Identification of maximum loadability limit and weak buses using security constraint genetic algorithm

Maximum loadability limit (MLL) is a realistic index to evaluate the steady state voltage stability because it provides system operator a better practical sense of security margin in the aspects of engineering parameter like system loading. If MLL is identified, load margin, voltage stability, security margin can be determined and precaution can be taken. Conventional power flow methods like Newton-Raphson, Gauss-Siedel, fast decoupled power flow methods suffer to provide proper MLL under security constraints as Jacobian matrix becomes singular when system loading approaches its loadability limit. Hence evolutionary techniques such as Particle Swarm Optimization (PSO), Genetic Algorithm (GA) are now-a-days applied to solve the non-linear complex MLL problem. Phase angles, active power, voltage magnitudes of load buses, reactive power of PV buses are considered as security constraints for MLL problem. New simple real coded Security Constraint GA (SCGA) is developed to solve the problem. MLL problem is formulated as maximization problem. As handling of real coded power flow variables are easier than binary coding, real coding of SCGA parameters is applied. Novel formulas are developed to update power flow parameters considering corresponding power mismatches. Utilizing decoupling properties of power system, mutation is implemented. To provide diversity, new parent selection in crossover section is adopted. Weak buses are also identified for the application of FACTS devices. The developed method is compared with general PSO (GPSO) technique for test systems of IEEE 14, 30, 57 and 118 bus. Showing characteristics and results, the effectiveness and efficiency of the proposed method is established. (C) 2011 Elsevier Ltd. All rights reserved.