Abstract
The aim of the present study is to propose an alternative artificial neural network model based on response surface methodology over conventional approach to estimate the unconfined compressive strength of artificially cemented bauxite residue. The artificial neural network model uses molding moisture content (w), curing time (t) and porosity/volumetric lime (η/Lv′) as input parameters and unconfined compressive strength as the output parameter. Bayesian regularization as training function with sigmoid and pure linear at hidden and output layers is used for modeling the artificial neural network. The proposed response surface methodology designed ANN model is comparable with the conventional designed ANN model and can be used effectively with significantly less number of data set. Sensitivity analysis, to make out the significant input factors based on connection-weight approach, is also discussed. Further, neural interpretation diagram is incorporated to study the effects of individual input parameters over the response. Finally, a predictive equation is presented based on response surface methodology designed artificial neural network model for the range of parameters studied.
Similar content being viewed by others
Abbreviations
- AAE:
-
Average absolute error
- ANN:
-
Artificial neural network
- BBDANN:
-
Box–Behnken designed ANN
- CCC:
-
Circumscribed central composite
- CCDANN:
-
Central composite designed ANN
- CONVDANN:
-
Conventional designed ANN
- FCCD:
-
Face-centered composite design
- FFBPANN:
-
Feed forward back propagation artificial neural network
- FIS:
-
Fuzzy interface system
- GA:
-
Genetic algorithms
- ICC:
-
Inscribed central composite
- MAE:
-
Maximum absolute error
- MAPE:
-
Mean absolute percentage error
- MSE:
-
Mean square error
- OFAT:
-
One factor at time
- RMSE:
-
Root-mean-square error
- RSM:
-
Response surface methodology
- SEM:
-
Scanning electron micrograph
- SVM:
-
Support vector machine
- trainbr:
-
Bayesian regularization training function
- UCS:
-
Unconfined compressive strength (qu)
- XRD:
-
X-ray diffraction
- b hk :
-
Bias at the kth neuron in the hidden layer
- b o :
-
Bias at the output layer
- H :
-
Number of hidden layers
- K :
-
Number of neurons
- L :
-
Lime content
- m :
-
Number of hidden neurons
- η/L v′ :
-
Porosity/volumetric lime ratio
- q u :
-
Measured unconfined compressive strength (UCS)
- q umax :
-
Predicted maximum unconfined compressive strength
- q umin :
-
Predicted minimum unconfined compressive strength
- q up :
-
Predicted unconfined compressive strength
- R 2 :
-
Coefficient of correlation (R-squared)
- q un :
-
Normalized predicted unconfined compressive strength
- t :
-
Curing time
- w :
-
Moisture content
- w ik :
-
Connection weight between ith input variable and kth neuron in hidden layer
- w k :
-
Connection weight between kth neuron in hidden layer and single output neuron
- X i :
-
Normalized input variable i
- f :
-
Activation function
- γ d :
-
Dry density of the specimen
- Z :
-
Number of input factors
- G L :
-
Specific gravity of lime
- G RM :
-
Specific gravity of bauxite residue
- γ w :
-
Density of water
References
Li J, Xu L, Sun P, Zhai P, Chen X, Zhang H et al (2017) Novel application of red mud: Facile hydrothermal-thermal conversion synthesis of hierarchical porous AlOOH and Al2O3 microspheres as adsorbents for dye removal. Chem Eng J 321(Supplement C):622–634
Sutar H, Mishra SC, Sahoo SK, Maharana H (2014) Progress of bauxite residue utilization: an overview. Am Chem Sci J 4(3):255–279
Khale D, Chaudhary R (2007) Mechanism of geopolymerization and factors influencing its development: a review. J Mater Sci 42(3):729–746
Klauber C, Gräfe M, Power G (2011) Bauxite residue issues: II. options for residue utilization. Hydrometallurgy 108(1):11–32
Liu R-X, Poon C-S (2016) Utilization of red mud derived from bauxite in self-compacting concrete. J Clean Prod 112:384–391
Thakur R, Sant B (1983) Utilization of red mud. 2. Recovery of alkali, iron, aluminum, titanium and other constituents and the pollution problems. J Sci Ind Res 42(8):456–469
Vangelatos I, Angelopoulos G, Boufounos D (2009) Utilization of ferroalumina as raw material in the production of ordinary portland cement. J Hazard Mater 168(1):473–478
Satayanarayana P, Naidu G, Adiseshu S, Rao C (2012) Characterization of lime stabilized redmud mix for feasibility in road construction. Int J Eng Res Dev 3(7):20–26
Deelwal K, Dharavath K, Kulshreshtha M (2014) Evaluation of characteristic properties of bauxite residue for possible use as a geotechnical material in civil construction. Int J Adv Eng Technol 7(3):1053–1059
Kushwaha S, Kishan D (2016) Stabilization of bauxite residue by lime and gypsum and investigating its possible use in geoenvironmental engineering. Geo Chic 2016:978–988
Sabat AK, Mohanta S (2015) Efficacy of dolime fine stabilized bauxite residue-fly ash mixes as subgrade material. ARPN J Eng Appl Sci 10(14):5918–5923
