Search

Summary The paper presents artificial neural networks (ANNs) based methodology for ascertaining the structural parameters of water harvesting structures (WHS) at the conceptual stage of design. The ANN is trained using exemplar patterns generated using an in-house MSExcel based design program, to draw a functional relationship between the five inputs design parameters namely, peak flood discharge, safe bearing capacity of strata, length of structure, height of structure and silt factor and four outputs namely, top width, bottom width, foundation depth and flood lift representing the structural parameters of WHS. The results of the study show that, the structural parameters of the WHS predicted using ANN model are in close agreement with the actual field parameters. The versatility of ANN to map complex or complex unknown relationships has been proven in the study. A parametric sensitivity study is also performed to assess the most significant design parameter. The study holistically presents a neural network based decision support tool that can be used to accurately estimate the major design parameters of the WHS at the conceptual stage of design in quick time, aiding the engineer-in-charge to conveniently forecast the budget requirements and minimize the labor involved during the subsequent phases of analysis and design.

Summary The increasing population and large scale growth with the development of modern science and technology has indicated very high stress on water sector in Rajasthan in India. Availability of water and uniformity of rainfall distribution is changing day by day due to shifting of monsoon in Rajasthan. The spatial and temporal variations in the rainfall in different river basins in Rajasthan are drastic due to which flood situation arises in the tributaries of Chambal river basin every year. Simultaneously on the other hand drought situation arises in the other different places in the most of the river basins which are also under dark zone. In totality natural water resources are also limited in Rajasthan so there is an urgent need for its proper planning, development and management of the available water duly taking care of environmental, ecological, socio-economic aspects and their inter-relationships. Annual surplus water of about 1437 MCM in the river Chambal is going waste and ultimately reaches to sea after creating flood situations in various places in India including Rajasthan, while on the other hand 1077 MCM water is a requirement in the four other basins in Rajasthan i.e. Banas, Banganga, Gambhir and Parbati at 75% dependability. Interlinking and water transfer from Chambal to these four river basins is the prime solution for which 372 km link channel including 9 km tunnel of design capacity of 300 cumec with 64 m lift is required.

With rapid growth in the construction industry, Ready Mix Concrete (RMC) is playing a key role in offering high-quality customized concrete to contractors and builders. The workability of concrete involves early stage operations of concrete: placing, compaction, and finishing. Since RMC is manufactured at a plant and transported to the construction site, loss of workability is of prime concern due to the considerable time interval between mixing and placing of concrete. Workability of concrete is measured using a slump test to evaluate the life of the concrete during its transportation phase and the uniformity of the concrete from batch to batch. The proportions of cement, fly ash, coarse aggregates, fine aggregates, water, and admixtures in the concrete govern its workability or slump value. In this study, an Artificial Neural Networks (ANNs) learning from past examples gathered from a RMC plant were used to model the functional relationship between the input parameters and the slump value of concrete. The ANN model provided promising results compared to first-order and second-order regression techniques for modeling the unknown and complex relationships exhibited by the design mix proportions and the slump of concrete. The neural network synaptic weights that control the learning mechanism of ANN were further used to compute the percentage of relative importance of each constituent of RMC on the slump value, providing insight into the composite nature of concrete. The technique presented in the study will enable technical staff to quickly estimate the slump of RMC based on its design mix constituents without having to perform multiple design mix trials in order to achieve a customized slump value.

The paper explores the usefulness of hybridizing two distinct nature inspired computational intelligence techniques viz., Artificial Neural Networks (ANN) and Genetic Algorithms (GA) for modeling slump of Ready Mix Concrete (RMC) based on its design mix constituents viz., cement, fly ash, sand, coarse aggregates, admixture and water-binder ratio. The methodology utilizes the universal function approximation ability of ANN for imbibing the subtle relationships between the input and output variables and the stochastic search ability of GA for evolving the initial optimal weights and biases of the ANN to minimize the probability of neural network getting trapped at local minima and slowly converging to global optimum. The performance of hybrid model (ANN-GA) was compared with commonly used back-propagation neural network (BPNN) using six different statistical parameters. The study showed that by hybridizing ANN with GA, the convergence speed of ANN and its accuracy of prediction can be improved. The trained hybrid model can be used for predicting slump of concrete for a given concrete design mix in quick time without performing multiple trials with different design mix proportions.

