Вы находитесь на странице: 1из 12

ANN prediction of some geotechnical

properties of soil from their index


parameters

Parichehr Tizpa, Reza Jamshidi Chenari,


Mehran Karimpour Fard & Sandro
Lemos Machado

Arabian Journal of Geosciences

ISSN 1866-7511

Arab J Geosci
DOI 10.1007/s12517-014-1304-3

1 23
Your article is protected by copyright and
all rights are held exclusively by Saudi
Society for Geosciences. This e-offprint is
for personal use only and shall not be self-
archived in electronic repositories. If you wish
to self-archive your article, please use the
accepted manuscript version for posting on
your own website. You may further deposit
the accepted manuscript version in any
repository, provided it is only made publicly
available 12 months after official publication
or later and provided acknowledgement is
given to the original source of publication
and a link is inserted to the published article
on Springer's website. The link must be
accompanied by the following text: "The final
publication is available at link.springer.com”.

1 23
Author's personal copy
Arab J Geosci
DOI 10.1007/s12517-014-1304-3

ORIGINAL PAPER

ANN prediction of some geotechnical properties of soil


from their index parameters
Parichehr Tizpa & Reza Jamshidi Chenari &
Mehran Karimpour Fard & Sandro Lemos Machado

Received: 30 September 2013 / Accepted: 22 January 2014


# Saudi Society for Geosciences 2014

Abstract This paper presents artificial neural network pre- Introduction


diction models which relate compaction characteristics, per-
meability, and soil shear strength to soil index properties. In Permeability and shear strength are two vital parameters need-
this study, a database including a total number of 580 data sets ed for almost all geotechnical designs. These two parameters
was compiled. The database contains the results of grain size are governed by the degree of compaction. During the com-
distribution, Atterberg limits, compaction, permeability mea- paction process, the shear strength of soil increases and per-
sured at different levels of compaction degree (90–100 %) and meability decreases due to the reduction in the void ratio.
consolidated–drained triaxial compression tests. Comparison Moreover, the shear strength of soils decreases with increasing
between the results of the developed models and experimental water content. In some cases, shear strength and hydraulic
data indicates that predictions are within a confidence interval conductivity requirements should be addressed simultaneous-
of 95 %. To evaluate the effect of each factor on these ly, which means they should be achieved at the peak point of
geotechnical parameters, sensitivity analysis was performed compaction curve of soil. Thus, both the maximum dry den-
and discussed. According to the performed sensitivity analy- sity (MDD) and optimum moisture content (OMC) are essen-
sis, Atterbeg limits and the soil fine content (silt+clay) are the tial data for earthwork projects. Since laboratory tests for
most important variables in predicting the maximum dry determining permeability, maximum dry density, optimum
density and optimum moisture content. Another aspect that moisture content, and shear strength are time-consuming, it
is coherent from the sensitivity analysis is the considerable is desirable to develop models to predict compacted soil
importance of the compaction degree in the prediction of the characteristics based on the classification properties soils.
permeability coefficient. However, it can be seen that effective Many attempts have been made to relate these key parameters
friction angle of shearing is highly dependent on the bulk to the physical properties of soils. The physical properties
density of soil. used generally include grain size distribution, specific gravity,
and plasticity characteristics (liquid limit, plastic limit, shrink-
Keywords Artificial neural network . Compaction age limit, and plasticity index).
characteristics . Permeability . Soil shear strength . Sensitivity Rowan and Graham (1948), Davidson and Gardiner
analysis (1949), Turnball (1948), Jumikis (1946), Ring et al. (1962),
Ramiah et al. (1970), Nagaraj (1994), etc., are among the
researchers who tried to relate the compaction characteristics
P. Tizpa : R. Jamshidi Chenari : M. Karimpour Fard (*)
of soils to their index properties.
Department of Civil Engineering, Faculty of Engineering,
University of Guilan, P.O. Box 1841, Rasht, Iran The permeability of soils depends greatly on soil structure,
e-mail: mehran.karimpour@guilan.ac.ir void ratio, soil density, water content, degree of saturation, and
R. Jamshidi Chenari the type of permeant which the soil is exposed to. Various
e-mail: jamshidi_reza@guilan.ac.ir relationships between the permeability and grain size distri-
bution of the soil have been reported, such as those of Hazen
S. Lemos Machado
(1911), Zunker (1930), Carman (1937), Burmister (1954),
Department of Materials Science and Technology, Federal University
of Bahia, 02 Aristides Novis St., Salvador 40210-630, BA, Brazil Michaels and Lin (1954), Olsen (1962), and Mitchell et al.
e-mail: smachado@ufba.br (1965). Wang and Huang (1984), using a data bank including
Author's personal copy
Arab J Geosci

