JTUS, Vol. 02, No. 10 October 2024
E-ISSN:
2984-7435, P-ISSN: 2984-7427
|
DOI: https://doi.org/10.58631/jtus.v2i10.112 |
Rainfall
Prediction using Backpropagation Method and GIS for Disaster Mitigation Mapping
Farhan Fadhilah1*, Muhamad Reza2,
Awliya Tribhuwana3*
Universitas Swadaya Gunung
Jati, Indonesia
Email: [email protected]1,
[email protected]2, [email protected]3
INTRODUCTION
Cirebon City has an average elevation of 5 metres above
sea level (B. K. Cirebon, 2024). The terrain in Cirebon Regency is divided into lowland
areas and partly highland areas (B. P. S. K. Cirebon, 2024). The altitude in Majalengka Regency is between 19 - 857
metres above sea level (Majalengka, 2024). The increase in rainfall is caused by changes in weather
patterns due to the ever-changing atmosphere and the presence of cold air flow (Aslim et al., 2023).The problem is that due to changes in weather or climate
that often change at this time, coupled with the topographic conditions of
Cirebon and Majalengka which are partly low-lying and close to the sea, it is
feared that if disaster mitigation is not carried out early on, flood disasters
will occur due to lack of early preparation for the prevention of runoff
discharge. With the expected results of monthly rainfall values, runoff
discharge, and mapping for the next 2, 5, 10, and 25 years. With the research
location in Cirebon and Majalengka. This research aims to determine the
prediction of rainfall, runoff discharge, and its mapping as a flood mitigation
effort in the Cirebon and Majalengka areas.
Figure 1. Research Location
Source: Google
Earth
Rainfall is the height of rainwater that falls on a land
surface which is considered not to evaporate, not to seep, and not to flow (Triwahyuni et al., 2020). Rain that falls to the ground surface will form a runoff
that will eventually flow back to the sea, some will be absorbed into the soil
surface (infiltration), and others will continue to flow down (percolation) to
go to a saturated place below the groundwater surface (Ikhwan et al., 2022). The origin of incoming flow is from rainfall, and the
origin of outgoing flow is from surface runoff, evaporation, and infiltration (Muttaqin & Farhan, 2021).
Prediction is a systematic activity in estimating
something with the possibility of happening in the future or has been estimated
based on pre-existing data (Adiguno et al., 2022). Rainfall prediction is one of the important activities
in climate forecasting, because precise and accurate results will greatly help
in planning and management for water resources, flood warnings, construction
activities, flight operations, and so on (Yusuf et al., 2022). One of the factors included in weather forecasting is
water, where water is very important for the survival of living things to meet
their needs and activities (Zahran, 2023). Digital mapping is an activity in processing digital
format maps, and uses hardware such as computers, as well as related software
in processing (Thariq, 2020). Mapping can present information related to a place by
displaying location data according to the coordinates on the map. (Sitio et al., 2021). Mitigation is an activity carried out to minimise and
reduce the risk of damage caused by disasters (Afrian, 2020). Disaster mitigation is a process to increase community
knowledge in understanding and dealing with hazard characteristics, changing
actions and mindsets so that the quality of natural resources is maintained (Qurrotaini & Nuryanto, 2020). Mitigation and control of natural disasters related to
meteorological forecasts must be carried out from the beginning with high
accuracy and in an easy-to-understand delivery (Oktaviani & Afdal, 2013). Examples of mitigation for flood disasters that can be
carried out include the construction of flood embankments, and issuing a
regulation to minimise the occurrence of disasters (Tribhuwana et al., 2021).
In the research conducted (Sheikhi et al., 2023) The content of the discussion is to compare
six hybrid artificial intelligent models developed to predict daily rainfall in
urban areas, by combining the firefly optimisation algorithm (FA), invasive
weed optimisation algorithm (IWO), genetic particle swarm optimisation
algorithm (GAPSO), neural network (ANN), group method of data handling (GMDH),
and wavelet transformation. Research (Kyojo et al., 2024) discussed the analysis of annual maximum
rainfall data for 31 years, starting from 1990 to 2020, and the Generalised
Extreme Value (GEV) model proved to be the best for modelling extreme rainfall
at all stations. Using three methods to estimate it, namely L-moments method,
Maximum Likelihood Estimation (MLE), and Bayesian Markov chain Monte Carlo
(MCMC) are used to estimate GEV parameters and future return rates. The content
of the discussion in the research (Silva et al., 2023) was to predict monthly rainfall, one month in
advance, in four municipalities in the Belo Horizonte City region using ANNs
trained with different climate variables. (Aslim et al., 2023) discusses the identification of the right
combination of network architecture, learning rate, and epoch in predicting
each rainfall post in Maros Regency. And also predicts the monthly rainfall
profile in 2021 - 2025 in Maros Regency. Other research by (Setiawan & Barokah, 2022) which predicts rainfall using the auto
correlation function in artificial neural networks with the backpropagation
method to deal with the impact caused by high rainfall such as hampering
population mobility and distribution of goods, especially in the port area.
