A fuzzy active disturbance rejection control method for a carbon fiber diagonal loom based on an improved black-winged kite optimization algorithm is proposed to improve warp tension control accuracy. Piecewise chaotic mapping is introduced to enhance the diversity of the initial population and accelerate convergence, while an adaptive spiral flight strategy and a longitudinal-transverse crossover strategy are employed to balance global exploration and local exploitation and avoid premature convergence. The improved black-winged kite optimization algorithm is used to iteratively optimize the quantization and proportional factors of the fuzzy controller, as well as the initial gains of the extended state observer and nonlinear state error feedback in the active disturbance rejection controller. Simulation results show that, compared with other control methods, tension overshoot is reduced by at least 85% and the settling time is shortened by at least 52.6%. Under disturbance and parameter perturbation conditions, the peak tension is reduced by at least 2.2% and 0.9%, while the recovery time is shortened by at least 46% and 49.2%.