Kumar S, Prasad A (2017) Parameters controlling strength of red mud-lime mix. Eur J Environ Civil Eng 1–15. https://doi.org/10.1080/19648189.2017.1304280
Mozumder RA, Laskar AI (2015) Prediction of unconfined compressive strength of geopolymer stabilized clayey soil using artificial neural network. Comput Geotech 69:291–300
Narendra B, Sivapullaiah P, Suresh S, Omkar S (2006) Prediction of unconfined compressive strength of soft grounds using computation[20]al intelligence techniques: a comparative study. Comput Geotech 33(3):196–208
Das SK, Samui P, Sabat AK (2011) Application of artificial intelligence to maximum dry density and unconfined compressive strength of cement stabilized soil. Geotech Geol Eng 29(3):329–342
Güllü H, Fedakar HI (2017) On the prediction of unconfined compressive strength of silty soil stabilized with bottom ash, jute and steel fibers via artificial intelligence. Geomech Eng 12(3):441–464
Suman S, Mahamaya M, Das SK (2016) Prediction of maximum dry density and unconfined compressive strength of cement stabilised soil using artificial intelligence techniques. Int J Geosynth Ground Eng 2(2):1–11
Kalkan E, Akbulut S, Tortum A, Celik S (2009) Prediction of the unconfined compressive strength of compacted granular soils by using inference systems. Environ Geol 58(7):1429–1440
Besalatpour A, Hajabbasi M, Ayoubi S, Afyuni M, Jalalian A, Schulin R (2012) Soil shear strength prediction using intelligent systems: artificial neural networks and an adaptive neuro-fuzzy inference system. Soil Sci Plant Nutr 58(2):149–160
Sharma LK, Singh R, Umrao RK, Sharma KM, Singh TN (2017) Evaluating the modulus of elasticity of soil using soft computing system. Eng Comput 33(3):497–507
Sakizadeh M, Mirzaei R, Ghorbani H (2017) Support vector machine and artificial neural network to model soil pollution: a case study in Semnan Province, Iran. Neural Comput Appl 28(11):3229–3238
Moayedi H, Rezaei A (2017) An artificial neural network approach for under-reamed piles subjected to uplift forces in dry sand. Neural Comput Appl. https://doi.org/10.1007/s00521-017-2990-z
Erzin Y, Gul TO (2014) The use of neural networks for the prediction of the settlement of one-way footings on cohesionless soils based on standard penetration test. Neural Comput Appl 24(3):891–900
Yilmaz I, Marschalko M, Bednarik M, Kaynar O, Fojtova L (2012) Neural computing models for prediction of permeability coefficient of coarse-grained soils. Neural Comput Appl 21(5):957–968
Erzin Y, Turkoz D (2016) Use of neural networks for the prediction of the CBR value of some Aegean sands. Neural Comput Appl 27(5):1415–1426
Erzin Y, Ecemis N (2017) The use of neural networks for the prediction of cone penetration resistance of silty sands. Neural Comput Appl 28(1):727–736
Edincliler A, Cabalar AF, Cagatay A, Cevik A (2012) Triaxial compression behavior of sand and tire wastes using neural networks. Neural Comput Appl 21(3):441–452
Ikizler SB, Vekli M, Dogan E, Aytekin M, Kocabas F (2014) Prediction of swelling pressures of expansive soils using soft computing methods. Neural Comput Appl 24(2):473–485
Tsai H-C, Tyan Y-Y, Wu Y-W, Lin Y-H (2013) Determining ultimate bearing capacity of shallow foundations using a genetic programming system. Neural Comput Appl 23(7):2073–2084
Buragohain M, Mahanta C (2008) A novel approach for ANFIS modelling based on full factorial design. Appl Soft Comput 8(1):609–625
Güllü H, Fedakar Hİ (2017) Response surface methodology for optimization of stabilizer dosage rates of marginal sand stabilized with sludge ash and fiber based on UCS performances. KSCE J Civil Eng 21(5):1717–1727
Güllü H, Fedakar Hİ (2016) Use of factorial experimental approach and effect size on the CBR testing results for the usable dosages of wastewater sludge ash with coarse-grained material. Eur J Environ Civil Eng 22(1):42–63
Dutta JR, Dutta PK, Banerjee R (2004) Optimization of culture parameters for extracellular protease production from a newly isolated Pseudomonas sp. using response surface and artificial neural network models. Process Biochem 39(12):2193–2198
Olgun M (2013) The effects and optimization of additives for expansive clays under freeze–thaw conditions. Cold Reg Sci Technol 93:36–46
Bayramov F, Taşdemir C, Taşdemir M (2004) Optimisation of steel fibre reinforced concretes by means of statistical response surface method. Cement Concr Compos 26(6):665–675
Chavalparit O, Ongwandee M (2009) Optimizing electrocoagulation process for the treatment of biodiesel wastewater using response surface methodology. J Environ Sci 21(11):1491–1506
Güneyisi E, Gesoğlu M, Algın Z, Mermerdaş K (2014) Optimization of concrete mixture with hybrid blends of metakaolin and fly ash using response surface method. Compos B Eng 60:707–715
Kobya M, Demirbas E, Bayramoglu M, Sensoy M (2011) Optimization of electrocoagulation process for the treatment of metal cutting wastewaters with response surface methodology. Water Air Soil Pollut 215(1–4):399–410
