PREDIKSI PASANG SURUT AIR SUNGAI MUSI DENGAN JARINGAN SARAF TIRUAN MENGGUNAKAN ALGORITMA BACKPROPAGATION DAN OPTIMASI CONJUGATE GRADIENT
Total View This Week0
Total View This Week0
Institusion
Universitas Sriwijaya
Author
FITRIYANTI HAKIKI (STUDENT ID : 09021381520066)
Efendi Rusdi (LECTURER ID : 8826630017)
Miraswan Kanda Januar (LECTURER ID : 0009019002)
Efendi Rusdi (LECTURER ID : 8826630017)
Miraswan Kanda Januar (LECTURER ID : 0009019002)
Subject
T1-995 Technology (General)
Datestamp
2019-10-15 08:17:59
Abstract :
Musi River is one of the largest rivers in Sumatera. The Musi River is the most important water transportation route in Palembang City. Therefore in the Musi River prediction of tides are needed in various sectors, especially sailing. Tides are a phenomenon of the rise up and down of the water surface due to the attraction of space objects, especially moon and sun. Tidal measurements are usually done manually, therefore this study developed software that can predict the tidal Musi River. Artificial neural network are one of the most commonly used method for prediction. The most commonly used algorithm is backpropagation. But this algorithm sometimes requires so much data, therefore optimization is needed. This research successfully developed the tide prediction software of Musi River using backpropagation algorithm and conjugate gradient optimization. Using input variables in the form of tidal data in the last twenty-four hours. Three tests were conducted on 184 tidal data and obtained MSE values of 1.286755, 0.964799 and 0.413152.
Musi River is one of the largest rivers in Sumatera. The Musi River is the most important water transportation route in Palembang City. Therefore in the Musi River prediction of tides are needed in various sectors, especially sailing. Tides are a phenomenon of the rise up and down of the water surface due to the attraction of space objects, especially moon and sun. Tidal measurements are usually done manually, therefore this study developed software that can predict the tidal Musi River. Artificial neural network are one of the most commonly used method for prediction. The most commonly used algorithm is backpropagation. But this algorithm sometimes requires so much data, therefore optimization is needed. This research successfully developed the tide prediction software of Musi River using backpropagation algorithm and conjugate gradient optimization. Using input variables in the form of tidal data in the last twenty-four hours. Three tests were conducted on 184 tidal data and obtained MSE values of 1.286755, 0.964799 and 0.413152.
File :
RAMA_55201_09021381520066.pdf
RAMA_55201_09021381520066_8826630017_0009019002_01_front_ref.pdf
RAMA_55201_09021381520066_8826630017_0009019002_02.pdf
RAMA_55201_09021381520066_8826630017_0009019002_03.pdf
RAMA_55201_09021381520066_8826630017_0009019002_04.pdf
RAMA_55201_09021381520066_8826630017_0009019002_05.pdf
RAMA_55201_09021381520066_8826630017_0009019002_06_ref.pdf
RAMA_55201_09021381520066_8826630017_0009019002_07_lamp.pdf
Download
book
BibTex
Latex, Jabref
cloud_download
Original Resource
url resource Institution
