<p>Bulging deformation not only aggravates equipment wear but also easily causes center segregation and internal cracks inside the slab. Constant bulging will lead to periodic abnormal fluctuations in the mold, increasing the risk of slag entrainment. In this study, the data was collected in the continuous casting process of peritectic steel. The wavelet transform method is used to analyze the time-frequency characteristics of the mold level fluctuation and inner-bow-driven roller torque (IDT). The mechanism of the effect of bulging on mold level fluctuation was revealed. It finds that periodic abnormal fluctuations are only an indirect expression of bulging. The abnormal periodic fluctuations and the IDT at different segments caused by bulging have strong periodicity and consistency in time-frequency distribution. The wavelength of the bulging does not have a clear relationship with the roll pitch, and it is mainly affected by the uneven solidification inside the shell and the deformation of the slab. The change of the IDT at different segments can reflect the uneven solidification characteristics inside the shell intuitively. However, it is difficult and inaccurate to determine which specific roll pitch is responsible for bulging using only the wavelength formula. For this purpose, a rapid identification model based on wavelet transform was established for bulging. The main advantage of this model is to identify unsteady bulging faster and more accurately. The model obtains the main wavelength of bulging and then can be quickly adjusted manually after the anti-bulging model fails. It is of great significance to improving the quality of the peritectic steel slab and controlling the mold level fluctuation.</p>