ARBA MINCH UNIVERSITY SCHOOL OF POST GRADUATE STUDIES

HYDRAULICANDWATERRESOURCEDEPARTEMENT
SPECIALIZATION ON DAM ENGINERING
INSTRUMENTATION & DAM SAFETY ANALYSIS ASSIGNEMENT
PRESENTATION AND EVALUATION OF LITERATURE REVIEW
PAPER

By: Habtamu Fenta--------------------------------Id: PSAMIT/2371/10

Submitted to Elias Gebeyehu (Dr.Eng.)

FEB, 2021
EARTH DAM-BREACH ANALAYSIS DUE TO OVERTOPPING AND FLOOD INUDATION MAP

EARTH DAM-BREACH ANALAYSIS DUE TO

OVERTOPPING AND FLOOD INUDATION MAP

BY

Habtamu Fenta
Department of Hydraulic and Water Resource Engineering,
Arba Minch University, Ethiopia,
Email: [email protected]
FEB,2021

i
 EARTH DAM-BREACH ANALAYSIS DUE TO OVERTOPPING AND FLOOD INUDATION MAP

Contents
List of Table...............................................................................................................................................iii
List of figures.............................................................................................................................................iii
Abstract......................................................................................................................................................iv
1. Introduction.........................................................................................................................................1
1.1 SIGNIFICANCE OF THE STUDY.............................................................................................2
1.2 MAIN CAUSES OF EMBANKMENT DAMS FAILURE.........................................................2
1.3 Objective of the Study.................................................................................................................2
1.4 DAM-BREAK FLOOD FORECASTING History......................................................................3
1.5 Breach Parameter Estimation.......................................................................................................4
2. MATERIALS AND METHODOLOGIES..........................................................................................6
2.1 Materials......................................................................................................................................6
2.2 Methodologies.............................................................................................................................6
2.1.1 Data collection.....................................................................................................................6
2.1.2 Data analysis........................................................................................................................7
2.1.3 Study Area...........................................................................................................................7
2.1.4 Land Cover..........................................................................................................................8
2.1.5 Inflow..................................................................................................................................9
2.1.6 Reservoir Bathymetry..........................................................................................................9
2.1.7 Dam Failure Scenario..........................................................................................................9
3. Result and Discussion........................................................................................................................10
3.1 Reservoir Routing............................................................................................................................10
3.2 Flood Depth...............................................................................................................................10
3.3 Flood Arrival...................................................................................................................................11
4. Conclusions.......................................................................................................................................11
5. References.........................................................................................................................................12

ii
 EARTH DAM-BREACH ANALAYSIS DUE TO OVERTOPPING AND FLOOD INUDATION MAP

List of Table
Table 1-1 Classification Based on Potential Consequence.........................................................................3
Table 2-1 HEC-HMS data used...................................................................................................................8
Table 2-2 HEC-HMS data used...................................................................................................................9
Table 3-1 Flood Area.................................................................................................................................11

List of figures
Figure 1-1: Dam Breach..............................................................................................................................4
Figure 1-2: Dam breach cross section..........................................................................................................4
Figure 2-1Flow chart...................................................................................................................................7
Figure 2-2:Study area. BNPB [1] Figure 2-3: Before dam failure Figure 2-4:After dam failure........8
Figure 2-5:Land cover map of the study area Reservoir Routing................................................................8
Figure 2-6:Inflow hydrograph.....................................................................................................................9
Figure 2-7: Storage-Elevation curve............................................................................................................9
Figure 3-1: Outflow hydrograph due the dam failure................................................................................10
Figure 3-2: Flood depth map.....................................................................................................................11
Figure 3-3: Flood arrival time map............................................................................................................11

iii
 EARTH DAM-BREACH ANALAYSIS DUE TO OVERTOPPING AND FLOOD INUDATION MAP

Abstract
In earlier times historical study of dams conceived is vital. To continue advancing, the
engineering profession must review past problems periodically. “The issue of the safety of dams,
dikes and levees is one which is still topical. Disasters that occurred in the 20th century which
were connected with the failures of large dams, such as e.g., those of Malpasset (France, 1959),
Vajont (Italy, 1963), Teton (USA, 1976), Machhu II (India, 1979) and others, are well known
among specialists. A number of small embankment dams fail every year. During a flood in the
year 1997, several dikes almost 4 m high failed, and consequently, the urbanized areas behind
the dikes were flooded” [CITATION Zak151 \l 1033 ] . Dam-break due to overtopping is one of the
most common types of embankment dam failures. During the floods in August 2002 in the Czech
Republic, several small dams collapsed due to overtopping.
The Objective of this Literature review is to summaries knowledge about the resistance of
embankment dams to failures due to overtopping, about the mechanism of their failures and to
analyses particular methods of modelling for breaching process in embankment dams during
overtopping.
Keywords: Dam Breach, Model, HEC-RAS, HEC-HMS, Hydrograph

