On the premise of ensuring safety and reliability in electricity market environment, the goal of State Grid Corporation is that purchase AGC ancillary service charges of reducing cost. This paper first takes total expense from many AGC units as an objective function, , which synthetically considers total regulation MW amount and total regulation speed constraints. A novel hybrid particle swarm optimization (PSO) algorithm is applied to solve the problem. Numerical simulation results show that the improved PSO algorithm has advantages both in the calculation accuracy and the convergence speed. Therefore, it is concluded that the algorithm is supposed to be an effective way to deal with the optimized issue in the power market.