The terahertz spectra of gibberellin, forchlorfenuron and thidiazuron in the range of 0.2 to 2.0 THz were obtained by THz time-domain spectroscopy, based on which their absorption and refraction spectra were calculated. The results showed that all of them had obvious characteristic absorption peaks in the terahertz band. The terahertz absorption spectra of nineteen different substances were classified and tested by least squares support vector machine. Particle swarm optimization and genetic algorithm were applied for the parameter optimization of the least squares support vector machine, and the stability of the model was tested with noise addition. This study demonstrates the feasibility of terahertz spectroscopy for identification of plant growth regulators and will provide a new approach for the detection of plant growth regulators.