Based on the above background, the purpose of
this research is to predict rainfall, runoff discharge, and map the potential
for flooding in the Cirebon and Majalengka areas as a disaster mitigation
effort. This research is expected to provide accurate information about
rainfall and runoff discharge predictions for the next 2, 5, 10, and 25 years.
This information will be useful in assisting regional spatial planning,
infrastructure development that is more resilient to floods, and minimizing the
risk of damage due to natural disasters in the research area. The benefit of
this research is to make an important contribution to water resources
management and flood disaster mitigation efforts. With accurate prediction
results, local governments and related agencies can formulate more effective
policies in reducing flood risks, such as building embankments, improving
drainage systems, and controlling uncontrolled land use changes.
METHOD
The method used is the Backpropagation Method. The
backpropagation method is a method that resembles the way the neurons system
works in the brain of humans to learn patterns and is used to optimise weights
and refractions, then the error rate (mean square error) gets smaller, saying
that the estimated value better represents the actual value (Siregar, 2019). For the stages of predicting rainfall in this research,
it starts from inputting the original data, namely entering the monthly
rainfall amount data that has been obtained for the three rainfall posts,
namely BMKG Jatiwangi, BMKG Kertajati, and BMKG Penggung. This rainfall amount
data is entered and arranged in different excel or each for each rainfall post
which is used as initial data or original data. Next, normalise the original
data, namely creating and compiling the amount of rainfall data in excel,
select one of the rainfall station excel first to be entered into the
programming with the help of Matlab 2021a software to get the normalisation
value. When determining the training data and training target data, we first
test the accuracy of the prediction results with existing data. In the neural
network programme, the artificial neural network design that will be used is
12-1000-1, meaning that the network starts with 12 values for the input layer
(rainfall data for 12 months), 1000 neurons in the hidden layer, and only one
value for the output layer, which is the rainfall data in the target year.
Finally, test and predict by inputting the test target data in the form of
rainfall data that has been pre-selected from the beginning. Then calculate the
RMSE (Root Mean Square Error) value to determine the prediction validation
level of the rainfall pattern.
In this research, quantitative research methods are used
because they focus on collecting numerical data and statistical analysis to
objectively measure and analyse phenomena. This method is used to analyse data
and problems using sources that can be considered in writing articles. In this research, the type of data
used is secondary data, namely data obtained from related agencies, namely BMKG
Kertajati, Jatiwangi, and Penggung Stations, as well as from some literature
such as BMKG Online related to rainfall data. For the stages of data analysis
in this research such as predicting rainfall, calculating rainfall runoff
discharge, and map modelling.
In the backpropagation artificial neural
network (JST), which is used is a binary sigmoid activation function (valued
between 0 and 1), can use the formula (Tamaji et al., 2022):
........................................................................................ �(1)
Calculating the RMSE (Root Mean Square Error) value to
determine the level of validation of predictions of rainfall patterns, with the
formula:
RMSE = 
.................................................................................. (2)
There are several methods that can be used to calculate
rainfall for return periods, such as:
Gumbel Method
................................................................................................... �(3)
Iwai Kadoya Method
.................................................................................................... (4)
Log Pearson III Method
.................................................................... (5)
Rainfall intensity is the amount of rainfall that occurs
in a period and the rainwater is concentrated. (Tribhuwana & Prasetyo, 2020). The intensity formula is:
................................................................................................... � (6)
Runoff water discharge is the amount of rainwater per unit
time that does not absorb into the soil (infiltration), so that it is carried
out by drainage channels. (RAMDHANI & YUSTIANA, 2023). The formula used is the rational method discharge
formula:
Rational Method
�....................................................................................... (7)
RESULT AND DISCUSSION
The results of the sum of rainfall prediction
values as shown in tables 1, 2, and 3 per month for the years 2025, 2028, 2033,
and 2048 at BMKG Kertajati, BMKG Jatiwangi, and BMKG Penggung stations.
