Abstract: Restoration of phase aberrations are crucial for addressing atmospheric turbulence involved light propagation.Traditional Zernike polynomial methods face high computational complexity and poor capture of high-frequency components, so we propose a Principal Component Analysis-based representation method. This paper analyzes factors affecting restoration accuracy, focusing on the size of sample space and sampling interval of D/r0 ,with r0 being the atmospheric coherence length and D being the pupil diameter, Results show PCA outperforms Zernike methods, especially in strong turbulence, and larger sampling intervals improve accuracy with less data.These findings pave a way to use PCs of phase aberrations with less orders than traditional ZPs to achieve data dimensionality reduction, and offer a reference to accelerate and stabilize the model based and deep learning based adaptive optics correction.