Predictive safety assessment for storage tanks of water cyber physical systems using machine learning
Cyber physical systems (CPS) are critical to the infrastructure of a country. In addition to being vulnerable to hardware and software failures, and physical attacks, they are now becoming vulnerable to cyber attacks because of their use of off the shelf servers and industrial network protocols. Availability on World WideWeb, for monitoring and reporting, has further aggravated their risk of being attacked. Once an attacker breaches the network security, he can affect the operations of the system, which may even lead to a catastrophe. Variousmachine learning, mathematical and formal models try to detect the departure of the system from its expected behaviour. However, little or no work predicts how long the system would take to become unsafe. We here propose a machine learning predictive safety assessment approach that quickly calculates the time to being unsafe (TTBU) of a water-based CPS. We validate our results on a complete replicate of the physical and control components of a real modern water treatment facility. Our approach is fast, scalable and robust to noise. The model can be easily updated to match the changing behaviour of the system and environment.