iv
 1. Introduction
From the early period of civilization, construction of dams conceived has been a long practice.
Dams are constructed for economic development, and their construction involves large
investments of money, natural and human resources. These benefits provided by dams come at a
risk due to their potential to fail and cause catastrophic flooding. Because the various types of
dams constructed around the globe, earth dams are the most common type and constitute the vast
majority of dams and overtopping is the most common cause of the failure of embankment dams [
CITATION VIJ96 \l 1033 ] . So, Mitigation of this risk is essential by simulate the potential failure
and protect against them for the benefit of the society as floods resulting from the failure of dams
produce devastating disasters. This dam failure caused for a number of reasons and its
consequence leads to requirement for preparation of dam breach due to over toping inundation
modeling and mapping to identify the flood risk and mitigate the consequences. In fact, Costa
reports that the average number of fatalities per dam failure is 19 times greater when there is
inadequate or no warning. Major causes of failures identified by Costa are overtopping due to
inadequate spillway capacity (34 percent), foundation defects (30 percent), and piping and
seepage (28 percent)[CITATION Ton98 \l 1033 ] .

Key questions related to embankment dam breach that must be known in the case of a risk
assessment are:

 Will dam failure occur? What are the loading thresholds that cause failure, and what is the
probability of failure given a particular loading?

 What are the consequences of failure in terms of loss of life and property damages?
Therefore, detailed information about the failure is needed, such as the amount of warning time,
and inundation levels and velocities at downstream locations. Most approaches rely either on
case study data from past dam failures or numerical models that do not simulate the erosion
mechanisms and flow regimes that are relevant to a dam breach. Case study data provide only
limited information (i.e., ultimate depth, width, and shape; peak discharge; maximum
overtopping depth; total time to fully fail embankment or drain reservoir), based on a relatively
small database of dam failures, primarily of small dams. Case study data are especially weak for
making predictions of the time needed to initiate a breach, the rate of breach formation, and the
total time required for failure. This is due to the difficulty of defining the exact point of failure
and the variations in interpretation of failure by the lay person who often is the only eyewitness
to a dam failure.
So, while planning and implementing dams, taking a good care of their safety is currently
becoming an important issue. Potential consequences of a dam failure should be understood for
planned dams and for those which are already built since the failure phenomena is unexpected an
immediate mitigation measures cannot be taken to hinder the breaching process dam for dams
under planning stage, one can use dam breach inundation information for classifying the dam
(and its hazard class) and this classification can then be used to estimate spillway discharge
capacity, seismic parameters, and others. Moreover, for both planned and existing dams, this
information is essential for preparedness and emergency action planning related to dam
failure[ CITATION Kib16 \l 1033 ].
The purpose of the dam breach analyses has been to illustrate how the flood wave propagates and
attenuates along the river valley from dam. In the present analyses the HEC-RAS model is used
for simulation of the flood wave caused by dam failure. This model is one of the most widely
accepted models of its kind.

1.1 SIGNIFICANCE OF THE STUDY

Preparing dam safety plans and hazard management strategies are unquestionably vital, since lots
of human lives have been lost and tremendous amount of economic crisis have been recorded
from dam failure events throughout the world in history. Setting out risk management,
emergency action plans or evacuation planning system to protect both lives and materials during
sudden dam failure phenomena and resulting flood waves is highly essential as it has been found
to be effective when applied during Baldwin Hill and San Fernando dam failures occurred in Los
Angeles, California USA where thousands could be evacuated.
Hence conducting pre-event analysis of Dam break and its consequences and forwarding the
hazard extent to public offices have great significances among which some are listed below:
Downstream settlement of inhabitants can be planned in such a way that evacuation of
the people during dam breach can be done rapidly saving as much lives as possible.
Protection dikes can be provided at both downstream banks with their top level
considerably higher than the propagating flood level, especially on the stretches along
residence areas.

