This paper presents a novel self-sensing steel fiber-reinforced polymer composite bar (SFCB). The SFCB combines damage control, self-sensing, and structural reinforcement functions using distributed fiber optic sensing (DFOS) technology. By combining DFOS strains with theoretical and numerical models, a multilevel performance method for damage assessment is proposed from the perspectives of safety, suitability, and durability. Stiffness is a metric used to assess the complete service history of the reinforced concrete (RC) structure, which was used to define the damage variables. Initially, a basic correlation is created between the SFCB strain and several performance characteristics, such as moment, curvature, load, deflection, stiffness, and crack breadth, at characteristic points. The threshold values of damage variables for safety, serviceability, and durability were determined based on loading peak, mid-span deflection limits, and crack width limits corresponding to the damage variables. Then, a modified fiber damage model based on DFOS strain data is proposed to improve identification, quantification, and tracking for fiber damage. Finally, the reliability of the proposed theoretical and numerical models was verified by three-point flexural tests of SFCB-RC beams, and the test beams were analyzed using the proposed method. The results show that increasing the reinforcement ratio can lower the threshold at all levels and improve the ability of the flexural beams to control damage. This paper contributes to advancing the intelligence of RC structures and offers valuable insights for the design of intelligent RC structures.