Distributed intelligent agents for service restoration and control applications

Typically, the electricity power network is divided into two basic networks: the transmission network and the distribution network. Other than the voltage range of operation, the distribution network differs from the transmission network in that it is mostly radial and also has many points of connection that are normally open. This allows various configurations by the operating utility by the simple control of closing and opening switches. Under normal circumstances, the system is operated and coordinated from the control center. With new technological advances the operation of these switches can now be remotely controlled from a control center. Every time that the power network fails the central control center determines which system elements and control actions should be implemented to either save the system from collapse or to reconfigure or restore the system after an outage. In this research, the concept of distributed agents connected to substations instead of controlling the system by a control center is being studied. These agents would be in charge of starting local system response in the case of a system blackout or any other abnormal situation. Agents would be required to communicate among themselves to coordinate the control actions. Here, simulations scenario on power system restoration and losses minimization using the distributed agents simulated in Matlab will be presented.