To achieve high-precision and high-stability detection of wind speed and direction in complex environments, this research proposes a dual closed-loop control scanning technique for the wind sensor system based on the acoustic resonance principle. This technique has been found to significantly enhance the system’s performance indicators. The acoustic resonance method used on wind sensors allows for the simultaneous modulation of frequency and intensity of signals generated by the transducer, resulting in linear scanning of the ultrasonic transducer. Frequency modulation resolves the issue of a resonance frequency shift caused by environmental factors like pressure and temperature, while intensity modulation addresses transducer performance degradation over time and can significantly improve the signal-to-noise ratio. However, when confronted with issues such as wind shear, the rapid change in the ambient pressure of the wind sensor may lead to the failure of the frequency modulation, followed by the change in the rate of wind shear, resulting in significant errors in wind speed detection. Therefore, the dual closed-loop control method is used to combine the frequency scanning modes—the slow and long scanning and the short and fast scanning. The slow and long scanning is used to solve the resonance frequency shift caused by various slow external changes and achieve frequency following, while the short and fast scanning resolves the resonance frequency shift resulting from rapid changes in wind shear and achieves rapid frequency following. Experimental results demonstrate that the scanning method employing dual closed-loop control can accurately measure wind speed and direction. The wind speed measurement range is 0–50 m/s, with a measurement accuracy of ±0.3 m/s (≤15 m/s)/±4% (>15 m/s), while the wind direction measurement range is 0°–360°, with a measurement accuracy of ±3°. After improvements, the system has high accuracy and stability and strong anti-interference ability and is less affected by environmental changes in complex environments.