The subsynchronous harmonics generated by a renewable energy collection area will affect the safety and stability of a power grid, and the harmonic detection accuracy of the traditional Fourier algorithm is easily affected by spectrum leakage and the fence effect. In order to improve the detection accuracy of subsynchronous harmonics, a new algorithm based on the Nuttall window and all-phase Fourier analysis (apFFT) is proposed. By dividing the sampled data into N segments and adding 4 terms 5 orders Nuttall windows twice, the pre-processed N-point data segments are obtained and FFT is performed. Further, based on the square relationship of amplitude between apFFT and traditional FFT and the “Phase invariance” of apFFT, the apFFT spectrum analysis results of amplitude and frequency are corrected. Compared with the eigenvalue analysis method and the interpolation correction FFT method, this method improves the accuracy of subsynchronous harmonic detection while maintaining the speed of FFT. The simulation results verify the effectiveness of the proposed method. This work is supported by National Natural Science Foundation of China (No.51607112).