Table 1. Recapitulation of Predicted Monthly Rainfall
Amount at BMKG Kertajati Station (mm)
|
YEAR |
JAN |
FEB |
MAR |
APR |
MAY |
JUN |
JUL |
AUG |
SEP |
OCT |
NOV |
DES |
|
2025 |
375 |
474 |
322 |
326 |
221 |
85 |
18 |
29 |
34 |
83 |
143 |
248 |
|
2028 |
338 |
364 |
378 |
307 |
257 |
109 |
40 |
18 |
42 |
56 |
112 |
173 |
|
2033 |
192 |
334 |
372 |
373 |
317 |
262 |
114 |
39 |
18 |
37 |
54 |
109 |
|
2048 |
368 |
367 |
521 |
394 |
162 |
65 |
11 |
49 |
93 |
280 |
318 |
294 |
Source: Calculation
Result
Table 2. Recapitulation of Predicted Monthly Rainfall
Amount at BMKG Jatiwangi Station (mm)
|
YEAR |
JAN |
FEB |
MAR |
APR |
MAY |
JUN |
JUL |
AUG |
SEP |
OCT |
NOV |
DES |
|
2025 |
427 |
489 |
342 |
358 |
222 |
78 |
17 |
26 |
30 |
74 |
138 |
293 |
|
2028 |
425 |
410 |
413 |
334 |
255 |
86 |
30 |
14 |
44 |
57 |
114 |
223 |
|
2033 |
464 |
419 |
491 |
347 |
329 |
109 |
47 |
8 |
57 |
36 |
103 |
209 |
|
2048 |
630 |
331 |
567 |
232 |
332 |
28 |
67 |
2 |
153 |
71 |
225 |
212 |
Source: Calculation
Result
Table 3. Recapitulation of Predicted Monthly Rainfall
Amount at BMKG Penggung Station (mm)
|
YEAR |
JAN |
FEB |
MAR |
APR |
MAY |
JUN |
JUL |
AUG |
SEP |
OCT |
NOV |
DES |
|
2025 |
436 |
487 |
506 |
326 |
215 |
54 |
21 |
9 |
33 |
37 |
79 |
191 |
|
2028 |
428 |
492 |
487 |
378 |
252 |
62 |
21 |
8 |
25 |
37 |
72 |
155 |
|
2033 |
422 |
481 |
529 |
378 |
285 |
69 |
24 |
6 |
25 |
29 |
67 |
141 |
|
2048 |
440 |
484 |
531 |
378 |
276 |
62 |
22 |
5 |
25 |
29 |
68 |
151 |
Source: Calculation
Result
Figure 2 Rainfall Prediction Chart
Source: Calculation Result
It
can be seen from the recapitulation table of rainfall prediction results above,
at BMKG Kertajati Station in 2025 the highest amount of rainfall occurred in
February (474 mm), in 2028 the highest in February (364 mm), in 2033 the
highest in April (373 mm), and in 2048 the highest in March (521 mm). At BMKG
Jatiwangi Station in 2025 the highest amount of rainfall occurred in February
(489 mm), in 2028 the highest in January (425 mm), in 2033 the highest in March
(491 mm), and in 2048 the highest in January (630 mm). And at BMKG Penggung
Station in 2025 the highest amount of rainfall occurred in March (506 mm), in
2028 the highest in February (492 mm), in 2033 the highest in March (529 mm),
and in 2048 the highest in March (531 mm).
Mapping of monthly rainfall prediction results for 2025,
2028, 2033, and 2048 at BMKG Kertajati, BMKG Jatiwangi, and BMKG Penggung
stations.
Figure 3 Rainfall Prediction Map
Source: Calculation
Result
It can be seen in Figure 3, for the distribution of the
amount of rainfall prediction results in 2025, with a maximum rainfall amount
of 2494 mm and a minimum of 2358 mm. In 2028, the maximum rainfall amount will
be 2417 mm, and the minimum will be 2194 mm. In 2033, the maximum rainfall
amount will be 2619 mm and the minimum will be 2221 mm. And in 2048 with a
maximum rainfall amount of 2922 mm and a minimum of 2471 mm.
