Article Per Year (5 Year)
p-Index From 2021 - 2026
0.408
P-Index
This Author published in this journals
All Journal Buletin GAW Bariri ( BGB)
Helminah Herawati
Pusat Meteorologi Penerbangan, Badan Meteorologi Klimatologi dan Geofisika, Jakarta Pusat, DKI Jakarta, 10610
Agriculture, Biological Sciences & Forestry Astronomy Computer Science & IT Earth & Planetary Sciences Environmental Science

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
Prediksi Harian Suhu Udara Permukaan dengan Jaringan Syaraf Tiruan : Studi Kasus di Kawasan perkotaan dan Pesisir Jakarta, Indonesia Richard Mahendra Putra; Eka Fibriantika; Yetti Kusumayanti; Erlya Afrianti; Arifatul Hidayanti; Wishnu Agum Swastiko; Helminah Herawati; Atri Wiujiana
Buletin GAW Bariri (BGB) Vol 4 No 1 (2023): BULETIN GAW BARIRI
Publisher : Stasiun Pemantau Atmosfer Global Lore Lindu Bariri - Palu

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31172/bgb.v4i1.84

Abstract

The weather forecast is significant to protect life and property. A forecast of temperature is important to the agriculture sector because when high temperatures can cause the soil to dry out faster and reduce the availability of water for plants. Furthermore, understanding the temperature condition can help meteorologists forecast the other atmosphere condition. The purpose of this research is to make a modeling prediction of temperature conditions the next day using an artificial neural network model (ANN). To make the ANN model, the daily average temperature measured in the meteorological stations in the urban and the coastal areas of Jakarta over 2010 – 2019 was used as training data. The testing data using surface temperature during January – December 2020. This model uses the various number of neurons in the hidden lapisan between 3 and 15. Based on the result, the ANN model is good enough to predict the temperature condition in Jakarta with the correlation between 0.625 – 0.653 and mean absolute error (MAE) between 0.569 – 0.600 oC. The best model prediction was obtained when the neuron number was 4 in the Urban Area of Jakarta and 8 in the Coastal Area of Jakarta seen from the high correlation value with the observations and a low error rate.
Co-Authors Arifatul Hidayanti Atri Wiujiana Eka Fibriantika Erlya Afrianti Richard Mahendra Putra Wishnu Agum Swastiko Yetti Kusumayanti