Discrete event based hybrid framework for petroleum products pipeline activities classification

Abstract

The importance of timely detection, classification and response to anomalies on petroleum products pipeline (PPP) have attracted pragmatic researches in recent times. There is need for efficient monitoring and detection of activities on PPP to guide leak detections and remedy decisions. This paper develops an intelligent hybrid system, driven by discrete event system specification (DEVS) and adaptive neuro-fuzzy inference system (ANFIS) for detection and classification of activities on PPP. A dataset comprising 330 records was used for training, validation and testing of the system. Result of sensitivity test shows that inlet pressure, inlet temperature, inlet volume and outlet volume have cumulative significance of 71.72% on flowrate of PPP. Hybrid learning algorithm was observed to converge faster than the back propagation algorithm in the detection of pipeline activities. ANFIS hybrid learning algorithm with training and testing errors of 0.11980 and 0.010233 yielded a correlation of 0.916 between the computed and the desired output and produced optimal consequent parameters to boost the intelligence of DEVS. A testing error of 0.0303 was observed in the evaluation of DEVS-ANFIS system on 33 test data sample, 32 precise detections were made with one incorrect detection, this gives 96.97% level of confidence in the DEVS-ANFIS model for detection, classification and localization of PPP activities.

Machine learning approach for classification of <i>Dalium guineense</i> fruits

“…The modifications were done in backward direction from the output layer through each hidden layer down to the first hidden layer and the process was repeated until an acceptable low error is attained. In this case, the network is able to learn the data patterns (Udoh, 2016).…”

“…These modifications are done in backward direction from the output layer through each hidden layer down to the first hidden layer. The back-propagation algorithm is discussed in (Acharya et al, 2003;Han et al, 2012;Obot, 2007;Udoh, 2016;George, 2019).…”