Data Driven Diagnosis for Large Cyber-Physical-Systems with Minimal Prior Information

TL;DR

This paper introduces a novel diagnostic method for large cyber-physical systems that requires minimal prior knowledge, combining neural networks and graph algorithms to identify causal components effectively.

Contribution

The paper presents a new diagnostic approach that operates with minimal prior knowledge, integrating neural symptom generation and graph diagnosis for complex systems.

Findings

Achieves 82% inclusion of true causal components in diagnosis

Reduces search space in 73% of cases

Demonstrates effectiveness on simulated and real-world datasets

Abstract

Diagnostic processes for complex cyber-physical systems often require extensive prior knowledge in the form of detailed system models or comprehensive training data. However, obtaining such information poses a significant challenge. To address this issue, we present a new diagnostic approach that operates with minimal prior knowledge, requiring only a basic understanding of subsystem relationships and data from nominal operations. Our method combines a neural network-based symptom generator, which employs subsystem-level anomaly detection, with a new graph diagnosis algorithm that leverages minimal causal relationship information between subsystems-information that is typically available in practice. Our experiments with fully controllable simulated datasets show that our method includes the true causal component in its diagnosis set for 82 p.c. of all cases while effectively reducing the search space in 73 p.c. of the scenarios. Additional tests on the real-world Secure Water Treatment dataset showcase the approach's potential for practical scenarios. Our results thus highlight our approach's potential for practical applications with large and complex cyber-physical systems where limited prior knowledge is available.