Abstract
In Additive Manufacturing (AM), auditing layer-by-layer images can detect infill defective attacks effectively. However, the auditing process itself can become a target of inside or outside attackers in Cyber-Physical Manufacturing System (CPMS) environments because pervasive connection through various types of computer networks in CPMS opens new doors for adversaries to compromise various components in an attack detection system. To maintain an effective attack detection system and prevent data from malicious data injection, this paper presents a Layer Image Auditing System (LIAS) secured by the Blockchain technology in CPMS. LIAS consists of a pre-processing system and a Multilayer Perceptron Neural Network (MLP). To evaluate the prediction accuracy of LIAS, a set of simulated infill images and physical images were used for training and testing. The effectiveness of the Blockchain implementation is demonstrated by presenting the comparative performance analysis of the proposed attack detection system with and without the Blockchain.
Original language | English (US) |
---|---|
Pages (from-to) | 585-593 |
Number of pages | 9 |
Journal | Procedia Manufacturing |
Volume | 53 |
DOIs | |
State | Published - 2021 |
Event | 49th SME North American Manufacturing Research Conference, NAMRC 2021 - Cincinnati, United States Duration: Jun 21 2021 → Jun 25 2021 |
Keywords
- Additive manufacturing
- Blockchain
- Cyber-physical manufacturing system
- Image classification
- Machine learning
- Neural networks
ASJC Scopus subject areas
- Artificial Intelligence
- Industrial and Manufacturing Engineering