There are several conditions that must be met before
choosing a rainfall distribution method.
Table 4. Type and Condition of Rainfall
Distribution
|
Distributin Type |
Terms |
|
Gumbel |
Cs ≤ 1.1396 |
|
Ck ≤ 5.4002 |
|
|
Iwai Kadoya & Log Pearson III |
Cs ≠ 0 |
Source: Calculation
Result
Since
the methods used in the calculation of rainfall distribution are the Gumbel,
Iwai Kadoya, and Log Pearson III methods, only eligible methods can be used for
further calculations before using any of the methods for rainfall distribution.
The results of the calculation of rainfall distribution at
BMKG Kertajati, BMKG Jatiwangi, and BMKG Penggung stations against the type and
conditions of rainfall distribution can be seen in Table 5.
Table 5. Results of Rainfall Distribution Calculation
|
Calculation Result |
Description |
||||||
|
Kertajati |
Jatiwangi |
Penggung |
Maksimum |
Kertajati |
Jatiwangi |
Penggung |
Maksimum |
|
3.551 |
0.388 |
-1.415 |
0.546 |
Unqualified |
Qualified |
Unqualified |
Qualified |
|
15.829 |
2.371 |
5.576 |
2.589 |
||||
|
3.551 |
0.388 |
-1.415 |
0.546 |
Qualified |
Unqualified |
Memenuhi |
Unqualified |
Source: Calculation
Result
It
can be seen in Table 5, that the results of the calculation of the Cs and Ck
values at each station, for stations that qualify using the Gumbel Method are
the Jatiwangi BMKG Station and the maximum average of the three BMKG stations.
While those that qualify using the Iwai Kadoya and Log Pearson III methods are
Kertajati BMKG Station and Penggung BMKG Station.
After determining the method that qualifies the rainfall
distribution for each BMKG station, the next step is to conduct a Chi-Square
Test to determine which rainfall data or values can be accepted or used.
Table 6. Recap of Chi-Square Test
|
Chi - Square Test Recap |
||||
|
Station |
Methods |
Count |
Critical |
Description |
|
Maximum |
Gumbel |
4.00 |
5.991465 |
Accepted
(Result ≤ Critical) |
|
Jatiwangi |
4.67 |
Accepted
(Result ≤ Critical) |
||
|
Kertajati |
Iwai
Kadoya & Log Pearson III |
34.67 |
Not
Accepted (Result ≥ Critical) |
|
|
Penggung |
8.67 |
Not
Accepted (Result ≥ Critical) |
||
Source: Calculation
Result
The
recapitulation results of the Chi-Square Test on each rainfall data in the
Gumbel Method, Iwai Kadoya, and Log Pearson III between the calculated results
and the critical value. And rainfall data that can be accepted or used is
rainfall data at BMKG Jatiwangi Station and the maximum average of the three
BMKG stations.
Because the data can be accepted or used for the
calculation of rainfall intensity, the data to be used is the maximum average
rainfall data because it is considered to represent all areas in the research
area and the highest possible amount of rainfall that occurs. For the
calculation of rainfall intensity using the Mononobe Method and the results of
the calculation of rainfall intensity can be seen in Table 7.
Table 7. Mononobe Method Rainfall Intensity
(mm/h)
|
t (Minutes) |
Rainfall Intensity |
|||
|
I2 |
I5 |
I10 |
I25 |
|
|
10 |
243.516 |
262.849 |
275.650 |
282.872 |
|
20 |
153.405 |
165.585 |
173.649 |
178.198 |
|
30 |
117.070 |
126.365 |
132.519 |
135.991 |
|
40 |
96.639 |
104.312 |
109.392 |
112.258 |
|
50 |
83.281 |
89.893 |
94.271 |
96.741 |
|
60 |
73.750 |
79.605 |
83.482 |
85.669 |
|
70 |
66.547 |
71.830 |
75.328 |
77.302 |
|
80 |
60.879 |
65.712 |
68.912 |
70.718 |
|
90 |
56.282 |
60.750 |
63.708 |
65.377 |
|
100 |
52.464 |
56.629 |
59.387 |
60.943 |
|
110 |
49.234 |
53.143 |
55.731 |
57.191 |
|
120 |
46.459 |
50.148 |
52.590 |
53.968 |
Source: Calculation
Result
The
time used or planned in the calculation of rainfall intensity is for two hours
(120 minutes) for each annual return period of 2, 5, 10, and 25 years. The
graph of the calculation results of rainfall intensity values in Table 7 for
maximum rainfall data using the Mononobe Method can be seen in Figure 4 below.
