Abstract—In this paper, an efficient neural network based estimation technique has been studied to estimate the magnetic field strength near any power substation, and to assess the possible exposure to electromagnetic radiation received by the residents living near that substation. The measurement and the estimation were carried out in close proximity to different high powered equipment at four different substations near Brunei Darussalam. Initially, the measurement was performed using the TM-191 gaussmeter for all four 66/11kV substations. In the measurement process the highest magnetic field of 12.5mG was recorded near the lightning arrestor at Telisai substation and the lowest value of 0.1mG was recorded at Lamunin substation for the same equipment. Later on, the magnetic field strengths were estimated using single-layer and two-layer feed-forward artificial neural networks (ANN). The highest value of coefficient of determination was found to be 98% using single-layer ANN estimation while the coefficient of determination was found to be around 99% by using two-layer ANN estimation. These coefficients of determination values indicate that the artificial neural network can predict the magnetic field strength with high accuracy.

 

Index Terms—substation, magnetic field, coefficient of determination, single-layer and two-layer ANN

Cite: Md. A. Salam, Swee P. Ang, Quazi M. Rahman, and Owasis A. Malik, "Estimation of Magnetic Field Strength near Substation Using Artificial Neural Network," International Journal of Electronics and Electrical Engineering, Vol. 4, No. 2, pp. 166-171, April 2016. doi: 10.18178/ijeee.4.2.166-171