Controlled system islanding, also called system splitting, can effectively prevent blackouts of power systems. The splitting strategy determining how to split a power network into islands should be given in real time. Previous papers proposed an ordered binary decision diagram method to search for the splitting strategies satisfying necessary steady-state constraints, e.g. generation-load balance and transmission-line capacity constraint, and also showed that, without any other corrective controls, the splitting strategies may produce unstable islands according to a further simulation study. A modified method to find feasible splitting strategies in real time for large power network is presented. Each feasible splitting strategy not only satisfies necessary steady-state constraints but also easily produces stable islands to prevent a blackout. The method also introduces new techniques into strategy searching to increase its efficiency and practicality, e.g. network partitioning and parallel processing, generator classifying, deriving splitting strategies from cut-set splitting strategies, etc. Simulations on the IEEE 118-bus system show that the real-time portion of the modified method finds feasible splitting strategies in less than one second.

/pdf/searching-for-feasible-splitting-strategies-of-controlled-tvopf7dhwl.pdf

Searching for feasible splitting strategies of controlled system islanding

Controlled islanding using transmission switching and load shedding for enhancing power grid resilience

Power system stability analysis becomes more important in the presence of ever increasing variations in operating conditions. Traditionally, the operation of power systems is verified for specific operating conditions. In this work, the stability analysis is performed for a set of operating conditions using reachability analysis, which makes it possible to compute the bounds of all possible system trajectories. Thus, reachability analysis can be used to rigorously check specifications. Contrary to previous work, the presented approach does not require model simplifications when the system is described by semi-explicit, nonlinear, index-1 differential-algebraic equations. The main obstacle in reachability analysis is the scalability towards larger systems, which is addressed by investigating compositional techniques. As a result, transient stability and variable energy production can be analyzed for the IEEE 14-bus and 30-bus benchmark systems, for which the computation times are orders of magnitude faster than the simulation of all cases starting in the corners of the set of possible initial states.

/pdf/formal-and-compositional-analysis-of-power-systems-using-1bg9qi3ds5.pdf

Formal and Compositional Analysis of Power Systems Using Reachable Sets

Network partitioning using harmony search and equivalencing for distributed computing

Electric utilities are quite interested in operational cost saving measures. This implies that the global optimization technique is more attractive in power system operation and planning. As typical meta-heuristics, the tabu search (TS), genetic algorithm (GA) and simulated annealing (SA) have widely been adopted. TS is an extension of the hill-climbing method in a sense that TS has adaptive memory in the hill-climbing method. It is noteworthy that TS is a transition-type algorithm like chaos search. GA comes from the natural selection of genes in biology. Genetic operations improve the quality of solution in the multi-point solution search process. SA is based on the analogy of annealing of crystal or metal in heat bath. As temperature of SA parameter cools down, the solution converges. In particular, TS is more attractive due to the good performance. In this paper, applications of TS to power systems are discussed.

L.R. Lu

Papers

Power system network partitioning using tabu search