Abstract: Axial flow pumps can serve as the core equipment for waterjet propulsion and water diversion. However, the flow separation-induced stall can cause the hump-shaped characteristics in the head-flow curves under low-flow conditions. Such unstable flow can pose significant operational risks to the pump. In this study, a flow control strategy was proposed with the endwall injection to suppress the rotating stall in an axial flow pump. Experimental studies were conducted in terms of the steady-state and pulsed endwall injection for the axial flow pump. A systematic investigation was made on the influence of the injected flow rate, injector numbers, and pulsed injection frequency on the performance of the axial flow pump. The injected flow rate was normalized against the design flow rate of the axial flow pump. The injected flow rate was tested in the range from 2.6% to 10.48% in the experiment. The number of the injectors varied between 1 and 6. The frequency of the pulsed injection covered a range of 0.05 to 0.5 Hz. The pressure pulsation test revealed that the endwall injection improved the stability of the axial flow pump. A test rig of the closed-loop axial flow pump was carried out at the State Key Laboratory of Water Engineering Ecology and Environment in Arid Area, Xi’an University of Technology, Shanxi Province, China. The test pump was designed with a flow rate of 168 m³/h, a head of 3.12 m, and a rotational speed of 1 800 r/min. The results indicate that the injected mass flow rate, injector number, and pulsed injection frequency shared a significant impact on the performance of the axial flow pump. In the steady-state endwall injection, the stall margin of the axial flow pump increased with the increase of the injected mass flow rate and the number of injectors. The pump efficiency exhibited negligible sensitivity to the injected mass flow rate when the injected mass flow rate was below 5% of the design. The measurable efficiency was enhanced significantly when the injected mass flow rate exceeded 8.4%. Parametric studies revealed that there was a non-monotonic relationship between the injector number and pump efficiency, indicating peaking at four injectors. The optimal injector number was four, considering both the stall margin and the efficiency under the design point. The pump efficiency and the stall margin were improved by 1.83% and 39.4% at an injection flow rate of 5.6%. Experimental studies were conducted on the pulsed endwall injection using six injectors. The greater stability was achieved by reducing the injected flow in the pulsed endwall injection, compared with the steady-state one. The frequency of the injection significantly impacted the pump performance. The stall margin generally increased as the injection frequency rose. However, there was little influence of the increasing injection frequency on the pump stability when the injection frequency exceeded 0.25 Hz. The injection frequency was elevated to significantly improve the pump efficiency. Once the critical threshold of the injection frequency was beyond 0.17 Hz, there was a negligible influence of the injection frequency on the pump efficiency. Both the pump efficiency and stall margin of the axial flow pump were balanced at the injection frequency over 0.25 Hz. Once the injection frequency was 0.5 Hz, the superior stability was enhanced in the pulsed endwall injection, compared with the steady-state one. The injection flow rate of 4.2% and the efficiency under the design point were also improved by 53.9% and 3.52% in the stall margin of the axial flow pump, respectively. Mechanistic analysis revealed that the head declined to initiate with the rotating stall onset, where the propagation velocity of the stall cell reached 72% of the impeller rotational speed. Endwall injection effectively delayed the stall inception caused by the rotating stall. The optimal parameters fully suppressed the rotating stall. Thereby, the operational stability was improved significantly to eliminate the associated hump-shaped characteristics in the head-flow curves. In conclusion, the endwall injection demonstrated the dual-benefit capability to maintain the pump efficiency under the design point. While the operational stability was enhanced significantly. A critical flow control strategy can be expected for the high-efficiency and high-stability operation in the axial flow pumps.