Fast and Flexible Design of Optimal Metering Systems for Power Systems Monitoring

Abstract

Metering systems planning in transmission networks can be modeled as an optimization problem in which the metering system cost is minimized, subjected to redundancy constraints that enable the state estimation function to observe system state and debug bad measurement data. However, the budget for investments in the metering system is commonly insufficient to achieve such a goal for the entire power system. On the other hand, there are areas of the power network considered more important for system operation than others. Consequently, a more qualified supervision of such areas is necessary to ensure adequate decision making regarding network operation and control. This work proposes a methodology flexible and easily adaptable to cope with limited investment budgets. An ant colony optimization algorithm (ACO-PCH) is employed to solve the optimal meter placement problem. A fast and cost-effective algorithm based on a decomposition strategy (ACO-PCH-DS) that takes into account the local nature of the state estimation problem is also proposed in order to improve computational efficiency of planning metering systems for large power networks. Tests with the IEEE bus systems and part of a real Brazilian system are carried out to validate the proposed methodology. If the stop criteria are the time limit (200 s), the ACO-PCH-DS reduces in 15% the cost of the measurement plan in IEEE 118-bus system, when compared to the ACO-PCH. For 300 min as a time limit, the ACO-PCH-DS do not provide a better cost only for reliability in the IEEE-300 bus system.