View Item 
      •   IPB Repository
      • Dissertations and Theses
      • Undergraduate Theses
      • UT - School of Data Science, Mathematic and Informatics
      • UT - Mathematics
      • View Item
      •   IPB Repository
      • Dissertations and Theses
      • Undergraduate Theses
      • UT - School of Data Science, Mathematic and Informatics
      • UT - Mathematics
      • View Item
      JavaScript is disabled for your browser. Some features of this site may not work without it.

      Prediksi Curah Hujan Harian Menggunakan Fuzzy Neural Network: Studi Kasus Kota Jakarta Utara

      Thumbnail
      View/Open
      Cover (1.821Mb)
      Fulltext (2.608Mb)
      Lampiran (2.326Mb)
      Date
      2025
      Author
      ALFARI, AHMAD EKA
      Supriyo, Prapto Tri
      Khatizah, Elis
      Metadata
      Show full item record
      Abstract
      Perubahan iklim global menyebabkan peningkatan ketidakpastian curah hujan harian, terutama di wilayah tropis seperti Jakarta Utara yang rawan banjir. Penelitian ini bertujuan membangun dan mengevaluasi model prediksi curah hujan harian menggunakan Fuzzy Neural Network (FNN) berbasis ANFIS. Model dikembangkan dalam lima lapisan utama yang mencakup input, fuzzifikasi, aturan fuzzy, inferensi, dan defuzzifikasi. Model dilatih menggunakan data tahun 2020–2023, diuji pada data tahun 2024, dan dievaluasi dengan tiga skenario, yaitu data latih, data uji dengan prediksi direct multi-step forecasting dan prediksi recursive forecasting. Hasil menunjukkan bahwa model FNN mampu mempelajari pola hujan nonlinier dengan baik pada data latih, serta mempertahankan performa baik pada data uji dengan prediksi direct multi-step forecasting. Performa kedua model menurun pada data uji dengan prediksi recursive forecasting, namun model FNN tetap unggul, yang menunjukkan bahwa model tetap mampu menangkap pola utama. Hasil ini menunjukkan bahwa model FNN berbasis ANFIS lebih unggul, efektif, dan interpretatif untuk prediksi curah hujan harian di wilayah tropis.
       
      Global climate change has led to increased uncertainty in daily rainfall, particularly in tropical regions such as North Jakarta, which is prone to flooding. This study aims to develop and evaluate a daily rainfall prediction model using a Fuzzy Neural Network (FNN) based on ANFIS. The model was developed in five main layers, including inputs, fuzzification, fuzzy rules, inference, and defuzzification. The model was trained using data from 2020–2023, tested on data from 2024, and evaluated with three scenarios: training data, test data with direct-multi step forecasting prediction, and recursive forecasting prediction. The results show that the FNN model is capable of learning nonlinear rainfall patterns well on training data and maintaining good performance on test data with direct-multi step forecasting prediction. The performance of both models decreases on test data with recursive forecasting prediction, indicating that the model is still capable of capturing the main trends. These results indicate that the ANFIS-based FNN model is superior, effective, and interpretable for daily rainfall prediction in tropical regions.
       
      URI
      http://repository.ipb.ac.id/handle/123456789/165360
      Collections
      • UT - Mathematics [89]

      Copyright © 2020 Library of IPB University
      All rights reserved
      Contact Us | Send Feedback
      Indonesia DSpace Group 
      IPB University Scientific Repository
      UIN Syarif Hidayatullah Institutional Repository
      Universitas Jember Digital Repository
        

       

      Browse

      All of IPB RepositoryCollectionsBy Issue DateAuthorsTitlesSubjectsThis CollectionBy Issue DateAuthorsTitlesSubjects

      My Account

      Login

      Application

      google store

      Copyright © 2020 Library of IPB University
      All rights reserved
      Contact Us | Send Feedback
      Indonesia DSpace Group 
      IPB University Scientific Repository
      UIN Syarif Hidayatullah Institutional Repository
      Universitas Jember Digital Repository