For the constrained non-linear dynamic environmental economic power dispatch (DEED), a multi-objective crisscross optimization algorithm (MOCSO) is proposed. The two conflicting, constraining objective functions of fuel cost and pollutant emission in the process of dynamic dispatch are optimized simultaneously. In the optimization process, combined with non-constrained domination principle, this paper presents a new double cross mechanism to enhance the ability of crossing feasible regions, so that the Pareto optimal solution may falls within the feasible region. With searching on the periphery, the algorithm's ability of global searching is improved greatly. Meanwhile, the proposed approach adopts an external elitist archive to retain non-dominated solutions and maintain the diversity by using the crowded-comparison operator. Finally, fuzzy theory is used to obtained the best compromise solution. Simulation results of 10 generator test system validate the effectiveness and advantages of the proposed method.