Expected security constrained reactive power planning

Abstract

The adequate reactive power resources are expected to maintain system integrity under both normal and post contingency operations. Nowadays, the power systems are leading to high-risk operating points, hence, the number of probable contingencies, which should be considered in security constrained reactive power planning (SCRPP), has widely increased. However, considering a large number of contingencies in SCRPP is not practically possible. This study presents an index named expected violation index to consider numerous contingencies in SCRPP. At first, a multi-objective algorithm for the proposed expected SCRPP approach is presented. This novel algorithm uses the probability of each contingency to invest on them more efficiently. Next, it presents a method for selecting candidate buses to install reactive power resources. Then, the effectiveness of the proposed method is analysed using a multiobjective particle swarm optimisation algorithm on IEEE 30-bus test system. Finally, a reliability-based quantitative method is presented for selecting among final Pareto optimal solution. Using the proposed method, the system designers are capable to consider a large number of contingencies in their SCRPPs, to invest more purposefully in reactive power equipment, and to select their most appropriate option among different feasible choices in the optimal Pareto solution.

Toward coordinated look‐ahead reactive power optimisation for distribution networks with minimal control

“…However, the issue regarding the existence of feasible reactive power controls for each coherent period should be addressed. The theory of feasible region developed in [20] can be used to identify whether there exists a solution for (6)- (18). This paper assumes that the proposed reactive power optimisation formulation admits a solution.…”

“…Constraints (14)- (16) are, respectively, the voltage limit, transformers limit, and capacitor bank limit at the time period t. Constraint (18) ensures that the status of any control variable is either 0 or 1 while constraint (17) imposes bounds on the movement of control variables.…”

“…Since transformer taps and capacitor banks are discrete variables, reactive power outputs of DGs are continuous variables; therefore, the proposed problem formulations (6)- (18) constitutes an MINLP problem.…”

“…It is possible that the reactive optimisation problem formulations (6)- (18) does not admit a solution. The theory of feasible region developed in [20] can be used to identify whether a solution exists for (6)- (18). This paper assumes that the proposed reactive power optimisation formulation admits a solution.…”

“…It is important to have sufficient hourly ADCM (alternatively, load margin) to fully support the penetration of DGs [13]. We envision that the metric ADCM can be used as an objective function, see for example [14][15][16], or as an operational constraint, see for example [17][18][19]. From an operational viewpoint, we believe that it is more appropriate to treat ADCM as an operational constraint since the forecasting information of renewable energy is available and maximising the value of ADCM may be unnecessary.…”