Case-Based Reasoning Approach to Voltage Security Assessment of Power Systems

This paper presents case-based reasoning (CBR) technique, a novel approach for voltage security assessment in real time environment. The CBR is an artificial intelligence methodology; extract the information of voltage security from large sets of possible operating conditions of the power system, which are screened off-line by randomly varying the real and reactive loads at PQ buses. In real time environment, data-base may be created by keeping the record of operating states by state estimation. The severity of contingency for each case with respect to different contingencies (single line outages) is assessed by number of secure and insecure operating states, which is predicted by CBR. Security classes are defined by threshold value of maximum loadability margins, calculated using the continuation power flow analysis. The effectiveness of the proposed approach is tested on 30-Bus IEEE test system. The proposed method gives fast and accurate voltage security information for unknown patterns and it is found to be suitable for on-line applications at energy management systems.