Abstract: Existing individual pitch control (IPC) strategies for wind turbines exhibit limited adaptability under wide-range operating condition variations and struggle to consistently and effectively suppress unbalanced blade loads. To address this issue, this paper proposes an adaptive IPC strategy that integrates a Back Propagation Neural Network (BPNN) and fuzzy logic. First, the Sparrow Search Algorithm (SSA) is employed to conduct offline multi-objective optimization of the IPC PI controller parameters, with the goals of reducing loads and stabilizing power output. Subsequently, the optimization data are used to train a BPNN to construct a wind-speed-information-based feedforward layer, enabling the adaptive setting of parameter baseline values. Concurrently, a fuzzy logic feedback layer, which takes real-time load information as input, is designed for online fine-tuning of the parameters. Finally, simulation studies based on an OpenFAST-MATLAB/Simulink co-simulation platform for wind turbines are conducted under typical conditions such as turbulent wind with varying turbulence intensities. The results demonstrate that, compared with existing fuzzy-PI control strategies, the proposed strategy shows superior adaptability to wide-range varying conditions. Through the synergistic mechanism of its feedforward and feedback layers, it effectively reduces blade damage-equivalent loads and also mitigates output power fluctuations to a certain extent.