Physics-based detection of cyber-attacks in manufacturing systems: A machining case study

The overlap between operational technologies and information technology has resulted in profound improvements in the manufacturing ecosystem, but it increases the risk of a non-conventional class of cyber-attacks capable of inflicting physical damages on manufacturing processes and/or products. If successful in penetrating traditional cyber-only defenses, such attacks may not be detected timely, leading to financial losses, and potentially endangering human safety. However, malicious alterations of products and/or processes intended by these attacks can be manifested as anomalous changes in process dynamics. Hence, monitoring physical process variables such as vibration and power consumption (known as side-channels in cybersecurity literature) can provide a physical-domain defense layer to detect such attacks. Focusing on product-oriented attacks, we propose a method to connect the product design, process design, and in situ monitoring to identify the physical manifestations of these attacks. The proposed approach can verify the geometric integrity of a machined part by observing cutting power signals during machining. We utilize the process and product knowledge to segment the power signal into the cutting cycles corresponding to specific geometrical features and extract process-related information accordingly. This work primarily focuses on extracting machining times for individual geometric features in parts. Next, we use the extracted information to construct quality control charts to use in detecting geometric integrity deviations of machined parts. Finally, we demonstrate our proposed method using a case study of cyber-physical attacks on machining processes aiming to tamper with different product's dimensional and geometrical features.