Water distribution systems analysis symposium-battle of the attack detection algorithms (BATADAL)

With the transition of water distribution systems (WDSs) to smarter ones, intelligent water networks' elements, such as programmable logic controls (PLCs), sensors, valves and supervisory control and data acquisition systems (SCADAs) have played a more significant role. Like other computer-based technologies, these smart elements will make WDSs more vulnerable to malicious intrusions due to cyber-physical attacks. Regarding the vulnerability of more intelligent WDSs, it is necessary to devise and apply anomaly detection algorithms being able to detect intrusions to water networks with the fewest false alarms. For the battle of the attack detection algorithms (BATADAL), to detect the manipulated data, a spectral domain method as a pre-processing technique is implemented to extract the important characteristics of the observed time series data and make them independent of time. Then, a supervised machine learning technique is used to classify the data and obtain the intrusion detection.