Murugesan K, Dhamija A, Nam I-H, Kim Y-M, Chang Y-S (2007) Decolourization of reactive black 5 by laccase: optimization by response surface methodology. Dyes Pigments 75(1):176–184
ASTM D1633-00 (2000) Standard test method for compressive strength of molded soil-cement cylinders. ASTM International. West Conshohocken. https://compass.astm.org/EDIT/html_annot.cgi?D1633+17. Accessed 15 Apr 2017
Gupta D, Kumar P, Mishra V, Prasad R, Dikshit P, Dwivedi S (2015) Bistatic measurements for the estimation of rice crop variables using artificial neural network. Adv Space Res 55(6):1613–1623
Hornik K, Stinchcombe M, White H (1989) Multilayer feedforward networks are universal approximators. Neural Netw 2(5):359–366
Witek-Krowiak A, Chojnacka K, Podstawczyk D, Dawiec A, Pokomeda K (2014) Application of response surface methodology and artificial neural network methods in modelling and optimization of biosorption process. Biores Technol 160:150–160
Salchenberger LM, Cinar E, Lash NA (1992) Neural networks: a new tool for predicting thrift failures. Decis Sci 23(4):899–916
Boger Z, Guterman H (1997) Knowledge extraction from artificial neural network models. Systems, man, and cybernetics. In: 1997 IEEE international conference on computational cybernetics and simulation, pp 3030–3035. IEEE
Sheela KG, Deepa SN (2013) Review on methods to fix number of hidden neurons in neural networks. Math Probl Eng 425740:1–10
Shibata K, Ikeda Y (2009) Effect of number of hidden neurons on learning in large-scale layered neural networks. In: ICCAS-SICE, 2009, pp 5008–5013. IEEE
Xu S, Chen L (2008) A novel approach for determining the optimal number of hidden layer neurons for FNN’s and its application in data mining. In: 5th international conference on information technology and applications (ICITA 2008), pp 683–686
Murata N, Yoshizawa S (1994) Amari S-i. Network information criterion-determining the number of hidden units for an artificial neural network model. IEEE Trans Neural Netw 5(6):865–872
Kulkarni P, Londhe S, Deo M (2017) Artificial neural networks for construction management: a review. Soft Comput Civil Eng 1(2):70–88
Consoli NC, da Silva Lopes L Jr, Heineck KS (2009) Key parameters for the strength control of lime stabilized soils. J Mater Civ Eng 21(5):210–216
Wei TK, Manickam S (2012) Response Surface Methodology, an effective strategy in the optimization of the generation of curcumin-loaded micelles. Asia Pac J Chem Eng 7:S125–S133
Miličević I, Šipoš TK (2017) Prediction of properties of recycled aggregate concrete. Građevinar 69(05):347–357
David FW (1992) Contouring: a guide to the analysis and display of spatial data with programs on diskette, 1st edn. Computer methods in the geosciences. Pergamon Press, Oxford, New York
Das SK, Basudhar PK (2006) Undrained lateral load capacity of piles in clay using artificial neural network. Comput Geotech 33(8):454–459
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The authors declare that there is no conflict of interest with any organization or entity with any financial interest (such as personal or professional relationships, affiliations, knowledge or beliefs) in the subject matter or materials discussed in this manuscript.
Electronic supplementary material
Below is the link to the electronic supplementary material.
Appendix A
Appendix A
1.1 Calculation of volumetric lime content (L v) and porosity of artificially cemented bauxite residue
In the present study, the following procedure has been followed:
-
a.
The amount of lime for each mixture was calculated based on the mass of dry bauxite residue.
-
b.
The dry density of the specimens was calculated as the dry mass of the bauxite residue and lime divided by the total volume of the sample.
1.2 Volumetric lime content (L V)
Defined as volume of lime with respect to total volume/volume of specimen and expressed in percentage.
Suppose,
We take 100 gm of bauxite residue and L gm of lime.
Then, total mass (mT) = (100 + L) gm
So,
where: γw = density of the specimen (defined as the weight of the water per its unit volume). γd = dry density of the specimen (defined as the weight of solid in a given volume). L = lime content (defined as the mass of lime with respect to total mass). GL = specific gravity of lime (defined as the ratio of the density of lime solids to the density of water). GRM = specific gravity of bauxite residue (defined as the ratio of the density of bauxite residue solids to the density of water).
Similarly,
Porosity (η) is defined as volume of voids with respect to total volume/volume of specimen and expressed in percentage.So,
Now,
Similarly, from Eq. (9), we can calculate,
Finally we get,
Rights and permissions
About this article
Cite this article
Kumar, S., Prasad, A. Strength retrieval of artificially cemented bauxite residue using machine learning: an alternative design approach based on response surface methodology. Neural Comput & Applic 31, 6535–6548 (2019). https://doi.org/10.1007/s00521-018-3482-5
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00521-018-3482-5