Summary The study investigates the use of crushed rock sand as viable alternative to Natural River sand that is being conventionally used as fine aggregate in cement concrete. Various mix designs were developed for different grades of concrete based on IS, ACI and British codes using Natural River sand and crushed rock sand. In each case, the cube compressive strength test, and beam flexure tests were conducted. The results of the study show that, the strength properties of concrete using crushed rock sand are nearly similar to the conventional concrete. The study has shown that crushed stone sand can be used as economic and readily available alternative to river sand and can therefore help to arrest the detrimental effects on the environment caused due to excessive mining of river sand.

Enormous cost and manpower utilization encountered in constructing a water resource project demands a great deal of attention in devising precise Rainfall-Runoff models for its successful performance. These models are dependent on the physiographic, climatic and biotic characteristics of the basin. These factors sometimes induce either a linear, non-linear or highly complex behaviour among the rainfall and runoff parameters. The unstructured nature of Rainfall-Runoff relations has diverted the attention of researchers towards Soft Computing tools which, harnesses reasoning, intuition, consciousness and wisdom possessed by human beings. Soft Computing being a multi-disciplinary field uses a variety of statistical, probabilistic and optimization tools which complement each other to produce its three main branches viz., Neural Networks, Genetic Algorithms and Fuzzy Logic. These techniques, whether complementing each other or working on their own, are able to model complex or unknown relationships which are either nonlinear or noisy. The review paper presents an introduction to these techniques and discusses their applications in modelling Rainfall-Runoff relations which to some extent have replaced time consuming conventional mathematical techniques with wiser and time saving computing tools.

field of engineering is a creative one. The problems encountered in this field are generally unstructured and imprecise influenced by intuitions and past experiences of a designer. The conventional methods of computing relying on analytical or empirical relations become time consuming and labor intensive when posed with real life problems. To study, model and analyze such problems, approximate computer based Soft Computing techniques inspired by the reasoning, intuition, consciousness and wisdom possessed by a human beings are employed. In contrast to conventional computing techniques which rely on exact solutions, soft computing aims at exploiting given tolerance of imprecision, the trivial and uncertain nature of the problem to yield an approximate solution to a problem in quick time. Soft Computing being a multi-disciplinary field uses a variety of statistical, probabilistic and optimization tools which complement each other to produce its three main branches viz., Neural Networks, Genetic Algorithms and Fuzzy Logic. The review paper presents the applications of two major Soft Computing techniques viz., Artificial Neural Networks and Genetic Algorithms in the field of Civil Engineering, which to some extent has replaced the time consuming conventional techniques of computing with intelligent and time saving computing tools.

Transformation of occurred rainfall into runoff generated within a catchment is a complex natural phenomenon that passes through various inter-related processes and influenced by many topographic, geographic, geologic and sociologic factors. To develop a model that can reliably imbibe the complex Rainfall-Runoff interaction, two different approaches namely, conventional regression and Artificial Neural Networks (ANN) have been attempted for Rajsamand Catchment located in the Banas River Basin of Rajasthan. The monthly weighted rainfall of the catchment and inflow received during the monsoon period in the Rajsamand Dam for last twenty years were used for the study. The weighted rainfall and corresponding runoff for the months July, August, September and the monsoon period were considered as inputs and outputs respectively for the model. The performance compared using three different statistical metrics reveal that, the Rainfall-Runoff Model based on ANN approach offer better prediction accuracy than the conventional regression model.