permeability, 320 data sets for modeling MDD and OMC, and
105 cases for modeling effective friction angle of shearing.
Different types of soils were used in these tests; therefore, the
results of this research could be valid for all types of soils.
As presented in Fig. 1, the database is obtained from
different sources, mainly from the geotechnical engineering
laboratory of the Federal University of Bahia (UFBA), Brazil.
Also, other cases from Wang and Huang (1984) and Mousavi
et al (2011) were added to the former source.
For each data set, the values of permeability, OMC, MDD,
compaction degree, friction angle, and soil index properties
(grain size curve, Atterbeg limits, and specific density) were
available. Table 1 gives the descriptive statistics of the vari-
ables used for the compaction characteristics and permeability
model developments. The variation ranges of the parameters
Fig. 1 Data sources used to compile the database used for effective friction angle model are summarized in
Table 2. It is noteworthy that the gravel content (Gc) was a
57 synthetic soils, developed several regression equations to coarse aggregate having a particle size coarser than 4.75 mm,
correlate the compaction characteristics and permeability with and the grain size of sand content (Sc) ranged from 4.75 to
the soil’s index properties. Koltermann and Gorelick (1995), 0.075 mm. As well, particles smaller than 0.075 mm were
Boadu (2000), Chapuis (2004), Sinha and Wang (2008), and, named as fine content (Fc). In addition, a quantity called
more recently, Cote et al. (2011) developed different numeri- fineness modulus (FM) is also computed. The value of fine-
cal and analytical models to estimate the hydraulic conductiv- ness modulus multiplied by 100 is equal to the sum of the
ity of soils based on their index parameters. percentages of particles coarser than 3/4 in., 3/8 in., no. 4, and
In recent years, some efforts have been made to develop a no. 100 mm. Note that the friction angle values in the database
correlation between the effective friction angle and the basic were determined by consolidated–drained triaxial tests.
properties of soil. Kayadelen et al. (2009) used artificial neural
network (ANN), genetic programming (GP), and adaptive Modeling method
neuro-fuzzy methods to predict the φ′ value of soils.
Mousavi et al. (2011) used GP and orthogonal least squares Artificial neural networks were employed to develop predic-
algorithm to present a correlation between the internal friction tion models. To evaluate the importance of each factor on the
angle and the physical properties of soils such as the fine and prediction models, a set of sensitivity analyses have been
coarse content, density, and liquid limit. Sezer (2013) performed.
employed nonlinear multiple regression, neuro-fuzzy, and Artificial neural networks are information-processing sys-
ANN methods to predict this soil parameter. tems whose architectures essentially mimic the biological
system of the brain (Goh 1994). ANNs have been successfully
applied to link independent variables to a series of dependent
Materials and methods ones, mainly where it is diverse to establish numerical equa-
tions. The use of ANNs has increased during the last decades
Database in various fields of geotechnical engineering, such as lique-
faction (Ural and Saka 1998; Najjar and Ali 1998), foundation
A database including a total number of 580 data sets was settlements (Sivakugan et al. 1998), reinforced soil (Ghiassian
compiled, in which 155 data sets were used for modeling the et al. 2006), and compaction characteristics of soils (Sinha and

Table 1 Descriptive statistics of the variables used for MDD, OMC, and permeability model

Parameter Gc (%) Sc (%) Fc (%) FM LL (%) PL (%) Gs Cd (%) MDD (kN/m3) OMC (%) Ka (cm/s)

Minimum 0 0 0 0.01 0 0 2.42 90 12.43 5.81 2.50E−11


Maximum 67 100 100 4.22 495 47 3.02 100 27.35 37.13 1.70E−03
Mean 7.39 40.05 52.31 1.71 77.58 24.60 2.70 95 17.31 17.44 3.49E−05
Standard deviation 10.15 21.13 22.83 1.17 104.48 10.47 0.06 4.04 1.97 5.67 1.91E−04
a
Permeability
Author's personal copy
Arab J Geosci

Table 2 Descriptive statistics of the variables used for effective friction


angle model

Parameter Cc Fc γ LL Shearing rate φ (deg)


(%) (%) (kN/m3) (%) (mm/min)

