Understanding the complex in-cylinder flow field under high swirl ratio conditions is crucial for optimizing combustion process in Spark-Ignition Direct-Injection (SIDI) engines. Large-Eddy Simulations (LES) have been applied to simulate the in-cylinder flow field and quantify the cyclic variations of engine flows under motoring or firing operations. In order to construct reliable LES models, the simulation results of in-cylinder flow fields with high temporal and spatial resolutions are commonly calibrated with highspeed Particle Image Velocimetry (PIV) data. [1] Typical comparisons between experimental PIV flow field measurement data and numerical LES simulation results are illustrated in Fig 1. This study employs data-driven sparse identification of nonlinear dynamics (SINDy) to extract a concise partial differential equation (PDE) representing the Navier-Stokes equation governing this intricate flow. [2] Specifically, the discrete PIV flow field data is analysed using sparse regression technique LASSO (Least Absolute Shrinkage and Selection Operator) to select nonlinear terms that most effectively characterize the system's dynamics. This data-driven methodology could bridge the gap between experimentation and simulation through discrepancy examination and coefficient analysis. Firstly, comparing PDE-reconstructed flow fields with experimental or simulation results helps identify whether disparities stem from simulation model simplifications or uncertainties in experimental measurements, which is crucial for optimizing simulation models and refining experimental setups. Additionally, the temporal evolution of the PDE coefficients provides valuable insights into the dynamic behavior of airflow at different phases of the engine cycle, shedding light on the intricate interplay of fluid forces during critical stages such as early intake stroke, near bottom-dead center, and late compression stroke. In summary, not only could this data-driven approach provide a reliable way for PDE reconstruction of the experimental data, it could also reveal the physical quantifications of the complex in-cylinder flow characteristics.