Garuda - Garba Rujukan Digital

Pemodelan Curah Hujan-Limpasan Menggunakan Artificial Neural Network (ANN) dengan Metode Backpropagation Hadihardaja, Iwan K.; Sutikno, Sugeng
Jurnal Teknik Sipil Vol 12, No 4 (2005)
Publisher : Institut Teknologi Bandung

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (720.384 KB)

Abstrak. Hubungan curah hujan-limpasan telah dikembangkan secara terus-menerus dengan menerapkan artificial intelligence sebagai alternatif pemodelan yang pada penelitian ini menggunakan black box model yakni Jaringan Syaraf tiruan (artificial neural network). Dengan menggunakan model kotak hitam (black box model) tersebut, maka dalam penerapanya tidak membutuhkan pengetahuan yang komplek antar elemen-elemen dalam suatu sistem Daerah Aliran Sungai (DAS) yang juga tidak secara eksplisit merepresentasikan hubungan antar elemen dalam DAS dan proses interaksi curah hujan-limpasan. Sehingga perubahan antar elemen di dalam suatu DAS tidak perlu diketahui sepanjang hujan dan limpasan diukur dengan akurat dari waktu ke waktu, oleh karena itu pemodelan dapat dilakukan secara lebih sederhana dengan hanya memiliki input hujan dan output limpasan didalam pengembangan model tersebut. Studi kasus yang digunakan adalah data curah hujan tengah bulanan dan debit limpasan DAS Way Sekampung â Pujorahayu, selama kurun waktu 19 tahun. Data curah hujan merupakan input sedangkan data debit merupakan variable output. Dari hasil penelitian ini diperoleh kofisien korelasi tertinggi sebesar 0,813 atau 81,3%. Dalam studi ini disimpulkan bahwa secara umum ANN dapat diterapkan dalam pemodelan curah hujan-limpasan, walaupun hasilnya belum terlalu akurat oleh karenamasih ada penyimpangan.Abstract. Rainfall-runoff relation has been developed continuously by applying artificial intelligence as one of the black box model alternative called Artificial Neural Network. By applying black box model, it is not necessary to apply complexity of knowledge due to interrelated elements in a river basin in which it is not explicitly representing the relation of the elements and interaction process of the rainfall-runoff modeling. Consequently, the changes of the elements in a river basin is not necessary to be quantified as long as rainfall dan runoff is observed accurately from time to time, furthermore, the modeling can be applied within less complexity due to rainfall and runoff data observation as an input and output, respectively. The case study applied to the river flow on the way Sekampung River in Lampung Province. The data used is rainfall data and stream flow discharge data in the middle of the month on the water level station Pujorahayu, for 19 years from 1983 up to 2001. The rainfall data is input and stream flow is a variable output. Learning method that is used reduced gradient. From the result of this research got correlation coefficient 0,813 or 81,3% the tallest. The conclusion of this research is the generally ANN can implementated in the rainfall run off modeling, although the result is not extremely accurate yet because of the deviation between observation and result of the model.

Perbandingan Hidrograf Satuan Teoritis Terhadap Hidrograf Satuan Observasi DAS Ciliwung Hulu Agus, Indra; Hadihardaja, Iwan K.
Jurnal Teknik Sipil Vol 18, No 1 (2011)
Publisher : Institut Teknologi Bandung

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (763.331 KB)

Abstrak. Banjir dan kekeringan dalam suatu wilayah (DAS) terjadi akibat fenomena iklim yaitu distribuasi curah hujan dengan intensitas tinggi atau periode kemarau yang terjadi lebih panjang. Penyebab banjir berasal dari masukan (hujan) dan sistem DAS. Masukan (hujan) meliputi faktor intensitas hujan, lama hujan dan distribusi hujan. Sedangkan sistem DAS meliputi faktor topografi, jenis tanah, penggunaan lahan dan sistem transfer hujan dalam DAS. Perkiraan debit banjir yang berdasarkan hujan lebat dapat diklasifikasikan dalam tiga cara yaitu, dengan cara rumus empiris, dengan cara statistik (kemungkinan) dan dengan cara hidrograf satuan. Curah hujan dan debit adalah dua hal penting untuk mendapatkan himpunan hidrograf hasil observasi dan teoritis. Curah hujan merupakan nilai yang efektif dan dihitung menggunakan metode indeks. Aliran langsung (limpasan langsung) didapat dari segregasi total limpasan dengan baseflow dan dengan menerapkan metode Straight Line. Ordinat hidrograf hasil observasi dihasilkan dengan membagi ordinat aliran langsung (direct overflow) dengan hujan efektif. Hasil dari satuan hidrograf observasi dibandingkan dengan satuan hidrograf teoritis dan dihitung dengan menggunakan metode Least Square, Forward Subtitution dan Linear Reservoar Cascade. Pebandingan dilakukan terhadap Time base (Tb), Time peaks (Ts), Q peak. Abstract. Flood and dryness in a region watershed was caused by a climate phenomenon, that is high intensity rainfall distribution or longer drought period. Flood is the existence of rainfall input and watershed system. Rainfall covers rain factor of stress intensity, rainfall duration and rain distribution. System watershed covers topography factor, soil type,land use and rain system transfer in watershed. Flooding debit approximation based on torrential rains can be classified in three ways such empiric formula, statistic or probability and unit hydrographer. Rainfall and discharge are important things in getting observed and theoretical hydrographer set. Applying Rainfall is the effective precipitation calculated using index method. Whereas direct overflow (directÂ run off) was earned by total run off segregation with base flow applies Straight Line Method. Observed Hydrographer ordinates is created by dividing direct overflow ordinate with effective rain. Observed unit hydrograph result was compared with teoritic unit hydrograph and calculated by using Least Square Method, Forward Subtitution Method and Linear Reservoar Cascade Method. Comparation was done to Time base (Tb), Time peaks (Ts), Q peak.

