Fault Detection System in Gas Metering Station Using Neural Network
N. S. Rosli, R. Ibrahim, and W. A. Syahira
Department of Electrical & Electronics Engineering, Universiti Teknologi PETRONAS, 31750 Bandar Seri Iskandar, Perak, Malaysia
Abstract—This research study focuses on the discussion regarding the development of fault detection in gas metering station using an Artificial Neural Network (ANN). The proposed model of fault detection applies ANN approach in order to provide a good detection method for billing purpose. However, one of the main problems faced by gas metering system is the undiagnosed faulty condition of measurement. Moreover, there are many researches regarding the Fault Detection and Diagnosis (FDD) that were conducted to enhance the reliability of the system in the plant process. Therefore, in order to address this issue, fault detection system using neural network is proposed to detect the fault data in the measured readings. The investigation of all faulty instruments was obtained from the detection model which was selected based on the performance of different ANN algorithms. Since, the artificial intelligence, such as neural network is one of the powerful tools in detecting and diagnosing the fault occurred. The ability of the neural networks to learn from the experience or past data has shown a great impact in the fault detection efficiency. Furthermore, such method based on the past data has also been established to improve the accuracy of the fault detection.
Index Terms—first neural network, fault detection, gas metering, fault diagnosis
Cite: N. S. Rosli, R. Ibrahim, and W. A. Syahira, "Fault Detection System in Gas Metering Station Using Neural Network," International Journal of Electronics and Electrical Engineering, Vol. 4, No. 1, pp. 52-55, February 2016. doi: 10.18178/ijeee.4.1.52-55
Cite: N. S. Rosli, R. Ibrahim, and W. A. Syahira, "Fault Detection System in Gas Metering Station Using Neural Network," International Journal of Electronics and Electrical Engineering, Vol. 4, No. 1, pp. 52-55, February 2016. doi: 10.18178/ijeee.4.1.52-55
Array