Typhoons are recurrent meteorological phenomena in the South-eastern coastal area of China. They often trigger debris flows and other types of slope failure which cause significant economic damage and loss of life in an area with dense population and high economic activities. Accurate prediction of Typhoon-triggered debris flows and determination of the potential risk zones are crucial for risk management. However, little effort has been devoted to risk assessment by constructing the physical vulnerability curves in the typhoon-affected area. To cope with this deficiency, this paper presented a quantitative method to build up the physical vulnerability curves of buildings by modeling the debris flow intensity and building damage features. In this study, the Wangzhuangwu (WZW) watershed was selected, which was impacted by a debris flow induced by Typhoon Lekima on 10 August 2019. At first, detailed field investigation and interpretation of remote sensing imagery were carried out to analyze the geological characteristics, mechanism of the debris flow, and construct a database of building damage features. The 2019 debris flow initiation, movement and deposition processes were modelled based on the Soil Conservation Service-curve number (SCS-CN) approach and a two-dimensional finite model (FLO-2D). The reconstructed debris flow depth and extend were validated with observed information. Then, we proposed physical vulnerability curves for different type of building structures by combining the damage degree of buildings and the modelled debris flow intensity (flow depth and impact pressure). Based on the validated rheological parameters, the potential intensity of future debris flows was modelled considering different recurrence frequency of the triggering rainfall. Finally, the vulnerability index and economic risk of buildings to debris flow events with different frequencies were calculated using different vulnerability functions. The uncertainty of quantitative risk assessment was considered in the intensity indictor and building structure. The RC (reinforced concrete) frame building has stronger resistance than the non-RC frame building under same intensity of debris flow, and the vulnerability function using the impact pressure as the intensity indictor is more conservative than which using the flow depth. The proposed approach efficiently generated the physical vulnerability curves and debris flow risk map that can be used for effective disaster prevention in debris flow-prone areas.