CYBER SECURITY FOR CHEMICAL PLANT USING ARTIFICIAL INTELLIGENCE

The adding number of cyber-attacks on diligence demands immediate attention for furnishing further secure mechanisms to guard diligence and minimize pitfalls. An administrative control and data accession (SCADA) system employing the distributed networks of detectors and selectors that interact with the physical terrain is vulnerable to attacks that target the interface between the cyber and physical subsystems. These cyber-attacks are generally vicious conduct that beget uninvited results in the cyber physical world, for illustration, the Stuxnet( 2010) attack that targeted Iran's nuclear centrifuges. An attack that hijacks the detectors in an attempt to give false readings to the regulator can be used to dissemble normal system operation for the control system, while the bushwhacker can commandeer the selectors to shoot the system beyond its safety range. AI result can identify shadow data, cover for abnormalities in data access and alert cybersecurity professional about implicit pitfalls by anyone penetrating the data or sensitive information. This proposes a process- apprehensive approach with the use of steady equations grounded on the physical and chemical parcels of the process and a Multiple Security sphere Nondeducibility (MSDND) frame to descry when a detector signal is being virulently manipulated. A system without any MSDND secure information flows between the AI and cyber observers has smaller sins that can be exploited.