A hybrid system robust diagnostic observer and fault tolerant control algorithm based on bond graph(BG) modeling are studied. On the basis of BG theory, there is the problem of parameter uncertainty for hybrid systems. First, the design of robust observer system, a bond graph in linear fractional transformation (BG-LFT) and proportional integral (PI) observer are combined to make robust fault diagnosis and fault estimation of the system. The observer can track the dynamic behavior of system variables in real time and effectively reduce false alarm rate and false negative rate to improve the detection effect. Then, an active fault tolerant controller (AFTC) based on state and fault estimation is proposed to ensure stable operation in the event of a system failure. Finally, the validity and feasibility of the method are verified by simulation.