Artificial Neural Networks (ANN) trained using backpropagation (BP) algorithm are commonly used for modeling material behavior associated with non-linear, complex or unknown interactions among the material constituents. Despite multidisciplinary applications of back-propagation neural networks (BPNN), the BP algorithm possesses the inherent drawback of getting trapped in local minima and slowly converging to a global optimum. The paper present a hybrid artificial neural networks and genetic algorithm approach for modeling slump of ready mix concrete based on its design mix constituents. Genetic algorithms (GA) global search is employed for evolving the initial weights and biases for training of neural networks, which are further fine tuned using the BP algorithm. The study showed that, hybrid ANN-GA model provided consistent predictions in comparison to commonly used BPNN model. In comparison to BPNN model, the hybrid ANNGA model was able to reach the desired performance goal quickly. Apart from the modeling slump of ready mix concrete, the synaptic weights of neural networks were harnessed for analyzing the relative importance of concrete design mix constituents on the slump value. The sand and water constituents of the concrete design mix were found to exhibit maximum importance on the concrete slump value., {"references":["P.K. Mehta and P.J.M. Monteiro, Concrete: Structure, Properties and\nMaterials. 3rd ed. New York: McGraw Hill, 2006.","Z. Li, Advanced Concrete Technology. 1st ed. New Jersey: John Wiley &\nSons, Inc, 2011.","W.P.S. Dias and S.P. Pooliyadda, \"Neural Networks for predicting\nproperties of concrete with admixtures,\" Construction and Building\nMaterials, vol. 15, no. 7, pp. 371-379, 2001.","I.-C. Yeh, \"Exploring concrete slump model using artificial neural\nnetworks,\" Journal of Computing in Civil Engineering, vol. 20, no. 3,\npp. 217-221, 2006.","A. Oztas, M. Pala, E. Ozbay, E. Kanca, N. Caglar and M.A. Bhatti,\n\"Predicting the compressive strength and slump of high strength\nconcrete using neural network,\" Construction and Building Materials,\nvol. 20, no. 9, pp. 769-775, 2006.","I.-C.Yeh, \"Modeling slump flow of concrete using second-order\nregressions and artificial neural networks,\" Cement and Concrete\nComposites, vol. 29, pp. 474-480, 2007.","A. Jain, S.K. Jha and S. Misra, \"Modeling and analysis of concrete\nslump using artificial neural networks,\" Journal of Materials in Civil\nEngineering, vol. 20, no. 9, pp. 628-633, 2008.","M. Saridemir, \"Prediction of compressive strength of concretes\ncontaining metakaolin and silica fumes by artificial neural networks,\"\nAdvances in Engineering Software, vol. 40, no. 5, pp. 350-355, 2009.","S.J. Kwon and H.W. Song, \"Analysis of carbonation behaviour in\nconcrete using neural network algorithm and carbonation modeling,\"\nCement and Concrete Research, vol. 40, no. 1, pp. 119-127, 2010.\n[10] R. Siddique, P. Aggarwal and Y. Aggarwal, \"Prediction of compressive\nstrength of self compacting concrete containing bottom ash using\nartificial neural networks,\" Advances in Engineering Software, vol. 42,\nno. 10, pp. 780-786, 2011.\n[11] M.I. Khan, \"Predicting properties of high performance concrete\ncontaining composite cementitious materials using Artificial Neural\nNetworks,\" Automation in Construction, vol.22, pp. 516-524, 2012.\n[12] O.A. Hodhod and H.I. Ahmed, \"Developing an artificial neural network\nmodel to evaluate chloride diffusivity in high performance concrete,\"\nHBRC Journal, vol.9, no. 1, pp. 15-21, 2013.\n[13] A.M. Diab, H.E. Elyamany, A.E.M.A. Elmoaty and A.H. Shalan,\n\"Prediction of concrete compressive strength due to long term sulphate\nattack using neural networks,\" Alexandria Engineering Journal, vol.53,\npp. 627-642, 2014.\n[14] C.L. Su, S.M. Yang and W.L. Huang, \"A two stage algorithm\nintegrating genetic algorithms and modified Newton method for neural\nnetwork training in engineering systems,\" Expert Systems with\nApplications, vol. 38, no. 10, pp. 12189-12194, 2011.\n[15] A. Johari, A.A. Javadi and G. Habibagahi, \"Modelling the mechanical\nbahaviour of unsaturated soils using a genetic algorithm based neural\nnetwork,\" Computers and Geotechnics, vol. 38, no. 1, pp. 2-13, 2011.