Enhancing PLC-SCADA System for tracking petroleum terminals in industry using Relational Bilevel Aggregation Graph Convolutional Network (RBAGCN)

The petroleum industry is a vital economic sector, and facilities that process crude oil and natural gas for end use are known as petrochemical or oil terminals. This manuscript presents Programming Logic Controller (PLC) with Supervisory Control and Data Acquisition (SCADA) System for Tracking Petroleum Terminals in Industry using RBAGCN-PLC-SCADA. This study utilises data from petroleum terminal dataset and employs tracking technique as PLC-SCADA compared to traditional ladder diagrams, to produce simpler structure, fewer rungs, less processor memory, and faster execution times. Then, RBAGCN is used for prediction, estimating ladder diagram number, program size, and scan time with weights optimized by Giraffe Kicking Optimization (GKO). The proposed RBAGCN-PLC-SCADA method is implemented in MATLAB and evaluated using accuracy, precision, error, sensitivity, specificity, F1-score, and computational time, achieving 98% accuracy, 0.05% error, 98% F1-score, 97% precision, 99% sensitivity, and 97% specificity in petroleum terminal industry. The proposed method shows better results in all existing systems like monitoring industrial reactors using capacitance with long-short term memory (MIREC-LSTM), supervised anomaly detection in physical model with multilayer perceptron (SADP-MLP) and PLC system for rumaila degassing stations with recurrent neural network (PLC-RNN). The results show that proposed RBAGCN-PLC-SCADA method achieves lower error rate than existing methods.