Cyber-Physical Operation Security with a Collaborative Knowledge Grid System Asesh Das 1 School of Business & Computer Technologies Pennsylvania College of Technology [Penn State] 1 Adjunct Faculty, CSEE Dept, W V University Williamsport, PA 17701, USA Email: adas@pct.edu Ramana Reddy 1 , Luyi Wang 2 & Sumitra Reddy 3 Department of Comp. Science & Electrical Engineering West Virginia University, Morgantown, WV 26506, USA Email: 1 ramana.reddy@mail.wvu.edu; 2 wangluyi1982@gmail.com 3 sumitra.reddy@mail.wvu.edu Abstract--A cyber-physical system is treated with a collaborative self-organizing knowledge network model, called Vijjana. While traditional communication between users of Vijjana forms a loosely coupled system, in cyber-physical system it is tightly- bound forming a policy-driven knowledge grid. If there is a compromised security in a cyber-physical system, the knowledge grid bootstraps to a subset grid, isolated from other nodes. A model for this bootstrap is proposed, which looks for carefully administered security policies in the cyber-physical system. Keywords--Cyber-Physical security, Security Theory, Interactive Knowledge Network, Vijjana Model. I. INTRODUCTION Eight priorities in the area of smart grid operations include [1]: demand response and consumer energy efficiency, wide- area situational awareness, energy storage, electric transportation, advanced metering infrastructure, distribution grid management, cyber security and network communication. It is also stressed that all technologies supporting these aforesaid items need to follow: “…development and maintenance through collaborative and consensus-driven processes—open to all relevant parties.” We are proposing to bring a collaborating tool [called Vijjana] which is developed to address domain-specific needs of any person at any time [2, 3]. This collaborative knowledge network model developed at the West Virginia University is essentially given as: Vijjana-X = {J, T, R, d A , o A , c A , v A , s A , r A } (1) Where: X = the domain name, J= the collection of Jans in the domain Vijjana-X, T = the taxonomy used for classification of Jans, R= the domain specific relations, d A = the discovery agent which find relevant Jans, o A = the organizing agent which interlinks the Jans based on R, c A = the domain consistency/completeness agent, v A = the visualization agent, s A = the Jans search agent, and r A = the Jans rating agent. In this model users are contributors too, and they share domains of common interest. This model provides appropriate knowledge based schemata and identifies all knowledge units creating the knowledge base. The knowledge units in Vijjana are called Jans (a derivative from the Sanskrit word Jnan, which means knowledge). In the model Vijjana, all types of Jans are discoverable in the domain ontology and classifiable categorically depending on their context of existence, which can be visually displayed. Since this collaborative, knowledge network model can actually capture domain ontology as desired and exercised— which also comes with user ranked [focusing necessity or importance of the knowledge captured] support—it is worthwhile to incorporate the product in a model cyber- physical operation as in the Figure 1 below. Secure Communication Flows Market Operations Service Provider Electrical Flows Bulk Generation Transmission Distribution The collaborative Knowledge network system: Figure 1 Customer VIJJANA 978-1-4673-1375-9/12/$31.00 ©2012 IEEE