Ilahi, Rizqi Noer (2024) Implementasi Aplikasi Artificial Neural Network (ANN) Backpropagation untuk Prediksi Debit Harian pada Stasiun Pos Duga Air, DAS Serang-Lusi, Jawa Tengah. Diploma thesis, Institut Teknologi Sepuluh Nopember.
Text
2036201070-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2026. Download (29MB) | Request a copy |
Abstract
Hubungan antara hujan dan limpasan sangat erat. Sebagian hujan yang turun dari permukaan bumi terserap ke dalam tanah yang memungkinkan terjadi infiltrasi, dan sebagian lainnya mengalir ke saluran kecil hingga mencapai aliran sungai. Di tempat ini, limpasan terjadi ketika daratan tergenang oleh air hingga dapat menyebabkan banjir. Adanya sistem analisis yang dapat memprediksi dengan baik diperlukan untuk mengatasi masalah yang ada. Dalam praktiknya, memilih model untuk menganalisis dan menilai sistem DAS sangat sulit, namun, ini tidak berarti bahwa model yang ada tidak baik, salah satu model yang tersedia adalah penggunaan Jaringan Syaraf Tiruan (JST) atau Artificial Neural Network (ANN). Studi ini menyelidiki hasil perhitungan debit pemodelan Artificial Neural Network yang dilakukan menggunakan Matlab dan Python. Tujuan dari Tugas Akhir ini adalah untuk menghasilkan debit pemodelan ANN yang optimal dari model arsitektur jaringan yang digunakan. Untuk masukan program ANN, digunakan data curah hujan, evapotranspirasi, dan koefisien aliran. Data hujan dilakukan Uji konsistensi kurva massa ganda untuk menguji konsisteni data curah hujan. Selanjutnya, diambil hujan rerata kawasan menggunakan Polygon Thiessen. Sementara untuk luaran debit pemodelan dari program ANN ini adalah data debit yang diambil dari stasiun pos duga air. Hasil penelitian menunjukkan bahwa dalam pemodelan debit menggunakan ANN metode backpropagation memiliki kinerja yang sangat baik dengan nilai mean square error (MSE) terbaik adalah 0,032 untuk training, sedangkan untuk testing memiliki nilai MSE 0,047. Nilai keandalan atau akurasi program ANN ini mencapai 98%, yang artinya program ANN ini layak dijadikan metode pendekatan debit lapangan.
=================================================================================================================================
The relationship between rain and runoff is very close. Some of the rain that falls from the earth's surface is absorbed into the ground, which allows infiltration to occur, and some of it flows into small channels until it reaches the flow of rivers. In this place, runoff occurs when the land is flooded by water so that it can cause flooding. The existence of an analysis system that can predict well is necessary to overcome existing problems. In practice, choosing a model to analyze and assess a watershed system is very difficult, however, this does not mean that the existing model is not good, one of the available models is the use of Artificial Neural Network (JST) or Artificial Neural Network (ANN). This study investigated the results of the calculation of the Artificial Neural Network modeling discharge conducted using Matlab and Python. The purpose of this Final Project is to produce optimal ANN modeling discharge from the network architecture model used. For the input of the ANN program, rainfall, evapotranspiration, and flow coefficient data are used. Rainfall data was carried out Double mass curve consistency test to test the consistency of rainfall data. Next, the average rainfall of the area was taken using Thiessen's Polygon. Meanwhile, the modeling discharge output of the ANN program is discharge data taken from the suspected water post station. The results show that in the discharge modeling using ANN, the backpropagation method has excellent performance with the best mean square error (MSE) value of 0.032 for training, while for testing it has an MSE value of 0.047. The reliability or accuracy value of this ANN program reaches 98%, which means that this ANN program is worthy of being used as a field debit approach method.
Item Type: | Thesis (Diploma) |
---|---|
Uncontrolled Keywords: | DAS, Curah Hujan, Limpasan, Artificial Neural Network, Backpropagation, Mean Square Error, Watershed, Rainfall, Runoff |
Subjects: | T Technology > T Technology (General) > T174 Technological forecasting T Technology > T Technology (General) > T57.5 Data Processing T Technology > T Technology (General) > T57.6 Operations research--Mathematics. Goal programming T Technology > T Technology (General) > T57.62 Simulation T Technology > T Technology (General) > T57.8 Nonlinear programming. Support vector machine. Wavelets. Hidden Markov models. T Technology > TC Hydraulic engineering. Ocean engineering > TC530 Flood control T Technology > TC Hydraulic engineering. Ocean engineering > TC812 Irrigation |
Divisions: | Faculty of Vocational > Civil Infrastructure Engineering (D4) |
Depositing User: | Rizqi Noer Ilahi |
Date Deposited: | 22 Aug 2024 03:23 |
Last Modified: | 22 Aug 2024 03:23 |
URI: | http://repository.its.ac.id/id/eprint/113000 |
Actions (login required)
View Item |