Compared with the classical integer-order PID controller, fractional-order [[EQUATION]] (FOPID) controller is more difficult to adjust due to the introduction of two additional parameters. Aiming at the problem, a novel improved sparrow search algorithm, abbreviated as IMSSA, is proposed to perform the selection of optimal parameters for the FOPID controller. The algorithm, integrating strategies of adaptive exponential weighting factor and the stochastic mutation, enhances the searching efficiency and convergence performance of the sparrow search algorithm (SSA). The results carried on 12 benchmark functions show that the IMSSA has excellent performance in terms of convergence speed, accuracy and robustness compared with other popular algorithms, proving the effectiveness of the improved strategies. Finally, the feasibility of the IMSSA algorithm for parameter tuning of the FOPID controller is further verified by two simulation cases. The IMSSA proposed in this paper offers a novel approach to apply intelligent algorithms to solve fractional-order control problems.