Figure 4. Mononobe Method Rainfall Intensity
Chart
Source: Calculation
Result
The
division of the research area using the thiessen polygon method using three
points from the BMKG stations reviewed, namely Kertajati BMKG Station,
Jatiwangi BMKG Station, and Penggung BMKG Station located in Cirebon and
Majalengka is shown in Figure 5.
Figure 5. Thiessen Polygon Method Area Division
Source: Calculation
Result
After dividing the area in the research area using the Thiessen
polygon method, the area value of each cut area was obtained in units of km2,
with one BMKG station per area. The data can be seen in Table 8.
Table 8. Area Data
|
No |
Station |
Area |
Area (Km2) |
|
1 |
Kertajati |
A1 |
355.194 |
|
2 |
Jatiwangi |
A2 |
935.136 |
|
3 |
Penggung |
A3 |
1028.284 |
Source: Calculation
Result
In addition to the
area within the research area, the area of watersheds included in the research
area is needed for further calculations. The watersheds that were selected
because they were included in the research area were Cibuaya DAS, Ciwaringin DAS,
and Cipager DAS. However, the watershed area used is the only one included in
the research area. The watershed area data can be seen in Table 9.
Table 9. DAS Area Data Per Area
|
No |
DAS |
Name Area |
Area (Km2) |
|
|
1 |
Cibuaya |
A1 |
96.735 |
|
|
2 |
Cibuaya +� Ciwaringin |
A2 |
146.541 |
|
|
3 |
Ciwaringin + Cipager |
A3 |
117.256 |
|
Source: Calculation
Result
The
coefficient values in Table 10 are derived from the percentage of built and
unbuilt area of each research area, and the data is available on the official
BPS website of the research area.
Table 10. Coefficient Value Data (C)
|
Location |
Area (Ha) |
Area (%) |
Coefficient |
Coefficient (%) |
Coefficient Value� ( C ) |
|
Majalengka |
120424.00 |
51.94 |
0.53 |
0.28 |
0.56 |
|
Cirebon |
111436.00 |
48.06 |
0.59 |
0.28 |
|
|
Total |
231860.00 |
100 |
Source: Calculation
Result
Daily
rainfall intensity data used in calculations based on maximum rainfall data per
return period of 2, 5, 10, and 25 years can be seen in Table 11.
Table 11. Rainfall Intensity Data
|
R24 |
I |
PUH |
|
212.731 |
46.459 |
2 |
|
229.620 |
50.148 |
5 |
|
240.802 |
52.590 |
10 |
|
247.111 |
53.968 |
25 |
Source: Calculation
Result
Mapping
of rainfall runoff discharge values for each cut area of the research area for
return periods of 2, 5, 10, and 25 years.
Figure 6. Rainfall Runoff Discharge Distribution Map by Area
Source: Calculation
Result
As shown in Figure 6, the first map is a
prediction map of the distribution of runoff discharge for return periods of 2,
5, 10, and 25 years in the research area. The second figure is the distribution
of runoff discharge for return periods of 2, 5, 10, and 25 years in Area 1. In
the third figure is the distribution of runoff discharge for return periods of
2, 5, 10, and 25 years in Area 2. The fourth figure shows the distribution of
runoff discharge for return periods of 2, 5, 10, and 25 years in Area 3.
A recapitulation of
the annual return period plan total runoff discharge for each cut-off area of
the research area for the 2, 5, 10, and 25-year return period plans is provided
in Table 12.
Table 12. Recapitulation of Runoff Discharge Annual Return Period
Plan
|
Area |
Q Return Period (Maximum) m3/sec |
|||
|
Q2year |
Q5year |
Q10year |
Q25year |
|
|
A1 |
704.96 |
760.93 |
797.99 |
818.89 |
|
A2 |
1067.92 |
1152.71 |
1208.84 |
1240.52 |
|
A3 |
854.51 |
922.35 |
967.27 |
992.61 |
Source: Calculation
Result
Table 12 shows the recapitulation of the total runoff
discharge of the annual return period plan in the study area, based on the
calculation of the maximum discharge for return periods of 2, 5, 10, and 25
years in three different areas (A1, A2, and A3). The runoff discharge is
measured in m�/second (cubic meters per second) and shows how the runoff
discharge is expected to increase with longer return periods.