1.2 MAIN CAUSES OF EMBANKMENT DAMS FAILURE

The main causes of failures of embankment dams are closely related to the erosion of
embankment materials caused by either overtopping or seepage erosion/piping. Ralston (1987)
discussed the mechanism of embankment erosion from overtopping. For non-cohesive
embankments, materials are removed from the embankment in layers by tractive stresses. The
erosion process from overtopping begins at a point where the tractive shear stress exceeds a
critical resistance that keeps the material in place. For cohesive embankments, breaching takes
place by head cutting. Usually, a head cut initiates near the downstream toe of the dam, and then
advances upstream until the crest of the dam is breached. The basic erosion mechanisms and
erosion rate as pointed out by Singh (1996) are different for granular and cohesive embankments.
For granular embankments, surface slips take place quickly due to the seepage existing on the
downstream slope; and hence granular materials are removed rapidly layer by layer. For cohesive
embankments, no seepage exists on the slope because of the low permeability. Instead, erosion
often begins at the embankment toe and advances upstream, undercutting the slope and in turn
causing the removal of large chunks of materials due to tensile or shear failure of the soil on the
overstepped slope[CITATION RPS13 \l 1033 ].

1.3 Objective of the Study

The general objective of this study is to model the dam breach process, apply it to the case of
Embankment dam and map the downstream area to be inundated by the flood.
The specific objectives are:
To determine the probable maximum flood (PMF), this may have the potential of
overtopping the dam and cause failure.
 To simulate the breach process applying different breach parameters using the model
HAC-RAS.
To run the model for different catastrophic scenarios and analyze the consequences of
the most hazardous one.
To predict the outflow hydrograph just at the outlet of the breach.
To prepare geographic data using HEC-GeoRAS and export to HEC-RAS for routing
process.
To route the peak outflow hydrograph on downstream channel and map the area that
shall be flooded.
To prepare various types of maps such as area, velocity, depth with time that can
describe the nature of flood propagation and coverage.
To simulate the breach process applying different breach parameters that shall be set up based on
regression equations which are developed by various scholars using historical dam break data.
To prepare various types of maps such as area, velocity, depth with time that can describe the
nature of flood propagation and coverage.

1.4 DAM-BREAK FLOOD FORECASTING History

The 1964 failure of Baldwin Hills Dam, near Los Angeles, California, and the near failure of
Lower van Norman (San Fernando) Dam in 1971 prompted the State of California to enact
statutes requiring dam owners to prepare dam failure inundation maps. The need for developing
procedures for estimating the breach hydrograph was thus born. Prior to the enactment of the
California statutes, very little was published regarding procedures for estimating dam breach
outflow hydrographs.
The numerous dam failures that occurred in the mid-1970's, including Buffalo Creek coal waste
dam (West Virginia, 1972), Teton Dam (Idaho, 1976), Laurel Run Dam and Sandy Run Dam
(Pennsylvania, 1977), and Kelly Barnes Dam (Georgia, 1977), led to comprehensive reviews of
Reclamation's dam safety program. Many of the reviews recommended that emergency
preparedness planning with inundation maps be emphasized. The Federal Guidelines for Dam
Safety, dated June 25, 1979, stated that inundation maps should be prepared. These events
highlighted the need for developing procedures for estimating dam breach outflow
hydrographs[ CITATION Ton98 \l 1033 ].
Table 1-1 Classification Based on Potential Consequence
 Loss of life Infrastructure,
Dam Class Loss of life economic loss Env. & cultural fac
Large potential for Estimated direct and indirect
multiple loss of life > (interruption of service) costs Loss or significant deterioration
Very high (Level 4) 1000 loss could exceed $100 million. Restoration LOW
Estimated loss of life Loss or significant deterioration
between 100 and 1 Restoration MEDIUM
high (Level 3) 000. > $1 million
Estimated loss of life Loss or significant deterioration
between 10 and Restoration HIGH
Moderate (Level 2) 100. > $100,000
Estimated loss of life Minimal economic loss, limited
low(level1) between 1 and 10. to dam owners property No Loss
No Potential for loss Minimal economic loss, limited
low(level 0) of life to dam owners property No Loss

