The problem of high false alarm rate when inductive monitoring stystem (IMS) was applied to monitor the system during trials or flights was analyzed

and an improved IMS method was proposed. The improved algorithm processed outliers that may occur during the test

reducing interference evoked by outliers of test data. The experiment was conducted by taking the trial data of a certain type of liquid-propellant rocket engine as an example. Simulation was performed and the result verified that the original IMS had a high false alarm rate

yet the improved IMS can decrease false alarm rate from 37.5% to 0 while ensuring timeliness

reliability and accuracy of fault detection.