CONCLUSION
It can be concluded that this research shows that the
highest rainfall predicted in 2048 in Cirebon is 2471 mm based on BMKG Penggung
Station data, while in Majalengka it is expected to experience the highest
rainfall of 2922 mm, based on BMKG Kertajati Station data. In addition, the
largest runoff discharge is in Susukan sub-district, Cirebon, with a discharge
of 68.648 m�/second (Q25), while in Majalengka, the highest discharge is in
Lemahsugih sub-district at 104.16 m�/second (Q25). Detailed predictions for the
BMKG Kertajati station estimate rainfall of 2358 mm in 2025, 2194 mm in 2028,
and 2922 mm in 2048. Similarly, for the Jatiwangi and Penggung BMKG stations,
the predicted rainfall shows a consistent trend. Runoff discharge predictions
for different districts further highlight areas of concern, such as Kertajati
and Lemahsugih, where significant runoff may occur in the coming years. This
research provides valuable input to flood mitigation strategies for the Cirebon
and Majalengka regions. The findings suggest that efforts such as regulating
development over drainage systems, increasing green space, preserving river
areas, and building flood embankments are critical to reducing future flood
risks. Future research should expand on these results by including additional
climate variables such as wind speed, humidity and temperature to improve the
accuracy of rainfall predictions. In addition, updating the administrative maps
of the study area can improve the accuracy of digital mapping, thus ensuring
more effective disaster mitigation planning.
REFERENCES
Adiguno, S., Syahra, Y., & Yetri, M. (2022). Prediksi
Peningkatan Omset Penjualan Menggunakan Metode Regresi Linier Berganda. Jurnal
Sistem Informasi Triguna Dharma (JURSI TGD), 1(4), 275�281.
Afrian, R. (2020). Kajian Mitigasi
Terhadap Penyebab Bencana Banjir di Desa Sidodadi Kota Langsa. Jurnal Georafflesia: Artikel Ilmiah Pendidikan Geografi, 5(2), 165�169.
Aslim, M. A. F. I., Jasruddin, J., Palloan, P., Helmi, H.,
Arsyad, M., & Triwibowo, H. (2023). Monthly Rainfall Prediction Using the
Backpropagation Neural Network (BPNN) Algorithm in Maros Regency. Scientific
Journal of Informatics, 10(1), 13�24.
Cirebon, B. K. (2024). Kota Cirebon Dalam Angka 2024 (Vol. 49). Cirebonkota.Bps.Go.Id.
https://cirebonkota.bps.go.id/publication/2024/02/28/f33f29bb47e64e9eeafec39a/kota-cirebon-dalam-angka-2024.html
Cirebon, B. P. S. K. (2024). Kabupaten Cirebon Dalam Angka
2024 (Vol. 19). Cirebonkab.Bps.Go.Id.
https://cirebonkab.bps.go.id/publication/2024/02/28/e12c1d992be02c5b2492c646/kabupaten-cirebon-dalam-angka-2024.html
Ikhwan, M., Musa, R., & Mallombassi, A. (2022). Kajian
Debit Limpasan Permukaan Akibat Intensitas Curah Hujan Lapangan: Studi Kasus
DAS Kiru-Kiru Kab. Barru.
Jurnal Konstruksi: Teknik, Infrastruktur Dan Sains, 1(7), 49�56.
Kyojo, E. A., Mirau, S. S., Osima, S.
E., & Masanja, V. G. (2024). Frequentist and
Bayesian Approaches in Modeling and Prediction of Extreme Rainfall Series: A
Case Research from Southern Highlands Region of Tanzania. Advances in
Meteorology, 2024(1), 8533930.
Majalengka, B. K. (2024). Kabupaten Majalengka Dalam Angka
2024 (Vol. 39). Majalengkakab.Bps.Go.Id.
https://majalengkakab.bps.go.id/publication/2024/02/28/4e71313ad18650d50ed2c667/kabupaten-majalengka-dalam-angka-2024.html
Muttaqin, Z., & Farhan, O. (2021). Analysis of Floating
Net Cages on The Carrying Capacity of The Darma Reservoir. Jurnal Teknik
Sipil Dan Perencanaan, 23(1), 74�82.