1.5 Breach Parameter Estimation

Breach is an opening or a breakthrough of a dam which leads to a spontaneous release of the
water held behind it. It is commonly caused by a rapid erosion of the embankment material in
embankment dams and a sudden removal of all or part of the structure in concrete dams.
Breach parameters can be estimated using regression equations that are developed based on
historical dam failure data. Due to insufficient data on breach formation time, spatial breach
development during the failure phenomena and inadequate number of large dams and reservoirs
in the domain of analysis, these equations hold uncertainties.
Another technique to determine the breach parameters is using analytical models. The model
BREACH is one of the available mathematical models that can predict the breach characteristics
which are the size, shape and time of formation. This physically based model is developed based
on the principles of hydraulics, sediment transport, soil mechanics, geometric and material
properties of the dam, and the reservoir properties.
The breaching process is accomplished in two-time phases. The initiation phase is the initial time
at which overtopping or piping flow is observed the dam. During this phase, the flowing water
can be stopped and hence the dam can be saved from failure. In no measures are taken to stop the
water, erosion of embankment material will be triggered and any mitigation actions that are
undertaken to stop will be value less, this phase is called breach formation time.
It was observed from the collected historical dam failure events that in most of the cases the
ultimate breach shape is of trapezoidal shape. They concluded that for embankment dams, the
breach shape can be assumed to be triangular up to the time that the base of the embankment is
reached. Once the apex of the triangle reaches the foundation level, the breach develops forming
a trapezoidal section extending due to lateral erosion[ CITATION Ton98 \l 1033 ].

Figure 1-1: Dam Breach

Figure 1-2: Dam breach cross section
Breach Depth (Hb)
The dam is taken to breach up to the foundation level. Hb is taken as height at the breach
location.
Breach Width (Bavg),
The best prediction equation for dimensionless average breach width B́∗¿ B́/ Hb was found from
multiple linear regression analysis to be

ln B́∗¿ = − 1.30 + 0.278 * Mode + 0.316 * ln Vw*-----------------------------------Eq.1

where Mode= 1 if the failure is from overtopping, and 0 otherwise, and Vw*=Vw/Hb 3. The
regression model is based on 69 cases from the assembled data (breach width is missing from
four of the dam failures summarized, and reservoir volume is missing from another). The
coefficient of determination of the expression for the logarithm of B́ * = 0.652, and ln ^ B́ * =
0.4602 = standard error of estimate of the regression model for ln B́ *. Transforming Eq.1 gives

B́ * = 0.27koVw* 0.32-------------------------------------------------Eq.2
were

ko= {1.31.0forforovertopping failures

other failure modes

is a factor that accounts for the effect of failure mode. Dimensionless variables have been used in
Eq. (1) primarily to make the formula easier to use with any system of consistent units.
Additionally, using dimensionless parameters in the analysis effectively scales the data, reducing
the possibility that one or more values will have inappropriate influence on the relation.
Expanding the dimensionless terms yields the following expression for B́:

B́ = 0.27koVw 0.32Hb 0.04--------------------------------------------Eq.3

Because dimensionless parameters are used in Eqs. (1) and (2), variables having consistent
length units need to be used in Eq. (3), that is, if the unit of B́ is meters, then the unit of Vw is
cubic meters and the unit of Hb is meters. The assembled data show that overtopping failures
produce breaches that are, on average, about 30% wider than those caused by other failure
modes, as reflected in the Nonetheless, final average breach widths given by the approximation

B́ = 0.27koVw1/3-------------------------------------------------------------------------------- Eq.4
 Breach Side-Slope Ratio
An approximate estimate of the side slope ratio obtained from 70 cases side slope is missing
from four of the dam failures presented by
ln z = − 0.416 + 0.389*Mode -------------------------------------Eq.5
which has a small coefficient of determination of only 5% and a standard error of estimate ˆln z=
0.8273. Nonetheless, the regression model coefficient for Mode is highly significant, showing
that embankment side slope is larger when failure is caused by overtopping. This general
dependence is given adequately by failure mode alone and is not improved by the addition of
impoundment volume or embankment height. Transforming Eq. (5) gives the simple relation
z = {1.0 for overtopping failures
z = {0.7 for other failure modes
Breach formation time (tf)

