<p>Sliding bearings are a vulnerable component of bridges, and monitoring their wear degree plays a key role in ensuring traffic safety. Therefore, this paper proposes an equivalent sliding friction coefficient to diagnose the wear degree of a bearing slide plate. First, the relationship between the bridge temperature difference and thermal-induced bearing displacement difference is analyzed, and a correlation model between the temperature difference and thermal-induced displacement is established by canonical correlation analysis. Then, after eliminating the changing trend of the friction coefficient of the bearing sliding plate under the effects of time-varying temperature, a diagnosis indicator of the equivalent sliding friction coefficient of the bearing slide is proposed, and a multilevel diagnosis method is established for the wear of the bearing slide based on an improved cumulative sum control chart. Finally, a case study of a long-span bridge verifies that a more reasonable temperature-displacement correlation model can be obtained through canonical correlation analysis, and the proposed equivalent sliding friction coefficient can effectively reflect the wear degree of bridge bearings. The results show that the proposed method can successfully conduct a multilevel diagnosis of bridge-bearing wear.</p>