Strengthening DCS Processors and Networks – Recovery from a Significant ICS Cyber Incident

No ratings

Presented at ICScybersecCon 2014 by

Anoop Kesavan

Distributed Control Systems (DCS) are extensively used in power plants for end-to-end plant requirements including control, protections, operation, monitoring, alarms, event analysis, trends and reports. The basic reliability of such a system is an unspoken faith for operation and maintenance executives. This paper discusses processor firmware upgrade and network hardening activities, which were necessitated by unique multiple-unit trip incidents involving complete and simultaneous failure of DCS of two interconnected units in a simultaneous manner, leading to dangerous plant conditions. The root cause for the DCS failure was a deficiency in processor firmware, which caused each one to abruptly restart and lose logic configuration. The initiating cause was a network storm that got initiated in one of the two redundant networks due to an inadvertent loop of five Ethernet switches by design. The paper provides details of comprehensive design validation tests for DCS processors along with networks carried out on an innovative basis to confirm the sturdiness of the upgraded processor firmware and its hardened network against overloads or storms. LEARNING OBJECTIVES It is emphasized as conclusion that the dual exercise of strengthening processors and networks independently will eliminate all vulnerabilities of any DCS, and safeguard against a complete DCS failure event with a bullet-proof two-tier security.