Neural network model with particle swarm optimization for prediction in gas metering systems

• Main Authors: Rosli, N.S., Ibrahim, R., Ismail, I.
• Format: Article
• Published: Institute of Electrical and Electronics Engineers Inc. 2017
• Online Access: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85012026140&doi=10.1109%2fICIAS.2016.7824049&partnerID=40&md5=9ca1f71bffbe4b7165d056bc5136e5bc; http://eprints.utp.edu.my/20173/

This research focuses on developing an intelligent system of prediction model to justify instrument's reliability. It is important to have an accurate prediction model in order to provide the reliable gas metering system. As the result, the billing integrity between the distributor and the customers are not affected. The application of particle swarm optimization (PSO) in optimizing the weights and biases of neural network (ANN) model is proposed to enhance the accuracy and performance of prediction model for gas metering system. This paper provides on the analysis on comparing the parameter prediction using ANN only with PSO-based ANN techniques. The results discover that the proposed instrument has the higher accuracy in estimating gas measurement with the errors lower than 1. © 2016 IEEE.