of Genetic Algorithm and Particle Swarm Optimization algorithm provided in the MATLAB software environment.In this study, the optimization of four two-dimensional trusses including the Six-node, 10-member truss, the Eight-node, 15-member truss, the Nine-node, 17-member truss and the Twenty-node, 45-member truss under different loadings derived from the literature are done by the Genetic Algorithm and Particle Swarm Optimization algorithm and the results are compared with those of the other researchers. The comparisons show the outputs of the Genetic Algorithm are the most generally economical among the different studies for the discrete size cases while for the continuous size cases, the outputs of the Particle Swarm Optimization algorithm are the most economical.Sun, 31 Mar 2019 19:30:00 +0100