Abstract: Abstract: Nowadays, the test methods of dynamic frequencies of typical vibrations of the wind wheel are badly in need of the wind turbine design, but the effective methods are few because installing the accelerometers on the blades will destroy the flow field and the structure field, and then greatly affect the test reliability. In this paper, an indirect test and identification method of dynamic frequencies of the wind wheel was developed based on the frequency holding characteristic of each sub signal in the multiple mixed vibration signals' transmission. The specific method was to get the acceleration signal from the accelerometer installed at the front end of the generator, and then to get its spectrum signature by the FFT (fast Fourier transform), and finally the dynamic frequencies of typical vibrations of the wind wheel were discriminated based on its mode shapes and spectrum signature. This method was carried out on a small horizontal axis wind turbine with 1.4 m wind wheel diameter while the wind velocity was 10 m/s and the tip speed ratio was 6, and achieved good results. The same tests were done while the wind velocities were 5-10 m/s and the tip speed ratios were 5-7 in order to prove the reliability of the above method, and the results could all prove its reliability. On the other hand, the numerical simulation was carried out based on one way fluid solid coupling. The vibration modes and vibration frequencies of the 2nd-order following typical vibrations were obtained while the wind velocity was 10 m/s and the tip speed ratio was 6, the results showed that the test accuracy of this method was relatively high because the relative errors between the calculated and experimental values were all less than 5%.Thus, it also proved that the above indirect test and identification method was reliable. Finally, considering the applicability of this method for different blade materials and structures, other 2 kinds of blades were tested by using the same method. One blade material was engineering plastic, and its material and structure had very big difference from the above measured blade. The other blade material was the same as the above measured blade, but its airfoil structure was NACA4415 which had big difference from the above measured blade. Both of the test results were good to exclude the above question. In other words, the test results confirmed that the indirect test and identification method of dynamic frequencies of typical vibrations of the wind wheel in this paper has a good adaptability to the wind wheels which have different materials or structures under the wind turbine operating conditions. Research also revealed that this method has a significant advantage in distinguishing the dynamic frequencies of the 1st-order and the following modes, but the recognition effect is obviously decreased along with the rise of the vibration order owing to the enhancement of the intensity of the torsional mode of blades and the decrease of the peak values of dynamic frequency curve. This method proposed in this paper can acquire the dynamic frequencies of typical vibrations of the wind wheel simply and accurately under the premise of not destroying the flow field and the structure field, because it does not need to install a large amount of sensors on the surface of the blade, so it can provide a solution for technical difficulties in monitoring and discriminating the dynamic vibration frequencies under the operating conditions of the wind wheel. At the same time, this research can provide a new reference method for the health monitoring of the wind wheel.