The invention relates to the technical field of core backbone network search of a power system, and particularly relates to an elastic guaranteed bottom network rack search modeling method based on a0-1 programming model, and the specific steps include: S1: establishing an objective function of a sum of a line number and a subnet search weight as reference factors; S2: establishing an equality constraint of the optimization model, establishing a branch power constraint and a branch current constraint based on the power flow constraint; S3: establishing an inequality constraint of the optimization model, including a voltage constraint, a generator power limiting constraint and a line power limiting constraint; S4: solving the model, so that the length of the final scheme grid line is smalland the electrical interface index is large. The method can solve the problem of the uncertainty of the traditional intelligent algorithm result and the poor robustness of the algorithm, can obtain the guaranteed bottom network rack with the fewest lines, and can obtain the situation that the generator is put into operation. The method has high flexibility and strong scalability and effectively improves the efficiency and practicability of the guaranteed bottom net rack.