The spanwise distribution of impeller exit circulation (SDIEC) has a significant effect on the impeller performance, therefore, there is a need for its consideration in the optimization design of mixed-flow pumps. In this study, a combination optimization system, including a 3D inverse design method (IDM), computational fluid dynamics (CFD), Latin hypercube sampling (LHS) method, response surface model (RSM), and non-dominated sorting genetic algorithm (NSGA-Ⅱ) was used to improve the performance of the mixed-flow pump after considering the effect of SDIEC on the performance of the impeller. The CFD results confirm the accuracy and credibility of the optimization results because of the good agreement the CFD results established with the experimental measurements. Compared with the original impeller, the pump efficiency of the preferred impeller at 0.8Q_des, 1.0Q_des, and 1.2Q_des improved by 0.63%, 3.39%, and 3.77% respectively. The low-pressure region on the blade surface reduced by 96.92% while the pump head difference was less than 1.84% at the design point. In addition, a comparison of the flow field of the preferred impeller and the original impeller revealed the effect of SDIEC on mixed-flow pump performance improvement and flow mechanism.

When considering the interaction between the impeller and diffuser, it is necessary to provide logical and systematic guidance for their matching optimization. In this study, the goal was to develop a comprehensive matching optimization strategy to optimize the impeller and diffuser of a mixed flow pump. Some useful tools and methods, such as the inverse design method, computational fluid dynamics (CFD), design of experiment, surrogate model, and optimization algorithm, were used. The matching optimization process was divided into two steps. In the first step, only the impeller was optimized. Thereafter, CFD analysis was performed on the optimized impeller to get the circulation and flow field distribution at the outlet of the impeller. In the second step of optimization, the flow field and circulation distribution at the inlet of the diffuser were set to be the same as the optimized impeller outlet. The results show that the matching optimization strategy proposed in this study is effective and can overcome the shortcomings of single-component optimization, thereby further improving the overall optimization effect. Compared with the baseline model, the pump efficiency of the optimized model at 1.2Q_des, 1.0Q_des, and 0.8Q_des is increased by 6.47%, 3.68%, and 0.82%, respectively.

In this study, the positive and negative power-off process of a bi-directional full-flow pump is investigated by model tests and numerical simulations, and the results show that under steady conditions, the head of the bi-directional full-flow pump in the positive direction is larger than in the negative direction. The positive power-off process of the bi-directional full-flow pump is slower than the negative power-off process. The clearance backflow rate of the bi-directional full-flow pump continues to drop during the power-off process until it remains stable under the runaway condition. The clearance backflow causes the vortex at the impeller inlet and the size of the vortex gradually decreases during the power-off process. Compared to the positive power-off process, the backflow vortex ratio is greater, and therefore, the flow pattern is poorer during the negative power-off process. The entropy production rate in the negative power-off process is significantly greater than that in the positive power-off process. The pumping condition has the largest hydraulic loss in the whole power-off process. The runaway rotational speed in the positive power-off process is higher than that in the negative power-off process, and the axial force in the positive runaway condition is 1.40 times greater than in the negative runaway condition. During the power-off process, the vibration and pressure pulsation in the negative operation is significantly greater than that in the positive operation, and the acceleration of the impeller vibration from large to small is the Y (vertical) direction, Z (axial) direction, and X (horizontal) direction. The research in this paper can provide an important reference for the design and operation of the bi-directional full-flow pump.

Loading distribution significantly affects the flow field in the impeller channel, but research on the effects of different loading distribution forms on the performance of mixed-flow pumps is relatively scarce. In this study, three typical mixed-flow pump models, namely front-loading model (FLM), middle-loading model (MLM), and back-loading model (BLM), were designed by changing the loading distribution at the shroud using the inverse design method. The energy losses within the three models were analyzed quantitatively and visually using the entropy generation method. The results show that using front-loading, middle-loading, and back-loading in the design of mixed-flow pumps improves efficiency under common, low, and high flow rate conditions, respectively; in particular, the efficiencies of the FLM, MLM, and BLM under design conditions were 88.72%, 87.82%, and 87.01%, respectively. The entropy generation analysis indicates that the loading distribution has a minor influence on the energy loss within the impeller, but its influence on the energy loss in the downstream components of the impeller can exceed 30%. Internal flow analysis shows that the effect of loading distribution on the relative velocity and adverse pressure gradient at the impeller outlet is the primary reason for the change in the performance of mixed-flow pumps.

The effectiveness of the inverse design method has been widely proven in previous studies, but research on the effect of different vortex designs on the performance of mixed flow pumps is relatively scarce. In this paper, the performances of models I1 and I2, which were designed with free and compound vortex designs, respectively, are compared to study the effect of different vortex designs on the energy characteristics and operational stability of mixed flow pumps. The results show that the efficiency of the compound vortex design at 0.8Qdes, 1.0Qdes, and 1.2Qdes is improved by 0.54%, 0.95%, and 5.91%, respectively, compared to that of the free vortex design, and the velocity and pressure pulsations under the design conditions are also significantly reduced. The internal flow analysis shows that the increased efficiency in the compound vortex design is mainly related to the reduction in the local entropy production from the hub to the mid-span and the wall entropy production from the mid-span to the shroud within the diffuser due to the improvement in the jet-wake structure near the hub. The increased operational stability is mainly related to the suppression of low-momentum fluid aggregation and H-S secondary flow caused by the increase in axial velocity near the hub and the spanwise uniformity of the total pressure.

This study experimentally investigated the transient flow characteristics of single and twin parallel circular impinging jets using Particle Image Velocimetry (PIV) and Proper Orthogonal Decomposition (POD). High spatiotemporal resolution data were collected to analyze and visualize the distribution and interactions of multi-scale vortex structures. The experimental data were evaluated for PIV uncertainty and statistical convergence. The flow field distribution of the fountain that may occur in the twin impinging jets was also discussed. The results revealed that, for the oblique single jet, high-energy large-scale structures primarily concentrate in downhill regions and gradually decompose into smaller-scale structures. In the combination of twin jets, vortices in the shear layers merge in an alternating pattern. The frequencies of the inner and outer shear layers are sensitive to jet spacing with different characteristic interaction patterns repeatedly observed based on varying spacings. Three distinct cases are defined based on the interaction of the twin jets with the impingement plate. Additionally, three distinct flow field structures of fountain are defined to facilitate, indicating the degree of combination as well. The velocity profiles of the fountain are similar and adhere to a Gaussian distribution. Furthermore, a semi-empirical equation was developed to describe the centerline velocity of the fountain.