[Objective] Inaccurate LiDAR measurements under harsh weather introduce uncertainty in feedforward pitch control

which can degrade turbine performance. To address this challenge

this paper proposes a novel fuzzyswitching feedforward pitch control algorithm for wind turbines. The algorithm enhances reliability by intelligently integrating direct Li DAR wind preview measurements with a Kalman-filter-based equivalent wind speed estimation

ensuring robust operation despite measurement inaccuracies. [Methods] First

a self-developed nacellemounted pulsed coherent Doppler LiDAR system captures line-of-sight wind speed data at multiple upstream points. These measurements are synthetically processed to derive a single equivalent wind speed that represents the effective aerodynamic inflow acting on the rotor. Second

a Kalman-filter-based estimation algorithm is designed to independently obtain an equivalent wind speed signal using standard turbine operational data and a simplified drive-train and aerodynamic model. A fuzzy-logic mechanism is then applied to dynamically evaluate the reliability of both wind speed sources based on real-time quality indicators

and to implement smooth

weighted switching between the Li DAR-measured and the Kalman-filter-estimated signals. This ensures the feedforward controller predominantly utilizes the more trustworthy input at any moment

thereby mitigating the impact of Li DAR inaccuracies. Finally

to fully coordinate the feedforward action with baseline feedback regulation

the parameters of the PI-based pitch controller are optimized offline using the Wind Driven Optimization(WDO) algorithm. WDO adjusts the controller gains to minimize a composite cost function that penalizes both generator speed deviations and blade root flapwise bending-moment fluctuations

thereby achieving an effective balance between speed regulation and load mitigation. [Results] The proposed control strategy was tested and validated using the high-fidelity NREL 5 MW wind turbine model under turbulent wind conditions with a mean speed of 18 m/s. Results show that the fuzzy-logic switching mechanism alone reduces the standard deviation of generator speed fluctuations by approximately 5.3% compared to a conventional Li DAR-only feedforward scheme. Following offline optimization of the feedback controller parameters

the performance gains are further enhanced: the standard deviation of generator speed is reduced by about 8.9%

and the blade-root flapwise bending moment(x-direction) is reduced by approximately 8.7%. [Conclusion] This work demonstrates that the integrated fuzzy-logic switching mechanism and offline parameter optimization effectively mitigate the adverse impacts of inaccurate LiDAR windpreview data on feedforward pitch control. By intelligently blending LiDAR measurements with a Kalman-filterbased estimator

the system maintains reliable wind preview input under uncertain measurement conditions. Consequently

the proposed strategy significantly reduces fluctuations in both generator speed and blade-root loads

enhancing both the power quality and structural longevity of the wind turbine.