This paper presents Variable Acceleration Coefficient-based Particle Swarm Optimization (VACPSO) method to solve the economic load dispatch for minimizing fuel cost while considering prohibited operating zones and valve point effect. The proposed VACPSO is a modified version of the conventional Particle Swarm Optimization (PSO). Three modifications are suggested in the proposed VACPSO. First, the cognitive behavior of particle is influenced by best and worst experience. Second, three acceleration coefficients, instead of two as in the conventional PSO, are suggested. Third, the acceleration coefficients are made variable rather than assigning them merely a fixed value. These suggested modifications facilitate better exploration of the search space with less computational burden. The proposed method is tested on two standard generating systems available in the literature. The application results and comparison with other recent approaches show that the proposed approach provides good quality solution with less computational time.