Smart Monitoring System Berbasis Artificial Neural Network (ANN) pada Wastewater Treatment Pengendalian Nilai BOD Limbah Cair Greywater

Chafsah, Anis Mahmuda (2024) Smart Monitoring System Berbasis Artificial Neural Network (ANN) pada Wastewater Treatment Pengendalian Nilai BOD Limbah Cair Greywater. Diploma thesis, Institut Teknologi Sepuluh Nopember.

Text 2042201078-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2026. Download (13MB) | Request a copy

Abstract

Rumah tangga menyumbangkan sekitar 42,23% limbah cair, terutama greywater, ke lingkungan tanpa pengolahan lanjutan, berpotensi menciptakan masalah kesehatan dan lingkungan. Kualitas greywater harus memenuhi standar Peraturan Menteri Lingkungan Hidup No 68/Menlhk/Setjen/Kum.1/8/2016, termasuk BOD maksimum 30 mg/L, DO 4-9 mg/L, TDS 200-400 ppm, NH3 0-3 ppm, dan pH 6.0-9.0, mendukung SDGs point 6 akses air bersih dan sanitasi. Kandungan BOD merupakan faktor terpenting dalam mengukur kualitas limbah cair, tetapi terdapat kendala dalam pengukuran BOD yang disebabkan mahalnya biaya dan keterbatasan aplikasi sensor sehingga tidak bisa dilihat secara real time. Sehingga diperlukan adanya proses pendekatan untuk mendapatkan nilai BOD salah satu caranya menggunakan model Artificial Neural Network. Berdasarkan permasalahan tersebut, diperlukan adanya inovasi wastewater treatment. Pertama, menggunakan sistem monitoring dengan sensor pH, MQ-135, TDS, dan DO pada limbah cair greywater. Kedua, proses filter filtration process dengan menggunakan zeolite sand, actived carbon, actived sand, manganese green sand, dan silica sand. Ketiga, metode prediktif BOD akan dibangun menggunakan model Artificial Neural Network (ANN). Salah satu jenis yang digunakan adalah algoritma backpropagation neural network. Arsitektur yang dihasilkan berjumlah 4 input layer, 6 hidden layer, dan 1 output layer. Hasil pengujian wastewater treatment menghasilkan nilai performa sensor pada pH, TDS, DO, NH3 di tangki 1 dan 2 mengalami penurunan selama proses, dengan error: 1,38% dan 1,39%, TDS 1,64% dan 2,70%, DO 2,86% dan 1,90%, MQ-135 2,69% dan 0,16%. Nilai prediksi BOD sebelum proses di tangki 1 dan 2: 58,8423 dan 58,1438 kemudian setelah proses: 28,21645 dan 27,47017. RMSE tangki 1 dan 2: 0,78 mg/L O2 dan 0,82 mg/L O2.
==============================================================================================================================
Households contribute about 42.23% of wastewater, especially greywater, to the environment without advanced treatment, potentially creating health and environmental problems. The quality of greywater must meet the standards of the Minister of Environment Regulation No. 68/Menlhk/Setjen/Kum.1/8/2016, including a maximum BOD of 30 mg/L, DO 4-9 mg/L, TDS 200-400 ppm, NH3 0-3 ppm, and pH 6.0-9.0, supporting SDGs point 6 access to clean water and sanitation. BOD content is the most important factor in measuring the quality of liquid waste, but there are obstacles in measuring BOD due to the high cost and limited sensor applications so that it cannot be seen in real time. So that an approach process is needed to get the BOD value, one way is to use the Artificial Neural Network model. Based on these problems, wastewater treatment innovation is needed. First, using a monitoring system with pH, MQ-135, TDS, and DO sensors in graywater wastewater. Second, a filter filtration process using zeolite sand, activated carbon, activated sand, manganese green sand, and silica sand. Third, the BOD predictive method will be built using an Artificial Neural Network (ANN) model. One of the types used is the backpropagation neural network algorithm. The resulting architecture consists of 4 input layers, 6 hidden layers, and 1 output layer. The results of wastewater treatment testing resulted in sensor performance values on pH, TDS, DO, NH3 in tanks 1 and 2 decreasing during the process, with errors: 1.38% and 1.39%, TDS 1.64% and 2.70%, DO 2.86% and 1.90%, MQ-135 2.69% and 0.16%. Predicted values of BOD before the process in tanks 1 and 2: 58.8423 and 58.1438 then after the process: 28.21645 and 27.47017. RMSE of tanks 1 and 2: 0.78 mg/L O2 and 0.82 mg/L O2.

Item Type:	Thesis (Diploma)
Uncontrolled Keywords:	Artificial Neural Network, Biological Oxygen Demand, Limbah Cair Greywater, Filtration process, SDGs, Greywater Liquid Waste
Subjects:	T Technology > TD Environmental technology. Sanitary engineering > TD433 Water treatment plants
Divisions:	Faculty of Vocational > Instrumentation Engineering
Depositing User:	Anis Mahmuda Chafsah
Date Deposited:	29 Aug 2024 03:15
Last Modified:	29 Aug 2024 03:15
URI:	http://repository.its.ac.id/id/eprint/109945

Actions (login required)

View Item