\n[16] H. Karimi and F. Yousefi, \" Application of artificial neural networkgenetic\nalgorithm (ANN-GA) to correlation of density in nanofluids,\"\nFluid Phase Equilibria, vol. 336, pp. 79-83, 2012.\n[17] R. Wang,C. Zhou, Z. Deng, B. Ni and Z. Zhao, \"Predicting foF2 in China\nregion using the artificial neural networks improved by the genetic\nalgorithms,\" Journal of Atmospheric and Solar-Terrestrial Physics, vol.\n92, pp. 7-17, 2013.\n[18] Y. Xue, L. Cheng, J. Mou and W. Zhao, \"A new fracture prediction\nmethod by combining genetic algorithm with neural network in lowpermeability\nreservoirs,\" Journal of Petroleum Science and Engineering,\nvol. 121, pp. 159-166, 2014.\n[19] M.M. Alshihri, A.M. Azmy and M.S. El-Bisy, \"Neural Networks for\npredicting compressive strength of structural light weight concrete,\"\nConstruction and Building Materials, vol. 23, no. 6, pp. 2214-2219,\n2009.\n[20] K. Hornik, M. Stinchcombe and H. White, \"Multilayer feed forward\nnetworks are universal approximators,\" Neural Networks, vol. 2, no. 5,\npp. 359-366, 1989.\n[21] S. Tamura and M. Tateishi, \"Capabilities of four layered feedforward\nneural network: four layer versus three,\" IEEE Trasactions on Neural\nNetworks, vol. 8, no. 2, pp. 251-255, 1997.\n[22] P.C. Pendharkar and J.A.Rodger, \"Technical efficiency based selection\nof learning cases to improve the forecasting efficiency of neural\nnetworks under monotonicity assumption,\", Decision Support Systems,\nvol. 36, no. 1, pp. 117-136, 2003.\n[23] K. Jinchuan and L. Xinzhe, \"Empirical analysis of optimal hidden layer\nneurons in neural network modeling for stock prediction,\" in\nProceedings of Pacific-Asia Workshop on Computational Intelligence\nand Industrial Applications, vol. 2, pp. 828-832, Dec. 2008.\n[24] D. Hunter, Y. Hao, M.S. Pukish, J. Kolbusz and B.M Wilamowski,\n\"Selection of proper Neural Network sizes and architectures-A\ncomparative study,\" IEEE Transaction on Industrial Informatics, vol. 8,\nno. 2, pp. 228-240, 2012.\n[25] S. Rajasekaran and G.A.V. Pai, \"Neural Networks, Fuzzy Logic and\nGenetic Algorithms: Synthesis & Applications,\" New Delhi: Prentice-\nHall of India Private Limited, 2003..\n[26] B.M. Wilamowski, Y. Chen, A. Malinowski, \"Efficient Algorithm for\nTraining Neural Networks with One Hidden Layer,\" in Proceedings of\nInternational Joint Conference on Neural Networks IEEE, pp. 1725-\n1728, 1999.\n[27] K. Wang, Computational Intelligence in Agile Manufacturing\nEngineering, in: Gunasekaran A, editor. Agile Manufacturing The 21st\nCentury Competitive Strategy, Oxford, UK: Elsevier Science Ltd, 2001,\npp. 297-315.\n[28] J.E. Nash and J.V. Sutcliffe, \"River flow forecasting through conceptual\nmodels Part I – a discussion of principles,\" Journal of Hydrology, vol.\n10, no. 3, pp. 282–290, 1970.\n[29] S. Srinivasulu and A. Jain, \"A comparative analysis of training methods\nfor artificial neural network rainfall-runoff models,\" Applied Soft\nComputing, vol. 6, pp. 295-306, 2006.\n[30] J.D. Olden and D.A. Jackson, \"Illuminating the \"Black Box\": a\nrandomization approach for understanding variable contributions in\nartificial neural networks.\" Ecological Modeling ,vol. 154, pp. 135-150,\n2002.\n[31] G. Acciani, E. Chiarantoni and G. Fornarelli, A neural network\napproach to study O3 and PM10 concentration, in: Kollias S, Staflopatis\nA, Duch W, Oja E, editors. ICANN'06 Proceedings of the 16th\ninternational conference on Artificial Neural Networks - Volume Part II,\nBerlin, Germany: Springer Verlag, 2006, pp. 913-922.\n[32] G.D. Garson, \"Interpreting neural network connection\nweights,\"Artificial Intelligence Expert , vol. 6, pp. 47-51, 1991.\n[33] M. Gevrey, I. Dimopoulos and S. Lek, \"Review and comparison of\nmethods to study the contribution of variables in artificial neural\nnetwork models,\" Ecological Modeling , vol. 160, pp. 249-264, 2003.\n[34] J.J. Montano and A. Palmer, \"Numeric sensitivity analysis applied to\nfeedforward neural networks,\" Neural Computing and Applications, vol.\n12, pp. 119-125, 2003."]}