Abstract: The Ni foam/polyurethane(PU) co-continuous composites (100PPI 0.8/PU, 50PPI 2.3/PU, 50PPI 1.3/PU, 50PPI 0.5/PU, 25PPI 1.7/PU and 25PPI 0.8/PU) were prepared by vacuum feeding method. The different co-continuous composites were named after pore size, volume density of the Ni foams with PU. According to ASTM standard G73-98, the high-speed droplet erosion device was purchased and modified. The plunger pump fed water into the pipeline, and the nozzle obtained liquid flow of different flow rates by adjusting the diverter valve. When the liquid flow was ejected from the nozzle, it was constrained by two parallel walls, and it scattered in a fan shape. The three test conditions of pipeline pressure 6.9 MPa, 8.3 MPa, 10.3 MPa for 30 min were chosen to conduct droplet erosion tests on the composites and pure PU. The PIV (particle image velocimetry) system was used to measure the velocity and size of the droplets. The PIV system was mainly composed of a laser generator, a camera, a synchronization controller and a computer. The laser generator emitted two slender, vertical laser beams, hitting the droplets below the nozzle. The synchronization controller enabled the camera to capture images and get two photos. Knowing the time interval of the two lasers and comparing the overall movement distance of the particles in the two photos, the velocity of the droplets could be calculated by the software Insight 4G. At the same time, the software Insight 4G recognized the droplet particles in the photo and could measure the droplet size.　Under 6.9 MPa pipeline pressure, the droplet velocity was about 70 m/s, the flow rate was 5.870 L/min, and the average droplet diameter was 480 μm; under 8.3 MPa pipeline pressure, the droplet velocity was about 90 m/s, the flow rate was 6.264 L/min, and the average droplet diameter was 512 μm; under 10.3 MPa pipeline pressure, the droplet velocity was about 115 m/s, the flow rate was 6.814 L/min, and the average droplet diameter was 543 μm. The higher the pipeline pressure was, the higher the flow rate was, the higher the droplet velocity was, and the higher percentage of large droplets in the total number of droplets was. When the pipeline pressure increased, the speed of the droplets increased, the size of the droplets increased, the total number of droplets per unit time increased, and the impact frequency of the droplets also increased. Therefore, as the pipeline pressure increased, the impact energy of the droplets increased significantly.　Under the condition of 6.9 MPa-30 min, the mass loss of a few composites was less than that of pure PU; under the condition of 8.3 MPa-30 min, the mass loss of most composites was less than that of pure PU; under the condition of 10.3 MPa-30 min, the mass loss of all composites was less than pure PU. The composites exhibited better droplet erosion resistance under higher impact energy. Under the condition of 10.3 MPa-30 min, the erosion craters of the composite 100PPI 0.8/PU were small and shallow, a small piece of resin was peeled off from the surface, and the metal arris were damaged slightly. The erosion craters of the 50PPI composite were large and deep, the resin phase was peeled off from the metal skeleton, and the metal arris had a small amount of plastic deformation and fracture. The erosion craters of the 25PPI composite were very deep with large pieces of resin peeling off from the metal arris, but the metal arris were damaged slightly. In general, the metal arris damage of all composites was relatively small, and resin peeling from the metal arris was the main source of damage for all composites. The erosion craters of pure PU were larger and deeper than that of all the composites.　When the droplets impacted, pure PU cracked under the action of water hammer pressure and stress wave. The crack widened and deepened under the action of lateral jet and hydraulic penetration. As impact energy of the droplets increased, the composites' resistance to high-speed droplet erosion was significantly better than pure PU. The Ni foam metal skeleton could block the impact of droplets, played a good protective role for the resin under the metal arris, which showed the shadow protection effect. The droplets of the forward and lateral jets rebounded when hitting the metal arris, and the rebounded droplets blocked the droplets that arrive later, which showed the carpet protective effect. The smaller the pore size of the Ni foam, the better the erosion resistance of the composite. The metal skeleton in the composite had strong resistance to the effects of water hammer pressure, stress wave, lateral jet and hydraulic penetration, and had shadow protection and carpet protection effects on the resin phase, while the resin phase can provide support effects on Ni foam absorbing the impact energy of droplets. The synergistic effect of the two phases improved the droplet erosion resistance of the composite. The composite 100PPI 0.8/PU exhibited the best droplet erosion resistance due to its dense metal skeleton. Application of Ni foam with small pore size and small volume density was beneficial to droplet erosion resistance of the Ni foam/PU co-continuous composites.