Disturbance Compensation‐Oriented Security Control for Delayed Singular Switched Semi‐Markovian Jump Systems With Hybrid Cyber‐Attacks

ABSTRACT This study sets out a framework to cope with the unified disturbance estimation and security control issue for singular switched semi‐Markovian jump systems (SSSMJSs) confronted with multiple stochastic delays, external disturbances and multiple cyber‐attacks. To accurately forecast the external disturbances, the proportional‐integral‐observer (PIO)‐based modified equivalent‐input‐disturbance (EID) estimator is first set up. Herein, the compensator and integral loop have been embedded in the traditional EID estimator and observer, respectively, aiming at improving disturbance estimation efficacy. Secondly, taking advantage of this disturbance estimate information, the disturbance rejection controller is designed to actively assist in dampening disturbances in the system model. Therein, the control segment notably accounts for multiple attacks and gain fluctuations, thus rendering the system's security and resiliency. In this context, multiple attacks embrace both denial‐of‐service (DoS) and deception attacks, which take place in a cohesive framework in a stochastic fashion and are independent of each other. Thirdly, sufficient criteria that ensure the sought‐after requirements are established using the linear matrix inequality method and Lyapunov stability theory. This is followed up by establishing a technique to design the controller gains by solving a string of linear matrix inequalities, centered on the stated criteria. Finally, the application example is presented, which shows the efflux of the design approaches and exhibits that the strategy enacted in this work can provide an intended performance.