<p>This paper proposes a vibration monitoring and health status recognition model for machine tool electric spindles to optimize efficiency. Laser Doppler technology is used to obtain vibration signals, which are analyzed through wavelet transform. A health status detection vector is calculated and compared with a standard vector in a database using Euclidean distance. The results showed that when spindle speed was below 5000 rpm, the vibration intensity growth rate was faster, while it slowed down above 5000 rpm. At 5000 rpm, the predicted and measured values of vibration intensity matched at 0.065. At 12,500 rpm, the values were 0.073 and 0.076, respectively. The test results indicated that the spindle was unbalanced when the test sequence number was between 24 and 40, and the bearing was faulty when the test sequence number was between 2 and 16. This method efficiently identifies fault data types and offers technical insights for vibration monitoring and health status recognition of machine tool electric spindles.</p>