IMPLEMENTASI MULTIPLE ADAPTIVE NEURO-FUZZY INFERENCE SYSTEM (MANFIS) PADA SISTEM PENDETEKSI KEBOCORAN PIPA DISTRIBUSI WATER TREATMENT PLANT

Dzaky, Nuriqbal (2026) IMPLEMENTASI MULTIPLE ADAPTIVE NEURO-FUZZY INFERENCE SYSTEM (MANFIS) PADA SISTEM PENDETEKSI KEBOCORAN PIPA DISTRIBUSI WATER TREATMENT PLANT. Bachelor thesis, Institut Teknologi Kalimantan.

[img] Text
04221076_Cover.pdf
Restricted to Repository staff only until 4 October 2028.
Download (291kB) | Request a copy
[img] Text
04221076_statement_of_authenticity.pdf
Restricted to Repository staff only until 4 October 2028.
Download (607kB) | Request a copy
[img] Text
04221076_publishing_agreement.pdf
Restricted to Repository staff only until 4 October 2028.
Download (575kB) | Request a copy
[img] Text
04221076_approval_sheet.pdf
Restricted to Repository staff only until 4 October 2028.
Download (430kB) | Request a copy
[img] Text
04221076_preface.pdf
Restricted to Repository staff only until 4 October 2028.
Download (198kB) | Request a copy
[img] Text
04221076_abstract_id.pdf
Restricted to Repository staff only until 4 October 2028.
Download (390kB) | Request a copy
[img] Text
04221076_abstract_en.pdf
Restricted to Repository staff only until 4 October 2028.
Download (323kB) | Request a copy
[img] Text
04221076_table_of_content.pdf
Restricted to Repository staff only until 4 October 2028.
Download (326kB) | Request a copy
[img] Text
04221076_illustrations.pdf
Restricted to Repository staff only until 4 October 2028.
Download (297kB) | Request a copy
[img] Text
04221076_tables.pdf
Restricted to Repository staff only until 4 October 2028.
Download (299kB) | Request a copy
[img] Text
04221076_notations.pdf
Restricted to Repository staff only until 4 October 2028.
Download (330kB) | Request a copy
[img] Text
04221076_chapter_1.pdf
Restricted to Repository staff only until 4 October 2028.
Download (590kB) | Request a copy
[img] Text
04221076_chapter_2.pdf
Restricted to Repository staff only until 4 October 2028.
Download (630kB) | Request a copy
[img] Text
04221076_chapter_3.pdf
Restricted to Repository staff only until 4 October 2028.
Download (1MB) | Request a copy
[img] Text
04221076_chapter_4.pdf
Restricted to Repository staff only until 4 October 2028.
Download (1MB) | Request a copy
[img] Text
04221076_approval_sheet.pdf
Restricted to Repository staff only until 4 October 2028.
Download (430kB) | Request a copy
[img] Text
04221076_conclusions.pdf
Restricted to Repository staff only until 4 October 2028.
Download (351kB) | Request a copy
[img] Text
04221076_bibliography.pdf
Restricted to Repository staff only until 4 October 2028.
Download (362kB) | Request a copy
[img] Text
04221076_enclosure.pdf
Restricted to Repository staff only until 4 October 2028.
Download (1MB) | Request a copy
[img] Text
04221076_paper.pdf
Restricted to Repository staff only until 4 October 2028.
Download (425kB) | Request a copy
[img] Text
04221076_presentation.pdf
Restricted to Repository staff only until 4 October 2028.
Download (7MB) | Request a copy
[img] Text
04221076_Form.TA-020.pdf
Restricted to Repository staff only until 4 October 2028.
Download (438kB) | Request a copy
[img] Text
04221076_approval_sheet.pdf
Restricted to Repository staff only until 4 October 2028.
Download (502kB) | Request a copy

Abstract

Kebocoran pada jaringan distribusi air Water Treatment Plant (WTP) merupakan salah satu penyebab utama terjadinya Non-Revenue Water (NRW) yang berdampak pada kerugian operasional dan menurunkan efisiensi distribusi air. Deteksi kebocoran secara konvensional masih memiliki keterbatasan, terutama pada jaringan pipa bercabang dan pipa bawah tanah yang sulit diamati secara langsung. Penelitian ini bertujuan merancang dan mengimplementasikan sistem pendeteksi kebocoran pipa distribusi air berbasis Multiple Adaptive Neuro-Fuzzy Inference System (MANFIS) pada prototipe Water Treatment Plant (WTP). Sistem dibangun menggunakan lima sensor Flow dan tiga sensor Pressure yang terhubung dengan Arduino Nano sebagai perangkat akuisisi data. Data sensor diolah melalui proses feature engineering untuk menghasilkan parameter ΔFLOW dan ΔPressure yang digunakan sebagai masukan pada tiga sub-ANFIS yang bekerja secara independen dan paralel. Setiap sub-ANFIS bertugas menganalisis satu segmen pipa dan menghasilkan keluaran berupa kondisi segmen dalam kategori normal, bocor kecil, atau bocor besar. Arsitektur paralel tersebut memungkinkan identifikasi kebocoran dilakukan secara spesifik pada masing-masing segmen tanpa dipengaruhi kondisi segmen lainnya. Dataset diperoleh dari 32 kombinasi kondisi operasi yang terdiri atas kondisi normal, bocor kecil, bocor besar, dan variasi pemakaian air. Proses pelatihan model menggunakan metode adaptive learning berbasis gradient descent dengan Gaussian Membership Function dan inferensi Sugeno orde nol. Hasil validasi model menunjukkan akurasi sebesar 85,94% pada Segmen 1, 92,71% pada Segmen 2, dan 90,62% pada Segmen 3. Pengujian realtime menggunakan 32 skenario operasi menghasilkan 275 prediksi benar dari total 288 prediksi sehingga diperoleh tingkat akurasi sistem sebesar 95,49%. Hasil penelitian menunjukkan bahwa metode MANFIS mampu membedakan kondisi normal, bocor kecil, dan bocor besar secara efektif berdasarkan perubahan debit dan tekanan pada jaringan distribusi air. Sistem yang dikembangkan berpotensi diterapkan sebagai solusi pendeteksian kebocoran dini untuk membantu mengurangi kehilangan air pada sistem distribusi Water Treatment Plant (WTP).

Item Type: Thesis (Bachelor)
Subjects: T Technology > TK Electrical engineering. Electronics Nuclear engineering
Divisions: Jurusan Teknologi Industri dan Proses > Teknik Elektro
Depositing User: Nuriqbal Dzaky
Date Deposited: 09 Jul 2026 06:27
Last Modified: 09 Jul 2026 06:27
URI: http://repository.itk.ac.id/id/eprint/26432

Actions (login required)

View Item View Item