||When tall buildings are subjected to strong wind, the sway caused by wind excitation may lead to the discomfort of occupants. In general, the discomfort can be minimized by reducing the acceleration of the structures. Active control is an effective way to alleviate the dynamic responses. However, the existing control algorithms for active structural control includes a series of complicated calculations and requires a long computation time; in turn, the time lag effect becomes serious. In this study, a fuzy logic control algorithm is presented with the displacement and acceleration inputs for control of the acceleration of tall buildings under wind excitation. The proposed fuzzy logic control algorithm is developed by use of the trial and error method. The effectiveness of the algorithm is studied by computer simulation. To simulate the fluctuating wind forces acting on tall buildings in along-wind and across-wind directions, the autoregressive and moving average (ARMA) recursive models are used by the two-stage matching approach. Based on the fuzzification, the formulation of inference rules and the defuzzification, the fuzzy logic control algorithm is derived and presented by a 13 x 13 numerical table. The control forces can be generated conveniently by matching the table with the on-line inputs. Hence, the computation time for the generation of control forces is greatly reduced. The fuzzy logic control algorithm is easily derived. It provides an effective way for control of the acceleration responses of tall buildings in both along-wind and across-wind directions within a safety limit of lateral drift. An example tall building of 160 m high with a square cross-section modeled in single mass system, three-lumped mass system and six-lumped mass system are investigated. It is shown that the proposed fuzzy logic control is one of the most effective and economical ways for active structural control.