A multi-objective optimization algorithm based on adaptive spatial division is proposed to solve the environmental/economic dispatch problem. As to keep the population diversity, the search space is divided into multiple regions, particles are guided by three kinds of local and global particles to rapidly near the Pareto optimal frontier, an age observer is used to record the contribution of guiders for particles near the Pareto optimal solution set real-timely, and guiders are changed in a certain cycle according to the contribution degree. The algorithm can fully explore the solution space, so as to quickly find a set of distribution with the best possible approximation. The experiment simulations on the international test function and the power system environment economic dispatch model are carried out. The results show that the improved algorithm can maintain the diversity of Pareto-optimal solutions and get better convergence at the same time.