PID and EPID types of Iterative Learning Control based on Evolutionary Algorithm

In order to reduce the number of iterations for convergence of Iterative Learning Control (ILC) system, this paper presents two types of ILC techniques, Evolutionary Algorithm (EA) based PID-type ILC and EA based EPID-type ILC. With the global optimization capability, EA is combined with PID and EPID types of ILC to select the optimal ILC parameters adaptively. In the EA based PID and EPID types of ILC, the encoding strategies and evolutionary operators are designed according to the ILC parameters characteristics. Because the convergent conditions of PID and EPID types of ILC are already known, population of EA is initialized to a certain interval. A comparative simulation example is provided to verify the effectiveness of the EA based PID and EPID types of ILC in reducing the number of iterations for a convergent ILC process.