An Integrated Trustworthy Detection and Classification of Cyber-Physical Attacks in the Presence of Disturbances Using Morphological Image Processing and Explainable AI
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Smart Digital Substations (SDSs), are cyber-physical systems (CPSs) that rely on communication networks to exchange information among physical devices, making such CPSs vulnerable to cybersecurity threats. The problem of detecting and classifying attacks in SDSs has been traditionally studied by considering machine learning as a black box with no interpretation of the decisions made, which has led to untrustworthy algorithms. The attack detection becomes more challenging in the presence of disturbances, as certain types of attacks may exhibit similar behavior to some disturbance events. Furthermore, some attacks may concurrently occur in the presence of disturbances, which may increase the misclassification rate. This paper presents a novel trustworthy approach for detecting and classifying attacks considering the simultaneous occurrence of disturbances in SDSs. This study uses Explainable Artificial Intelligence (XAI) to provide interpretability of the classification decisions using the cyber and physical features in SDSs. This method applies a series of processes, including the use of the Bartlett observation window and morphological image processing, to enhance the learning of the Convolutional Neural Network (CNN) to better extract the hidden features relevant to the attacks and the disturbances when applying the Continuous Wavelet Transform. The proposed approach achieved detection and classification accuracies of 99.37% and 98.44%, while reducing the computational time by 90%, due to the incorporation of a hardware acceleration of multiple graphics processing units (GPUs).
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it