Determining The Critical Degree of Reservoir Lifetime for The Saguling Reservoir Based on The Sediment Inflow Simulation Bakhtiar, Bakhtiar; Hadihardaja, Joetata; Hadihardaja, Iwan K.
Jurnal Teknik Sipil Vol 17, No 1 (2010)
Publisher : Institut Teknologi Bandung

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract. This study aims to find out the relationship between the critical degree of reservoir lifetime and sediment inflow. The critical degree of reservoir lifetime is defined as the ratio between the designated lifetime of the reservoir and its effective lifetime. The effective lifetime of a reservoir is affected by the amount of sediment flowing into the reservoir and trap efficiency. Soil and Water Assessment Tool (SWAT) version 2005 model was incorporated into this study to simulate hydrological processes taking place in the catchment. The reliability of of SWATmodel depends greatly on the data availability and some parameter adjustments. From the calibration and validation results on monthly data, the model is considerably of good performance. TheÂ modeling resultsrevealed that the mean sediment yield is 235.86 ton/ha/year, and the mean sediment inflow into the reservoir is 5,102,000 tons per year which is equivalent to 3,836,090 m3 per year resulting in 43.66 years of the reservoir lifetime (21.66 years remaining) which corresponds to the critical degree of 2.29 The graph plotting the values of the sediment inflow and the critical degree of reservoir lifetime reveals a linear relationship. Abstrak. Penelitian ini dimaksudkan untuk mencari hubungan antara tingkat kekritisan umur waduk dengan inflow sedimen. Tingkat kekritisan umur waduk didefinisikan sebagai perbandingan antara umur rencana waduk dengan umur efektifnya. Umur efektif waduk dipengaruhi oleh jumlah sedimen yang masuk ke waduk serta efisiensi pengendapan sedimen. Model Soil and Water Assessment Tool (SWAT) versi 2005 digunakan dalam penelitian ini untuk mensimulasikan proses-proses hidrologi yang terjadi di daerah tangkapan. Keandalan model SWAT sangat bergantung pada ketersediaan data dan penyesuaian beberapa parameter. Dari hasil kalibrasi dan validasi terhadap data bulanan, kinerja model dinilai cukup baik. Hasil simulasi menunjukkan bahwa sediment yield rata-rata sebesar 235,86 ton/ha/tahun, serta inflow sedimen rata-rata yang masuk ke waduk sebesar 5.102.000 ton per tahun yang ekivalen dengan 3.836.090 m3 per tahun yang menghasilkan umur waduk sebesar 43,66 tahun (21,66 sisa umur waduk) yang bersesuaian dengan tingkat kekritisan 2,29. Grafik yang menghubungan nilai inflow sedimen dengan tingkat kekritisan umur waduk menunjukkan hubungan yang bersifat linear.

Pemodelan Prioritas Pemanfaatan Potensi Air Situ Berbasis Konservasi Sumberdaya Air Hadihardaja, Iwan K.
MEDIA KOMUNIKASI TEKNIK SIPIL Volume 12, Nomor 2, Edisi XXIX, JULI 2004
Publisher : Department of Civil Engineering, Diponegoro University

Degradation function of Situ Gede (Lake Gede) becomes crucial at present time compared to its function at initial condition when it was used forÂ tourism, irrigation, and raw water supply within an optimal use. The crucial aspects are related to the decreasing reliability of supply to Situ Gede and increasing sedimentation accumulation. The research concerns with various aspects due to the Situ Gede conservation and restoration that is aimed to increase the potential uses, based on the quantitative and qualitative model by using optimization technique and Analitic Hierarchy Process (AHP). The optimization technique is applied in order to evaluate the amount of monthlyÂ water availability within 3 scenarios i.e. tourism ounly (Scenario 1), tourism and irrigation (Scenario 2), and, tourism, irrigation and raw water (Scenario 3), respectively, based on the potency of Situ Gede itself, new additional supply from other sources, and additional supply from Galunggung Crater. In addition, based on simulation of accumulated sedimentation model, that is aimed to identify the effect of firm outflow change of Situ Gede Release, indicates the decrease of firm outflow around 0,5% annually. The priority order based on AHP evaluation shows that the the first priority is Scenario 1, the second is Scenario 3 and the third is Scenario 2. Those priorities based on various criterias such as Investment Cost and Operation Maintenance, Additional Supply Reliability to Situ Gede, and Sedimentation. Sensitiviy analysis is also applied based on the weight change of each criteria in performing the shift possibilities due to the priority order. Key Words : Non Linear Programming Modeling,Â Simulation, Multi objective-Multi Criteria

