The tea tree is a type of horticultural crop and agricultural crop, its leaves are used to be an important ingredient in many beverages. However, the tea-leaf harvest is mainly accomplished by hands, which has an enormous impact on the yield of tea leaves. The wide application of tea-leaf mechanical harvester could improve the harvest efficiency of the tea-leaf, and then increase the harvest yield of the tea-leaf. To improve the harvest yield of the tea-leaf, the working parameters (structural and kinematic parameter) of the cutter on tea-leaf mechanical harvester need to be considered. And the cutting diagram of the cutter was used to optimize the working parameters. In this paper, the tooth top width, tooth root width, tooth depth, teeth space and cutting speed ratio were selected as working parameters of cutter that were considered to enhance the harvest efficiency. Taguchi method coupled with grey relational analysis (GRA) was used to determine the optimal combination of working parameters. Besides, the variance analysis method was performed to statistical analyse the influence of multiple parameters. The results showed that the key parameters of the variation of cutting, re-cutting and blank ratio was the teeth space. The optimal parameter combination was 4.5 mm tooth top width, 12.5 mm tooth root width, 26.0 mm tooth depth, 45.0 mm teeth space and 0.8 cutting speed ratio. Grey relational grade of the optimal parameter combination increased 0.367 compared to the grade with initial parameter combination. Based on the optimal parameter combination, the tea-leaf harvest flatness, efficiency and quality were completely in conformity with the requirements of the tea-leaf mechanical harvester. So that, the Taguchi method coupled with GRA is a practical one to optimize the working parameters of the cutter on the tea-leaf mechanical harvester.