Dimensionless breach formation time tf ∗¿ tf √( g /Hb) where g= gravitational acceleration, is

given by the regression equation
ln tf* = 4.15 + 0.495 ln Vw* ----------------------------------------Eq.6
which has a coefficient of determination of 0.929 and a standard error of estimate ˆln tf* =
0.2979. Breach formation time is missing from 51 of the cases reported, consequently Eq. (6) is
based on only 23 cases. Transforming Eq. (6) gives
tf* = 63.2Vw* 0.50 ------------------------------------------------------------Eq.7
which, after expanding the dimensionless terms, can be written as
tf = 63.2√(Vw/gHb 2) ------------------------------------------------------Eq.8

2. MATERIALS AND METHODOLOGIES

2.1 Materials
The materials and models used in the whole course of the study are:
Digital Elevation Model (DEM) with 30mx30m resolution, ArcGIS 10.1, HEC-RAS,
HEC-HMS, BREACH model, Google Earth and online MS Excel

2.2 Methodologies
2.1.1 Data collection
Primary data:
The primary data to be collected include:
 Surveying data to be collected along the cross section of the river at downstream of the
dam on every chainage that will be sufficient for the flood mapping process.
 Physical observation.
  Land use land cover data of the area prone to flooding. This can vary from land surface
formation type that will be used to estimate the manning's roughness 'n' to land use
pattern to evaluate the economic value that the area has rendered the inhabitants.
Secondary data:
All necessary secondary data such as:
 Daily rainfall data of meteorological stations located within and surrounding the
catchment.
 Land Cover
 Stream flow data of River
 Salient features of the dam
 Salient features of the reservoir
 Reservoir area-elevation-volume curves
 Geotechnical information on the composition of the dam and foundation.
 Laboratory test results.
2.1.2 Data analysis
The procedure followed to conduct is;
First hydrological study is conducted to determine the probable maximum flow (PMF)
that enters the reservoir and may overtop the dam thereby endangering the dam for
failure.
Dam breach analysis is then done to predict the breach size and estimate the outflow
hydrograph.
Having determined the outflow magnitude, the flood is routed through the downstream
channel and flood plain using HEC-RAS in order to determine the flood prone area, so
that flood early warning system can be set and emergency action plans can be made.

Figure 2-3Flow chart

 2.1.3 Study Area
“Way Ela Natural Dam is located in Negeri Lima Village, Leihitu District, Ambon Island.
The village had a mean elevation between 0-700 m above sea level. The main River are the
Way Ela River which in 2012 dammed to form a natural dam due to avalanches.

Figure 2-4:Study area. BNPB [1] Figure 2-5: Before dam failure Figure 2-6:After dam failure

2.1.4 Land Cover

In this study the land cover is used to define the Runoff coefficient (C). The Way Ela
Watershed Land is covered by forest with Runoff coefficient is 0.5

Figure 2-7:Land cover map of the study area Reservoir Routing

The reservoir routing was calculated using HEC-HMS to estimate the amount of discharge
when the dam failed. There are three main parameters used as the input for reservoir routing
in HEC-HMS, inflow, reservoir bathymetry and the dam failure scenarios. The output from
this reservoir routing is the outflow discharge due the failure of Way Ela Natural Dam.
Table 2-2 HEC-HMS data used

Data Values
Watershed Area (m2) 12.56
River Length (m) 6.85
Land Cover (Runoff Coefficient) 0.5
Inflow Hydrograph Figure
Dam Top Elevation (m) 215.66
 Dam Top Length (m) 300
Normal Water Level (m) 197
2.1.5 Inflow
The inflow of the reservoir was calculated using rainfall at the time of the dam failure. The
condition of this inflow at the time of dam failure was estimated with SynderAlexeyev Synthetic
Unit Hydrograph based on the characteristic of the catchment area as given in table 2.1. The
rainfall detected at the dam failure by Ambon Meteorological Agency (BMKG) was 432 mm.
The results are shown in the Figure 2-6 along with the capacity of the spillway of the natural
dam. It shows that inflow exceeds the spillway capacity.

Figure 2-8:Inflow hydrograph

2.1.6 Reservoir Bathymetry
Reservoir Bathymetry of the Way Ela natural dam was analyzed based on previous studies [7],
the result is given in figure 2-6 along with the dam elevation (table2.1).

