Amba, Tushar (2009-05). Genetic Algorithm Based Damage Control For Shipboard Power Systems. Master's Thesis. Thesis uri icon

abstract

  • The work presented in this thesis was concerned with the implementation of a damage control method for U.S. Navy shipboard power systems (SPS). In recent years, the Navy has been seeking an automated damage control and power system management approach for future reconfigurable shipboard power systems. The methodology should be capable of representing the dynamic performance (differential algebraic description), the steady state performance (algebraic description), and the system reconfiguration routines (discrete events) in one comprehensive tool. The damage control approach should also be able to improve survivability, reliability, and security, as well as reduce manning through the automation of the reconfiguration of the SPS network. To this end, this work implemented a damage control method for a notional Next Generation Integrated Power System. This thesis presents a static implementation of a dynamic formulation of a new damage control method at the DC zonal Integrated Flight Through Power system level. The proposed method used a constrained binary genetic algorithm to find an optimal network configuration. An optimal network configuration is a configuration which restores all of the de-energized loads that are possible to be restored based on the priority of the load without violating the system operating constraints. System operating limits act as constraints in the static damage control implementation. Off-line studies were conducted using an example power system modeled in PSCAD, an electromagnetic time domain transient simulation environment and study tool, to evaluate the effectiveness of the damage control method in restoring the power system. The simulation results for case studies showed that, in approximately 93% of the cases, the proposed damage algorithm was able to find the optimal network configuration that restores the power system network without violating the power system operating constraints.
  • The work presented in this thesis was concerned with the implementation of a
    damage control method for U.S. Navy shipboard power systems (SPS). In recent years,
    the Navy has been seeking an automated damage control and power system management
    approach for future reconfigurable shipboard power systems. The methodology should
    be capable of representing the dynamic performance (differential algebraic description),
    the steady state performance (algebraic description), and the system reconfiguration
    routines (discrete events) in one comprehensive tool. The damage control approach
    should also be able to improve survivability, reliability, and security, as well as reduce
    manning through the automation of the reconfiguration of the SPS network.
    To this end, this work implemented a damage control method for a notional Next
    Generation Integrated Power System. This thesis presents a static implementation of a
    dynamic formulation of a new damage control method at the DC zonal Integrated Flight
    Through Power system level. The proposed method used a constrained binary genetic
    algorithm to find an optimal network configuration. An optimal network configuration is
    a configuration which restores all of the de-energized loads that are possible to be restored based on the priority of the load without violating the system operating
    constraints. System operating limits act as constraints in the static damage control
    implementation. Off-line studies were conducted using an example power system
    modeled in PSCAD, an electromagnetic time domain transient simulation environment
    and study tool, to evaluate the effectiveness of the damage control method in restoring
    the power system. The simulation results for case studies showed that, in approximately
    93% of the cases, the proposed damage algorithm was able to find the optimal network
    configuration that restores the power system network without violating the power system
    operating constraints.

publication date

  • May 2009