Intelligent algorithms are easy to implement, have simple principle and good global convergence ability, so they are widely used in dynamic optimization of chemical processes in recent years. In this paper, an improved krill herd algorithm is proposed for solving chemical dynamic optimization problems. In the improved algorithm, the speed factor is updated according to the changes of the krill population to accelerate the convergence. Adaptive Cauchy mutation is employed to increase the ability of the algorithm escaping from local optimum. The control vector parameterization method is firstly used to convert the dynamic optimization problems into nonlinear programming problems, the changeable time interval distribution method is considered to optimize the time variable, then the resulting problems are solved by the proposed algorithm. Finally, the algorithm is applied to several dynamic optimization problems of chemical processes. The simulation results show the good feasibility of the algorithm for solving such problems.