Improving intrusion detection in SCADA systems using stacking ensemble of tree-based models

This paper introduces a stacking ensemble model, which combines three single models, to improve intrusion detection in supervisory control and data acquisition (SCADA) systems. The first layer of the proposed model is combination random forest, light boosting gradient machine, eXtreme models. We use an multilayer perceptron (MLP) network as meta-classifier model. optimized tested on international dataset (gas pipeline dataset). results show accuracy 99.72% with f1-score for binary classification tasks (attacked or non-attacked detection). For categorical tasks, rates almost all attack types are higher than 97.55% (except denial service (DoS)-95.17%), overall 99.62%.