Minimum 1 15 14.11 1 0.024 19


Maximum 85 99 21.54 105 0.350 35.27
Mean 38.37 63.16 17.64 45.34 0.240 27.49
Standard 21.54 21.08 1.52 18.75 0.142 3.26
deviation

Wang 2008; Gunaydın 2009). In this study, a multilayer


perceptron network has been utilized to present the prediction
models.
A multilayer perceptron (MLP) is a feed-forward artificial
neural network model that maps sets of input data onto a set of
appropriate outputs. MLP utilizes a supervised learning tech-
nique called back-propagation for training the network. In the
application of MLP, data are categorized as input layer, output Fig. 3 Comparison between the predicted values of MDD and the actual
data
layer, and one or more hidden layers. The input patterns are
fed to the network for feed-forward computations to calculate
output patterns. The output patterns are compared with corre- Since the manner in which the database is used in the
sponding output patterns and the summation of the square of training and testing sets has a significant effect on the results,
the errors is calculated. The errors are then back-propagated the database was divided into several combinations of training
through the network using the gradient-descent rule to modify and testing sets until a robust representation of the whole
the weights and minimize the summed squared errors. population was achieved. To select an optimal combination
Figure 2 illustrates the typical ANN structure and the relation of training and testing sets, a statistical analysis considering
between the input and output parameters. the maximum, minimum, mean, and standard deviation was
In this study, a MLP network consisting of three hid- performed on the input and output parameters. The aim of the
den layers with nine, ten, and one neurons for the first, analysis was to ensure that the statistical properties of the data
second, and third hidden layers, respectively, has been in each of the subsets were as close to each other as possible
used. The number of neurons in the hidden layer was deter- and therefore represented the same statistical population
mined by training several networks with different numbers of (Rezania et al. 2008).
hidden neurons and comparing the predicted results with the Sensitivity analysis concerns the mathematical model rep-
desired output. resentation of a physical system and attempts to evaluate the

Fig. 2 Typical structure of ANN


Author's personal copy
Arab J Geosci

sensitivity of the output patterns to variations of input


patterns. The main issues in designing methods for regres-
sion sensitivity analysis are the choice of perturbation
scheme and the way to assess and measure any influence.
An appropriate method for perturbation is to delete obser-
vations individually or in groups. This approach is known
as case deletion and aims to assess the influence of an obser-
vation on the final results.
In this research, the type of sensitivity analysis was ANN-
based. At the first step, a MLP network was trained in the case
of each parameter with all data. In the trained network, each
neuron in a specific layer is connected to other neurons via
weighted connections in which scalar weights show the
strength of the connections. In the second step, one of the
inputs is removed from the ANN model by setting its scalar
weight to zero and then an output is achieved. In this way, all
the weighted connections between this variable and other
variables will be dropped from the model; therefore, the effect
of the removed input on the prediction of outputs could be
Fig. 4 Comparison between the predicted values of OMC and the actual pictured.
data

Fig. 5 Comparison of the ANN results on MDD model by excluding each parameter
Author's personal copy
Arab J Geosci

Fig. 6 Comparison of the ANN results on OMC model by excluding each parameter

Xn
Evaluation method ðM i −Pi Þ
COD ¼ 1 − Xi¼1
n
ð1Þ
Different statistical approaches were used to evaluate the perfor- i¼1
M 1 −M
mance of the prediction model. These parameters were coeffi-
cient of determination (COD), root mean squared error (RMSE), sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
Xn
and coefficient of residual mass (CRM). The following equa- i¼1
ðPi −M i Þ2
RMSE ¼  100 ð2Þ
tions are the mathematical expressions of these parameters. n

Table 3 Summary of the ANN’s performance

Outputs Statistics Total Ex. w1a (Gc) Ex. w2 (Sc) Ex. w3 (Fc) Ex. w4 (Gs) Ex. w5 (LL) Ex. w6 (PL)

MDD COD 0.92 0.57 0.57 0.67 0.64 0.74 0.42


RMSE 54.42 128.23 184.29 390.94 162.58 387.32 235.57
CRM 1.50E−04 3.30E−02 8.40E−02 2.1E−01 −1.40E−02 2.10E−01 −8.70E−02
OMC COD 0.92 0.84 0.83 0.83 0.85 0.61 0.65
RMSE 199.11 236.89 245.06 246.76 223.39 373.11 355.93
CRM 4.92E−05 1.40E−04 6.80E−05 1.30E−03 −8.70E−05 4.50E−04 5.90E−04

