CAREER: Scalable Learning and Adaptation with Intelligent Techniques and Neural Networks for Reconfiguration and Survivability of Complex Systems

Computational Intelligence (CI) techniques are useful tools for modeling, dynamic reconfiguration, control and optimization. The nonlinear stochastic power network requires different mathematical models as system parameters and network topologies change. Intelligent system identification and control improves the dynamic and transient behavior of power networks, damps inter-area oscillations, and provides improved voltage and frequency stability. Increasing the power transfer across existing transmission lines with Flexible Alternating Current Transmission Systems (FACTS) devices and High Voltage Direct Current (HVDC) systems, improving the utilization of wind power plants, improving generation and grid reliability, are all important for local and wide area stability of the power network, and assisting the expert manpower in control rooms. CI techniques are suitable techniques for scheduling power system elements for maintenance taking in account a number of objectives and constraints. This paper briefly describes the activities and accomplishments during the 2007 summer IREE visit to the Corporate Research Center, ABB, Västerås, Sweden and the Royal Institute of Technology, Stockholm, Sweden. Preliminary work in the design of damping controllers for HVDC systems, the feasibility of CI techniques for the design of offshore wind park topology and generator maintenance scheduling have been carried out.