Figure 2-9: Storage-Elevation curve

2.1.7 Dam Failure Scenario
In this study a simulation was conducted for the dam failure with the assumption that the failure
was caused by overtopping based on previous studies.
Table 2-3 HEC-HMS data used
The scenarios parameter is given in table 2. The trigger elevation is set to + 200 m, is 3 m above
spillway = 197 m, because based on chronology when the dam failure sliding occurs on the dam
body because the spillway only finishes 70% and then collapses. The dam failure development
time is 5 hours but lengthened to 10 hours due to emergency spillway [7]. The Bottom Width
parameter is 2 × dam height.
Flood Propagation Model
The 2D flood propagation was modeled using HEC-RAS v5. The HEC-RAS v5 solves the full
2D Saint Venant equations.

where h is the water depth (m), p and q are the specific flow in the x and y directions (m2 /s), is
the surface elevation (m), g is the acceleration due to gravity (m/s), n is the Manning resistance,
is the water density (kg/m3 ), xx, yy and xy are the components of the effective shear stress and f
is the Coriolis (s-1 ).

3. Result and Discussion

3.1 Reservoir Routing
Outflow from dam failure is simulated uses HEC-HMS. The overtopping failure scenario is
selected refers to previous study. Inflow Hydrograph was analyzed using the Synder-Alexeyev
Synthetic Unit Hydrograph. The result of this routing process will be shown by Figure 3.3.

Figure 3-10: Outflow hydrograph due the dam failure

Figure 3.1 shows the outflow peak discharge due the dam failure by the overtopping failure is
1,268.4 m3 /s.

3.2 Flood Depth

The relative error of overtopping failure shown in Table 3.1. Table 3.1 shows the result that this
overtopping failure has a small relative error compared with the inundation of the orthophoto
image. It can be concluded that the previous dam failure was caused by overtopping. Time
requirement to execute this model is 30 minutes.
Table 3-4 Flood Area

Figure 3-11: Flood depth map

3.3 Flood Arrival

Figure 3-12: Flood arrival time map

Figure 3-3 shows that the arrival time of the flood caused by the dam failure when it reaches the
village is between 2 hours and in the downstream is between 5-5.5 hours.

4. Conclusions
A two-dimensional (2D) numerical model, HEC-RAS v.5, is used to simulate the over land flow
generated by a dam failure. The dam failure itself is simulated using HECHMSv.4. The model
was applied to the dam break case of the Way Ela natural dam. The results from the study are the
inundation area, depth, and arrival time. The inundation area was verified using the data from
orthophoto. It was found that both areas correspond well with a very small relative error (0.42%).
The inundation depth in the village there are up to 6 m, which agrees well with the reported
depth. The arrival time of the flood is 2 hours. These information’s are very important to propose
a mitigation plans with respect to possible dam break in the future[ CITATION Bag18 \l 1033 ]”.
 5. References

[1] J. J. a. J. Ř. Zakaraya Alhasan, "STUDY OF DAM-BREAK DUE TO OVERTOPPINGOF FOUR

SMALL DAMS IN THE CZECH REPUBLIC," Institute of Water Structures, Faculty of Civil
Engineering, Brno University of Technology, Veveří 95, 602 00, vol. 63, p. 717–729, 2015.

[2] V. Y. P. SINGH, DAM BREACH MODELING TECHNOLOGY, Baton Rouge, U.SA. : Louisiana
State University, 1996 .

[3] T. L. Wahl, "Prediction of Embankment Dam Breach Parameters," U.S. Department of the Interior
Bureau of Reclamation Dam Safety Office, pp. DSO-98-004, July 1998.

[4] K. Tesemma, "Dam Breach Modeling and Flood Inundation Mapping For Middle Awash Dam,"
Addis Ababa University Institute of Technology School of Civil and Environmental Engineering, Nov,
2016.

[5] A. K. R. P. Sharma, "Case Histories of Earthen Dam Failures," International Conference on Case
Histories in Geotechnical Engineering. 8., 2013.

[6] M. B. A. F. S. N. I. H. Bagus Pramono Yakti, "2D Modeling of Flood Propagation due to the Failure
of Way Ela Natural Dam," Graduate School of Civil Engineering, Institut Teknologi Bandung, Jalan
Ganesha 10, Bandung 40132, Indonesia, p. MATEC Web of Conferences 147, 2018.