Publication Date

8-2022

Date of Final Oral Examination (Defense)

4-15-2022

Type of Culminating Activity

Thesis

Degree Title

Master of Science in Computer Science

Department

Computer Science

Supervisory Committee Chair

Hoda Mehrpouyan, Ph.D.

Supervisory Committee Member

Bogdan Dit, Ph.D.

Supervisory Committee Member

Craig Rieger, Ph.D.

Abstract

Control logics, as part of the Industrial Control Systems (ICS), are used to control the physical processes of the critical infrastructures such as power plants, water, and gas distribution, etc. Most commonly, the Programmable Logic Controller (PLC) manages these processes through actuators based on information received from sensor readings. Any safety issues or cyberattacks on these systems may have catastrophic consequences on human lives and the environment. In an effort to improve the resilience and security of control logics, this thesis provides algorithms and tools to formally define the safety and security requirements w.r.t. the physical processes, and the industrial domain. Web Ontology Language (OWL) is utilized to create semantic relationships between the elements of industrial processes and knowledge mapping of the input and output of the control logic. Description Logic (DL) knowledge bases derived from OWL allow us to reason about the semantic security and safety concepts to ensure their consistency. Next, these formal specifications are translated to Timed Computational Tree Logic (TCTL) queries for the verification of the control logic modeled in UPPAAL as a network of timed automata (TA). In the second part of the thesis, boundary conditions are checked to perform a model verification. Boundary checking is essential in ICS because the sensor’s readings and actuator’s values need to be within the safe range to ensure secure ICS operation.

For the proof of concept, we have studied a part of an industrial chemical process to implement our proposed approach. Experimental results in this work proved that the proposed method detects the inconsistencies in safety and security requirements and ensures that the input and output variables of the control logic are within a safe and secure range. The performance study of our implementations shows that the time grows linearly with the number of axioms in the ontology and the number of iterations in TA model simulations. Hence, the approach is scalable to have a practical implementation to help the technicians and engineers to create a safer and more secure control logic for ICS processes.

DOI

https://doi.org/10.18122/td.1983.boisestate

Download

Share

COinS