PENELITIAN KERUSAKAN UNIT LAPIS LINDUNG PEMECAH GELOMBANG TERHADAP BERBAGAI SPEKTRUM GELOMBANG Sriyana, Sriyana; Hadihardaja, Iwan K.
MEDIA KOMUNIKASI TEKNIK SIPIL Tahun 15, Nomor 1, PEBRUARI 2007
Publisher : Department of Civil Engineering, Diponegoro University

Determination of the amount of damage in the armour layer unit of breakwater is highly affected by wave spectrum generated. This research is aimed at knowing the influence of many kind of wave spectrum toward the amount of damage in the armour layer unit. This study is carried out by means of irregular non-breaker waves on physical modeling. The spectrums used on irregular waves are Pierson Moskowitz, Bretschneider, Jonswap, and Darbyshire, employing tetrapod, cubes, and rubble mound as the armour layer. The result of the test model indicate that Darbyshire spectrum have largest potential to destroy the stability of armour layer more than spectrum Jonswap, Pierson Moskowitz, Bretschneider.

Aplikasi Metode Generalized Reduced Gradient dalam Pemodelan Curah Hujan-Limpasan Menggunakan Artificial Neural Network (ANN) Hadihardaja, Iwan K.; Sutikno, Sugeng
MEDIA KOMUNIKASI TEKNIK SIPIL Volume 13, Nomor 2, Edisi XXXII, JUNI 2005
Publisher : Department of Civil Engineering, Diponegoro University

The rainfall run off modelling is necessary until now days, for fulling data or make data longer. Artificial neural network can made the alternative rainfall run off modelling. The implementation Artificial neural networ for modelling on the water resources which is done by researcher to get an accurate result. Artificial neural networ is one of artificial intelligent that is imitation of representation from brain of human. This model is the black box modelling, so in the implementation were not need complecity of scient among the other aspect in the process of rainfall run off modelling. The case study applied to the river flow on the way Sekampung river in Lampung Province. The data used is rainfall data and stream flow data in the middle of the month on the water level station Pujorahayu, for 19 years from 1983 up to 2001. The rainfall data is input and stream flow is a variable output. Learning method that is used reduced gradient. From the result of this research got correlation coefficient 0,790 or 79 % the tallest. The conclution of this research is the generaly ANN can implementated in the rainfall run off modeling, although the result is not too accurate because of there is still deviation.Keywords: rainfall-runoff, artificial neural network, black box, generalized reduced gradientPermalink: http://ejournal.undip.ac.id/index.php/mkts/article/view/3907[How to cite: Hadihardaja, I.K. dan Sutikno, S., 2005, Aplikasi Metode Generalized Reduced Gradient dalam Pemodelan Curah Hujan-Limpasan Menggunakan Artificial Neural Network (ANN), Jurnal Media Komunikasi Teknik Sipil, Volume 13, Nomor 2, pp. 37-49]

Pengaruh Curah Hujan Rata-rata Tahunan terhadap Indeks Erosi dan Umur Waduk pada DAS Citarum Hulu Bakhtiar, Bakhtiar; Hadihardaja, Joetata; Hadihardaja, Iwan K.
MEDIA KOMUNIKASI TEKNIK SIPIL Volume 19, Nomor 1, JULI 2013
Publisher : Department of Civil Engineering, Diponegoro University

The aim of this study is to find out the effect of annual rainfall on the erosion index and the life of a reservoir. The erosion index is defined as the total soil loss divided by the tolerable soil loss. The life of a reservoir is affected by the amount of sediment flowing into the reservoir and sediment trap, which is a function of the effective capacity of the reservoir and the inflowing water. Soil and Water Assessment Tool (SWAT) version 2005 model was incorporated into this study to simulate hydrological processes taking place in the catchment. ArcSWAT (ArcGIS Interface for SWAT 2005) program was used as a preprocessing tool to write in input files to be executed by SWAT.SWAT model involves a great number of parameters. Hence, its reliability depends so much on the data availability and some parameter adjustments. From the calibration and validation results on annual data, the model is considerably of good performance. This was proven from the coefficient of correlation (r), coefficient of determination (R2), model efficiency (ME), and index of agreement (IA) which are close to 1 except for the sediment inflow. The difference between the observed and simulated sediment inflow resulted because the SWAT model accounts for the annual variation in precipitation and mean inflow discharge which is not manifested in the observed data. The modeling results revealed that the average inflowing sediment into the reservoir is 5,102,000 ton/year resulting in 46.18 year effective life of the reservoir which corresponds to the critical degree of 1.15. The graph plotting the values of the sediment inflow and the critical degree of the reservoir life reveals a linear relationship.

Title

Found 7 Documents
Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title Search

Found 7 Documents Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title

Found 7 Documents
Search