Ex. excluding
a
Scalar weights
Author's personal copy
Arab J Geosci

(PL). The only output was MDD. Hence, the input layer has
six neurons and the output layer has one neuron. Among 320
measured data sets, 290 sets (90 %) have been used for training
and 30 sets (10 %) have been used for testing the model.
Figure 3 shows the predicted values of MDD versus the
experimental data. As is clear from the graphs, the MLP model
gives very reliable estimates of the maximum dry density.
A similar ANN structure with the same input parameters
(Gc, Sc, Fc, Gs, LL, PL) has been used to estimate OMC.
Figure 4 shows a comparison between the results of the
developed ANN model and the experimental data. The ob-
tained values of COD and RMSE demonstrate the accuracy of
the developed model.
The results of the sensitivity analysis performed on the
MDD prediction model is presented in Fig. 5. In each graph,
one input variable is excluded (by setting its scalar weight to
zero) and the ANN model is trained for the five remaining
input parameters. As is obvious from the graphs, excluding
each parameter causes some extra scatter in the prediction
Fig. 7 Comparison between the predicted values of permeability coeffi- model for MDD.
cient and the actual data
The results of the sensitivity analysis performed on the OMC
prediction model are also presented in Fig. 6. The graphs
Xn
ðPi Þ illustrate that LL and PL are the most important variables in
CRM ¼ 1 − Xni¼1 ð3Þ the prediction of OMC. Moreover, it can be seen easily that
i¼1
ðM i Þ removing Fc leads to perturbation in the results.
Table 3 also presents a summary of ANN performance.
where Mi and Pi are the measured and predicted values, Based on the results, in the case of MDD, excluding Fc from
respectively, M is the mean of the measured values, and n ANN increases the RMSE to 390.9, indicating the significant
is the number of samples. The RMSE is the variance of the role of this parameter on the ANN predictions. Also, a CRM
residual error and should be minimized when the outputs fit a of 0.21 shows an underestimation of the MDD. LL and PL are
set of data. In the case of a perfect fitting, the RMSE is zero. the other important parameters in the prediction of MDD.
The lower the RMSE, the higher is the accuracy of the model Moreover, it is clear from the results that excluding Gs and
predictions. The coefficient of residual mass, CRM, is an PL causes an overestimation of the ANN prediction.
analysis of the difference between the measured and predicted Considering the OMC results, Fc, LL, and PL are the most
values. The optimum value of CRM is zero. Positive values of important variables in the ANN prediction.
CRM indicate underestimation and vice versa.
ANN prediction model for permeability

Results and discussions Six input variables were used for the ANN model for perme-
ability coefficient, including FM, LL, gravel content (Gc), sand
ANN prediction model for MDD and OMC content (Sc), fine content (Fc), and compaction degree (Cd). The
only output was LogK. Thus, the input layer has six neurons
Six input variables were used for the ANN model for MDD, and the output layer has one neuron. Among 155 measured data
including gravel content (Gc), sand content (Sc), fine content sets, 140 sets (90 %) have been used for training and 15 sets
(Fc), specific density (Gs), liquid limit (LL), and plastic limit (10 %) have been used for testing the model. Figure 6 shows the

Table 4 Summary of the permeability model performance

Output Statistics Total Ex. w1 (Gc) Ex. w2 (Sc) Ex. w3 (Fc) Ex. w4 (FM) Ex. w5 (LL) Ex. w6 (Cd)

LogK COD 0.99 0.51 0.54 0.78 0.88 0.33 0.48


RMSE 17.69 152.86 125.42 95.57 119.11 214.17 431.38
CRM 1.27E−05 1.01E−01 4.90E−02 −2.80E−02 −1.35E−01 2.20E−01 5.73E−01
Author's personal copy
Arab J Geosci

Fig. 8 Comparison of the ANN results on permeability model by excluding each parameter

