Abstract: Low-pressure wells still produce at high rates under very low pressure. The formation pressure coefficient of the ancient gas wells in the Changqing field is around 0.4, yet they still maintain economically viable output. Leak control during well workover is key to ensuring stable and high production from these wells. By collecting data from over 50 ancient low-pressure gas wells in the Changqing oilfield, with more than 80 geological, engineering, and material characteristic parameters, and extracting 17 sets of raw data, we introduced big data evaluation concepts to determine the weight coefficients of different factors under multifactor conditions. This guided the matching of plugging material performance under the main controlling factors for leakage in low-pressure ancient wells, leading to the development of three types of well killing fluid systems: fuzzy ball fluid, fuzzy ball fluid compound system, etc. From 50 formulations, we selected 5 sets of optimized formulas and evaluated their pressure-bearing capacity and self-degradation ability changes over 7-30 days. After 7 days, the pressure-bearing capacity decreased by 1.96% to 4.10% due to material self-degradation, and after 30 days, it decreased from 18.36% to 23.23%, with permeability recovery values around 95%. Field tests on wells G37-A, G31-B, and others, involving temporary plugging operations due to string replacement in three ancient low-pressure gas wells, showed that the pressure-bearing capacities were 23.93, 19.89, and 26.53 MPa, respectively, with water leak rates all below 2 m³/h. Combining geological, engineering, and material characteristic information with big data optimization concepts can effectively improve the efficiency of single-well plugging, providing a reference method for other leaking wells due to drilling, completion, etc.