Abstract: This study experimentally investigates the active control of a Mach 1.5 round jet using multiple steady radial microjets to enhance jet mixing. The control parameters include the expansion ratio of the main jet exit, the mass flow ratio of microjet to main jet, and the number of microjets. The effect of these parameters on supersonic jet mixing is revealed through experimental results. First, as the expansion ratio increases, the control performance of the supersonic jet gradually decreases from under-expanded to over-expanded conditions. Second, there is a complex nonlinear relationship between the mass flow ratio and the core length of the supersonic jet. Third, increasing the number of microjets does not enhance the supersonic jet mixing. Fourth, given the expansion ratio of the main jet is 2.5, 3, 3.67, 5, and 7, the optimal control parameters can reduce the length of the jet potential core by 62%, 60%, 53%, 46%, and 8%, respectively. This result indicates that the optimal control effect decreases with the increase of the expansion ratio of the main jet. Two prediction models are proposed to link the control parameters with the control targets based on the Taylor expansion and BP neural network algorithm. Both models exhibit better prediction ability near the optimal value. When the prediction results are far from the optimal value, the length of the core region of the supersonic jet predicted by the BP neural network is more accurate, with a prediction error that is only 6% of that based on the Taylor expansion method.