predicted values of LogK versus the experimental data. Figure 7 conducted hydraulic conductivity tests on compacted clays.
shows an almost perfect prediction of hydraulic conductivity They plotted the variation of permeability at different densities
based on the employed index parameters; however, in the (Fig. 9). As could be seen, for the same soil, permeability
testing cases, the predictions exhibit a higher scatter. could be changed almost in one order of magnitude. Ahangar-
Table 4 and Fig. 8 illustrate the results of the sensitivity Asr et al. (2011) have presented the same results for the effects
analysis performed on the permeability prediction model. In of compaction degree on the permeability of soils. The results
accordance with Table 4, removing Cd increases the RMSE to of their parametric analysis show a decreasing trend for per-
431.4, which indicates that the compaction degree is of great meability by increasing the degree of compaction. In addition,
importance in predicting the permeability of soils. In this case, Mesri and Olson (1971), by comparing the permeability of
the obtained value for CRM shows an overestimation of different type of clays, stated that the size and arrangement of
permeability. As is clear from the results, Gc, Sc, and LL are particles have a significant effect on the permeability coeffi-
the other important variables in the ANN prediction model. cient. According to their studies, decreasing the void ratio,
Another aspect that is obvious from the results is that exclud- which is a result of increasing the compaction degree, will
ing Fc and FM causes an underestimation of the ANN model. cause the permeability of soils to decline.
As mentioned above, the degree of compaction has the
highest effect on the prediction of permeability coefficient. It ANN prediction model for effective friction angle
is well known that the volume of voids decreases due to an
increase in the degree of compaction, which causes the per- Five input variables were used for the ANN model for effec-
meability of a soil to decrease. Boynton and Daniel (1985) tive friction angle of shearing, including coarse content (Cc),
Author's personal copy
Arab J Geosci

Fig. 9 Variation of the


permeability at different dry
densities (Boynton and Daniel
1985)

fine content (Fc), liquid limit (LL), soil bulk density (γ), and
shearing rate (Sr). The only output was φ′. Thus, the input
layer has six neurons and the output layer has one neuron.
Among 105 measured data sets, 95 sets (90 %) have been used
for training and 10 sets (10 %) have been used for testing the
model. Figure 10 shows the predicted values of φ′ versus the
experimental data.
Table 5 and Fig. 11 also present the results of the sensitivity
analysis performed on the ANN model of effective friction
angle. It can be concluded from the results that the soil bulk
density has the highest influence on the prediction of the
effective friction angle. Another aspect that is coherent from

Table 5 Summary of the effective friction angle model performance

Output Statistics Total Ex. w1 Ex. w2 Ex. w3 Ex. w4 Ex. w5


(Cc) (Fc) (γ) (LL) (Sr)

φ COD 0.97 0.27 0.06 0.01 0.18 0.67


RMSE 51.63 517.71 723.03 3600.2 452.82 340.93
Fig. 10 Comparison between the predicted values of effective friction CRM 0.0018 −0.06 −0.08 4.61 0.01 0.1
angle and the actual data
Author's personal copy
Arab J Geosci

Fig. 11 Comparison of the ANN results on friction angle model by excluding each parameter

the result is that excluding the soil bulk density increases University of Bahia (UFBA), Brazil was utilized; however
the value of CRM significantly and leads to an underes- some other experimental data cases from the literature were
timation of effective friction angle. Also, the sensitivity added to this data bank.
analysis conducted by Mousavi et al. (2011) indicates the The results of the prediction models have been compared
importance of the bulk density on the friction angle of with the experimental data. Comparison of the results demon-
soils. strates that the developed ANN models provide highly accu-
rate predictions and that the existing models can be improved
with increasing the database. A major strength of the ANN
Summary and conclusion prediction models is their ability to improve as more data
become available without repeating the development proce-
This paper presents a set of ANN models for predicting the dures from the beginning.
compaction characteristics, permeability, and effective fric- Furthermore, a set of sensitivity analyses have been
tion angle of shearing based on the soil index properties. performed to illustrate the influence of each parameter
Since laboratory tests to determine these key parameters are on the ANN’s performance. To evaluate the performance
laborious and time-consuming, it is desirable to develop of the prediction models, three statistical approaches have
prediction models to estimate these parameters based on been utilized: coefficient of determination (COD), root
index parameters which are easy to measure. To do this, mean squared error (RMSE), and coefficient of residual
the databank of geotechnical laboratory of the Federal mass (CRM).
Author's personal copy
Arab J Geosci

