In order to achieve high-precision detection of controller's workload, firstly, pre-experiments was designed to select representative and distinguishable control scene scenarios, based on which, experiments were designed to obtain the (electroencephalogram) EEG data and workload data of controllers in different work scenarios, and the validity of the selected experimental scenarios was verified by the Shapiro-Wilk (S-W) test. Subsequently, a series of EEG features were proposed and their correlation with workload was analysed. The results show that the correlation between the gamma wave-related indexes β + θ + α + γ and workload is the highest, which can be used for the workload detection of controllers. Finally, the "leave-one-out cross-validation method" was used to establish a decision tree regression model to detect the workload of controllers. The detection accuracy is up to 0.979 ~ 0.996, demonstrating that the power spectrum features associated with gamma waves can be used for controller workload detection.