This paper presents an improved Particle Swarm Optimization (PSO) algorithm for solving Transient Stability Constrained Optimal Power Flow (TSCOPF) problem through the application of Gaussian and Cauchy probability distributions. The modified PSO approach introduces new diversification and intensification strategy into the particles thus preventing PSO algorithm from premature convergence. The controllable system quantities are optimized to minimize fuel cost of the power generation. An IEEE 30-bus test system is taken for investigation. The transient stability constrained optimal power flow results obtained using the improved PSO models are compared with those obtained using standard PSO and GA algorithms. The investigations reveal that the proposed algorithm is relatively simple, reliable and efficient and suitable for on-line applications.