VALIDATION OF NOISE LEVEL IN OMU-ARAN TOWNSHIP USING ARTIFICIAL NEURAL NETWORK

OLAJIDE,, OPEYEMI SUNDAY (2022) VALIDATION OF NOISE LEVEL IN OMU-ARAN TOWNSHIP USING ARTIFICIAL NEURAL NETWORK. Masters thesis, Landmark University, Omu Aran, Kwara State.

Text OLAJIDE OPEYEMI SUNDAY.docx Download (912kB)

Abstract

The inconspicuous nature of noise pollution has made it undetectable in both developed and developing countries with associated health complications. In Nigeria, several studies have suggested different models to mitigate noise pollution, but there is none carried out on noise levels in Omu-Aran, therefore this study, was aimed at validating noise levels in Omu-Aran Nigeria using Artificial Neural Network. A sound level meter type SL4010 was used to measure, the noise levels in the morning, afternoon, and evening daily for the duration of three weeks from three zones. The zones are subdivided into seven locations namely: Oke-Agbede, Landmark junction, High Court junction, Latinwo Market, Ile-Nla, Falaye, Landmark Chapel, Central Market, Central Roundabout, Iganngu/Okeki, Ile-Olupo/Ile-Adee, Odo-Areyin, Egbe Garage, Otolorin/Federal Hospital Junction, GRA, Agamo, Taissa Junction, Bovas, Orolodo/Olomu Palace, Secretariat/Eco Bank, and Taiwo. Health risk assessment was evaluated using the recommended exposure limit (REL) and permissible exposure limit (PEL) established by the National Institute of Occupational Safety and Health while Artificial Neural Network was used to validate the noise levels. Descriptive statistics were used to manage the data at P< O.05 level of statistical significance The average mean noise levels for all locations were 67.82± 2.1, 68.7± 1.87, and 69.53± 2.24 dB for the mornings, afternoons and evenings respectively. There was a significant difference in noise levels across the period of the day where Central Roundabout, Central Market, and Landmark University Chapel were exposed to non-permissible noise levels of 87.24, 86.78, and 83.16 dB respectively. The noise level exposure at each location has health issues ranging from discomfort to cardiovascular effects, and an increase in physiological responses with a noise level of 60dB to 87.24 dB. The ANN model validated the noise levels in Omu-Aran township as one of the best results with 21 associated input features which showed high performance of 97.84% accuracy and the root means square error to be RMSE =0.1096. Population and human activities have an impact on the noise levels at the different locations in Omu-Aran with its health-related issues. Hence, there is a need for law enforcement at the local level to mitigate the health-related issues Key Words: Noise Levels, Health Risk Assessment, Artificial Neural Networks, Omu-Aran

Actions (login required)

View Item