The invention discloses a method for improving the blade vibration sampling rate under a single sensor, and the method comprises the steps: firstly carrying out the abnormal burr removal of an original
analog signal, and guaranteeing the
signal stability; then, n rising edge events are recognized by setting an initial threshold value, and every n blades are divided into a
processing unit group; fitting each group of data to obtain a smooth curve, and extracting the maximum value in the
voltage minimum values and the minimum value in the maximum values to form a dynamic trigger interval; dividing a plurality of
equidistant sub-thresholds in the interval, and executing upper and lower edge triggering on the
signal under each sub-threshold to obtain a multi-resolution triggering sampling point; and finally, performing
time alignment on the multi-threshold trigger points according to the blade number, and reconstructing a blade vibration displacement
signal, thereby improving the calculation accuracy of the blade vibration displacement. The method is suitable for on-line deployment of an FPGA platform, remarkably improves the resolution capability of the tiny
vibration response on the premise of not improving the hardware sampling rate, and has the advantages of simple structure, high robustness, high
engineering realizability and the like.