6 th Global Conference on Power Control and Optimization, Las Vegas, USA, 6-8 August 2012 AN APPLICATION OF PSO IN OPTIMAL LOAD SHEDDING CONSIDERING VOLTAGE STABILTY M. M. Hosseini Bioki 1 , M. Rashidinejad 2 ,R. Fadaeinedjad 1 and H.R. Esmaeilian 1 1 Department of Electricaland Computer Engineering, Kerman Graduate University of Technology, Kerman, Iran 2 Department of Electrical Engineering,Shahid Bahonar University of Kerman, Kerman, Iran IAEEE, Kerman Branch, Kerman, Iran m.hosseinibioki@kgut.ac.ir Abstract Optimal power system load shedding in restructured power system considering market issues and voltage stability usingParticle Swarm Optimization (PSO)is presented in this paper. A multi-objective function that comprises economic and technical sub-optimization problems is considered to provide an optimal load shedding scheme with maximumGenerating Company (GenCo) profit, and as a consequence maximum social welfareas well as maximum loadability limit. Particle swarm optimization (PSO) is utilized to find the optimal scheme of load shedding when a component outage occurs. The proposed methodology is obtained using modifications in MATPOWER 4.1 software integrating continuation power flow and PSO algorithm. In the smart market (SM) procedure, the generators offers and dispatchableloads bids have been considered where the Independent System Operator (ISO) as the smart market operator decides on the market clearing price and generators rescheduled amount of power generation in pre- contingency as well as post-contingency situations. The proposed methodology is applied to a modified IEEE 30-Bus test system. The results obtained show the effectiveness of the proposed method in procurement of an optimal load shedding satisfying both GenCo and ISO aims to maximize profit and social welfare as well as maintaining the power system voltage stability margin. Keywords: Continuation power flow, optimal load shedding pricing, particle swarm optimization (PSO). 1. Introduction Secure and economic operation of power systemin restructured environment involves balance of generation and load, continuity of response to its energy demand, quality of the power delivery, cost minimization or profit maximization of entities attending the power market, and maximization of social welfare. Any disturbance in power system Like generator or line contingency, sudden increase in power system load or any other component outage leads to insecure operation or voltage instability of power system.Voltage stability is considered as the ability of a power system to maintain steady acceptable voltages at all buses in the system under normal operating conditions and after being subjected to a disturbance from agiven initial operating condition [1]. Considering classification of power system as normal, alert,emergency, in extremis and restorative, load shedding is applied when the system is in the emergency or in extremis state and the system is driven to collapse [2]. Load shedding (LS) is generally categorized in two well-known methodologies of under-frequency load shedding (UFLS) and under-voltage load shedding (UVLS). UFLS or UVLS are performed when the frequency or voltage falls below a specified frequencyor voltage threshold using under-frequency or under- voltage relays. The load shedding procedure curtails an amount of load in a manner that the active and reactive generation and the demand become balanced and a widespread system black-outis prevented[3]. The main factors considered in load curtailment procedure are the location, amount, and time of load cut. Load shedding implementation can be categorized into three main classifications [4-6]. The first procedure involves shedding a fixed amount of load which is similar to UFLS [7]. This procedure involves time simulation analysis and studying instability from dynamic aspects of power system.One of the difficulties with this approach is integrating a time domain simulation to an optimization problem [8]. The second approach to load shedding considers shedding an amount of load with dynamic characteristics. However, this method involves precise modeling of dynamic load parameters. The third category utilizes optimal power flow (OPF) equations considering static model of power system to determine the minimum load shedding. This static approach has the advantage of proposing a good approximation of system voltage stability, since the dynamics associated with voltage stability are often slow. The fundamental idea of this approach is finding a feasible solution to the power flow equations [9, 10]. Load shedding considering static model of power system is presented in previous researchesand is solved using many mathematical techniques such as linear programming (LP), non- linear programing, and interior point methods [11- 13]. In some other researches, evolutionary algorithms such as genetic algorithm (GA) [14, 15], particle swarm optimization (PSO)[4-6, 16], and ant colony optimization(ACO) [17] have been utilized to tackle the load shedding problem. The advantages of these evolutionary algorithms are their accuracy in obtaining feasible solution and considering all problem constraints. However, in