Lifting coarse mineral particles from thousands of meters of seabed to the supportive vessSel is a crucial part of exploitation of deep-sea mineral resources, and the vertical transportation part is a key component of the deep-sea mining system. Three typical vertical transportation schemes are discussed and compared from the aspects of working mechanism, structural scheme, transportation capacity, system efficiency and implementation feasibility in the context of commercial exploitation of deep-sea polymetallic nodules. The conclusion is that the hydraulic pipeline lifting system with a centrifugal pump is a comprehensive scheme. Furthermore, the basic composition and function of the hydraulic lifting system are introduced, and the transportation performance indicators and technical requirements under commercial mining conditions are analyzed. As the key equipment of the lifting system, the structural characteristics, design theory, transportation performance analysis methods and research progress of the lifting pump are described. A 1000 m sea trial was carried out. The lifting system, the tests of the centrifugal pump and the sea trial are introduced.

In order to improve the hydraulic performance of a deep-sea mining pump, this research proposed a multi-objective optimization strategy based on the computational fluid dynamics (CFD) numerical simulation, genetic algorithm back propagation (GABP) neural network, and non-dominated sorting genetic algorithm-III (NSGA-III). Significance analysis of the impeller and diffuser parameters was conducted using the Plackett–Burman experiment to filter out the design variables. The optimum Latin hypercube sampling method was used to produce sixty sample cases. The GABP neural network was then utilized to establish an approximate model between the pump’s hydraulic performance and design variables. Finally, the NSGA-III was utilized to solve the approximation model to determine the optimum parameters for the impeller and diffuser. The results demonstrate that the GABP neural network can accurately forecast the deep-sea mining pump’s hydraulic performance, and the NSGA-III global optimization is effective. On the rated clear water conditions, the optimized pump has a 14.65% decrease in shaft power and a 6.04% increase in efficiency while still meeting the design requirements for the head. Under rated solid-liquid two-phase flow conditions, the head still meets the design requirements, the shaft power is decreased by 15.64%, and the efficiency is increased by 6.00%.