Oktaviani, C., & Afdal, A. (2013). Prediksi curah hujan
bulanan menggunakan jaringan syaraf tiruan dengan beberapa fungsi pelatihan
backpropagation (studi kasus: stasiun meteorologi tabing padang, tahun
2001-2012). Jurnal
Fisika Unand, 2(4).
Qurrotaini, L., & Nuryanto, N.
(2020). Implementasi pendidikan mitigasi bencana alam gempa bumi dalam
pembelajaran IPS SD. Implementasi Pendidikan Mitigasi Bencana Alam Gempa
Bumi Dalam Pembelajaran IPS SD, 2(1), 37�44.
Ramdhani, R., & Yustiana, F. (2023). Tinjauan Perhitungan
Debit Limpasan Permukaan Dengan Metode Rasional Modifikasi (Studi Kasus
Drainase Mikro Kampus Itenas). Prosiding FTSP Series, 50�55.
Setiawan, W., & Barokah, A. (2022). Rainfall Prediction
Using Backpropagation with Parameter Tuning. MATEC Web of Conferences, 372,
7003.
Sheikhi, Y., Ashrafi, S. M., Nikoo, M. R., & Haghighi, A.
(2023). Enhancing daily rainfall prediction in urban areas: a comparative research
of hybrid artificial intelligence models with optimization algorithms. Applied Water Science, 13(12), 232.
Silva, M. A. da, Merlo, M. N., Thebaldi,
M. S., Ferreira, D. D., Schwerz, F., & Deus, F. P. de. (2023). Climate-related variables may not improve monthly
scale rainfall predictions by artificial neural networks for the metropolitan
region of Belo Horizonte, Brazil. Revista Ambiente & �gua, 18, e2879.
Siregar, E. (2019). Analisis Metode
Backpropagation Dengan Fungsi Aktivasi Sigmoid Bipolar Dan Fungsi Linear Dalam
Prediksi Pertumbuhan Penduduk. Universitas Sumatera Utara.
Sitio, A. S., Poningsih, P., Parlina,
I., Gunawan, I., & Sumarno, S. (2021). Analisis Metode Sig (Sistem Informasi Geografis) untuk Pemetaan Klinik
Bpjs Dengan Pendekatan Rad (Rapid Application Development). Jurnal Sosial Teknologi, 1(5), 355�365.
Tamaji, T., Utama, Y. A. K., &
Sidharta, J. (2022). Jaringan Saraf Tiruan Menggunakan Metode Backpropagation
untuk Prediksi Curah Hujan. Telekontran: Jurnal Ilmiah Telekomunikasi,
Kendali Dan Elektronika Terapan, 10(1), 30�37.
Thariq, A. (2020). Pemetaan Titik Kelandaian Permukaan Jalan
Dengan Memanfaatkan Teknologi Google Maps 3D. Jurnal Simetrik, 10(2),
328�332.
Tribhuwana, A., & Prasetyo, A. (2020). Analysis of
Infiltration Relations With Land Physical Properties. Journal of Green
Science and Technology, 4(3).
Tribhuwana, A., Rohman, F., & Farhan, O. (2021). Analysis
of The Carrying Capacity of Urban Drainage Dimensions. Journal of Green
Science and Technology, 5(1).
Triwahyuni, T., Husna, I., & Andesti, M. (2020). Hubungan
Curah Hujan dengan Kasus Demam Berdarah Dengue di Bandar Lampung 2016-2018. ARTERI:
Jurnal Ilmu Kesehatan, 1(3), 184�189.
Yusuf, M., Setyanto, A., & Aryasa, K. (2022). Analisis
Prediksi Curah Hujan Bulanan Wilayah Kota Sorong Menggunakan Metode Multiple
Regression. J-SAKTI (Jurnal Sains Komputer Dan Informatika), 6(1),
405�417.
Zahran, B. (2023). A fuzzy based model for rainfall
prediction. International Journal of Data and Network Science, 7(1),
97�106.
|
Copyright
holder: Farhan Fadhilah, Muhamad Reza, Awliya
Tribhuwana (2024) |
|
First
publication right: Journal of Transnational
Universal Studies (JTUS) |
|
This
article is licensed under:
|
