Abstract: While active vibration control is carried out with piezoelectric smart structures, the control action of the actuators can generate structural internal positive feedback on the vibration reference signal collected by the reference sensor, leading to performance degradation or even failure of the controller. To ensure the stability of the internal feedback loop and enhance the adaptive and robust performance of the active vibration controller, this paper proposes a novel feedforward robust adaptive vibration controller based on the Youla-Kučera parameterization method by combining adaptive filtering control and robust control. By introducing historical data information to improve the parameter adaptation algorithm and incorporating a steady update strategy based on an error threshold, the accuracy of parameter estimation is enhanced. Furthermore, for system identification applications, an improved segmented parameter adaptation algorithm is proposed to separately estimate the parameters of different model segments, enabling more precise and rapid parameter estimation for models of varying orders. The effectiveness of the proposed Youla-Kučera parameterization-based feedforward robust adaptive controller with the improved parameter adaptation algorithm is verified under single-frequency, multi-frequency, and noise interference conditions using a piezoelectric smart blade of an aero-engine with bonded Macro-Fiber Composite (MFC) with xPC Target real time control system.