References Mesri G, Olson RE (1971) Mechanism controlling the permeability of


clays. Clay Clay Miner 19:151–158
Michaels AS, Lin CS (1954) The permeability of kaolinite. Ind Engng
Ahangar-Asr A, Faramarzi A, Mottaghifard N, Javadi AA (2011) Chem 46:1239–1246
Modeling of permeability and compaction characteristics of soils Mitchell JK, Hopper DR, Campanella RC (1965) Permeability of
using evolutionary polynomial regression. Comput Geosci 37(11): compacted clay. J Soil Mech Found Div ASCE 91(No. SM4):41–66
1860–1869 Mousavi SM, Alavi AH, Mollahasani A, Gandomi AH, Arab Esmaeili M
Boadu FK (2000) Hydraulic conductivity of soils from grain-size distri- (2011) Formulation of soil angle of shearing resistance using a
bution. J Geotech Geoenviron Eng 126(8):739–746 hybrid GP and OLS method. Eng Comput. doi:10.1007/s00366-
Boynton SS, Daniel ED (1985) Hydraulic conductivity of compacted 011-0242-x
clay. J Geotech Eng 111(4):465–478 Nagaraj TS (1994) Analysis and prediction of compaction characteristics
Burmister DM (1954) Principles of permeability testing of soils, sympo- of soils. Unpublished
sium on permeability of soils. ASTM Spec Tech Publ 163:3–26 Najjar YM, Ali HE (1998) CPT-based liquefaction potential assessment:
Carman PC (1937) Fluid flow through granular beds. Trans Inst Chem a neuronet approach. ASCE Geotech Spec Publ 1:542–553
Eng 15:150 Olsen HW (1962) Hydraulic flow through saturated clays. Clay Clay
Chapuis RP (2004) Predicting the saturated hydraulic conductivity of Miner 9:131–161
sand and gravel using effective diameter and void ratio. Can Ramiah BK, Viswanath V, Krishnamurthy HV (1970) Interrelationship of
Geotech J 41(5):787–795 compaction and index properties. Proceedings of the Second
Cote J, Fillion MH, Konrad JM (2011) Estimating hydraulic and South East Asian Conference on Soil Engineering, Singapore,
thermal conductivity of crushed granite using porosity and pp 577–587
equivalent particle size. J Geotech Geoenviron Eng ASCE Rezania M, Javadi AA, Giustolisi O (2008) An evolutionary-based data
137(9):834–842 mining technique for assessment of civil engineering systems. J Eng
Davidson DT, Gardiner WF (1949) Calculation of standard proctor Comput 25(6):500–517
density andnoptimum moisture content from mechanical analysis, Ring GW, Sallgerb JR, Collins WH (1962) Correlation of compaction
shrinkage factors, and plasticity index. Proceedings of the HRB29, and classification test data. HRB Bull 325:55–75
pp 447–481 Rowan HW, Graham WW (1948) Proper compaction eliminates curing
Ghiassian H, Jamshidi R, Poorebrahim G (2006) Neural network analysis period in construction fills. Civ Eng 18:450–451
of silty sand reinforced by carpet wastes. Kuwait J Sci Eng 33(1): Sezer A (2013) Simple models for the estimation of shearing resistance
119–139 angle of uniform sands. Neural Comput & Applic 22(1):111–123
Goh ATC (1994) Nonlinear modeling in geotechnical engineering using Sinha SK, Wang MC (2008) Artificial neural network prediction models
neural networks. Aust Civ Eng Trans CE36(4):293–297 for soil compaction and permeability. Geotech Geol Eng J 26(1):
Gunaydın O (2009) Estimation of soil compaction parameters by using 47–64
statistical analyses and artificial neural networks. J Environ Geol 57: Sivakugan N, Eckersley JD, Li H (1998) Settlement predictions using
203–215 neural networks. Aust Civ Eng Trans CE40:49–52
Hazen A (1911) Discussion of ‘Dams on sand foundations’ by A. C. Turnbull JM (1948) Computation of the optimum moisture content in the
Koenig. Trans ASCE 73:199–203 moisture density relationship of soils. Proceedings of the Second
Jumikis AR (1946) Geology and soils of the Newark metropolitan area. International Conference on Soil Mechanics and Foundation
Journal of the Soil Mechanics and Foundations Division, ASCE Engineering, Rotterdam, Holland, pp 256–262
93(SM2), pp 71–95 Ural DN, Saka H (1998) Liquefaction assessment by neural networks.
Kayadelen C, Gu¨naydın O, Fener M, Demir A, O¨ zvan A (2009) A Electronic Journal of Geotechnical Engineering. http://www.ejge.
modeling of the angle of shearing resistance of soils using soft com/Ppr9803/Ppr9803.htm
computing systems. Expert Syst Appl 36:11814–11826 Wang MC, Huang CC (1984) Soil compaction and permeability predic-
Koltermann CE, Gorelick SM (1995) Fractional packing model for tion models. J Environ Eng ASCE 110(6):1063–1083
hydraulic conductivity derived from sediment mixtures. Water Zunker F (1930) Das verhalten des bodens zum wasser. Handbuch der
Resour Res 31:3283–3297 bodenlehre 6:66–220

Вам также может понравиться