Search

To address the challenges of multiple design variables, long evaluation times, and poor global search capability of traditional surrogate-assisted algorithms that rely on the complete substitution of accurate function evaluations in the turbine aerodynamic optimization process, a pre-screening surrogate-assisted elitist preservation genetic algorithm (pre-SAEGA) optimizer is proposed. Pre-screening strategy in the pre-SAEGA can screen samples instead of directly estimating them, thereby reducing expensive evaluations in each generation. The directly manipulated free-form deformation (DFFD) method is applied to parameterized multi-stage axial turbines, and multi-degree-of-freedom flexible control is realized. Combining the pre-SAEGA with the DFFD method, a data-driven multi-stage axial turbine optimization platform is established. A two-stage axial turbine is the research object, and 44 design variables are selected for the combined optimization design of flow path and blade rows. The results show that the isentropic efficiency and flow rate improve by 1.33 % and 1.81 % respectively, and the pressure ratio decreases by 0.47 % at the turbine design point. The presented optimization platform not only improves the aerodynamic optimization effect but also significantly reduces the number of design variables and real evaluation samples, making it suitable for solving multi-stage turbine optimization problems with multiple degrees of freedom.

The intercooling recuperated cycle (ICR) is commonly employed in marine gas turbines to enhance thermal efficiency. However, the addition of an intercooler may lead to the increased dimension and structural complexity of marine ICR gas turbines. To address this issue, we propose an improved configuration of the intermediate pressure compressor – an axial-centrifugal combined compressor (ACC) with high inlet hub-tip ratio and high flow rate. The aerodynamic performance of the ACC at multi-operating points is optimized via an improved free-form deformation method for the parametric modeling of the flow paths and blades of turbomachinery. The result indicates that the number of stages decreases from 6 to 3 and the axial length is reduced by 38.3% after modification. The adiabatic efficiency of the optimized ACC at the design and low-speed operating points is improved by 1.18% and 6.48% respectively. Additionally, the ACC redesigned scheme can reduce the axial length, maximize the flow path space utilization, enhance the stage load capacity, and significantly improve the low-speed performance. This provides a reference for developing advanced marine ICR gas turbines with high power and low fuel consumption.

Summary: The eye is a complex organ consisting of multiple compartments with unique and specialized properties, and small disturbances in one eye region can result in impaired vision and blindness. Although there have been advancements in ocular research, the hierarchical molecular network in region-wide resolution, indicating the division of labor and crosstalk among different eye regions, is not yet comprehensively illuminated. Here, we present an atlas of region-resolved proteome and lipidome of mouse eye. Multiphoton microscopy-guided laser microdissection combined with in-depth label-free proteomics identifies 13,536 proteins across various mouse eye regions. Further integrative analysis of spectral imaging, label-free proteome, and imaging mass spectrometry of the lipidome and phosphoproteome reveals distinctive molecular features, including proteins and lipids of various anatomical mouse eye regions. These deposited datasets and our open proteome server integrating all information provide a valuable resource for future functional and mechanistic studies of mouse eye and ocular disease.

The compressor global geometric model including the endwall and the blade is parameterized based on a directly manipulated free-form deformation (DFFD) approach. A meridional flow streamline curvature method for solving inverse and forward problems is applied to the reverse design and performance analysis of the compressor. Combined with an improved Powell search algorithm and the multi-island genetic algorithm, a platform for the fast aerodynamic optimal design of the multistage axial compressor is established. The integrated automatic design and performance optimization of a nine-stage axial compressor in a gas turbine are performed. Meanwhile, the inlet transition section, strut and outlet expansion section of the compressor are retrofitted and optimized considering the connection with the front and rear of the test section. The result indicates that the platform can automatically and effectively realize the process of the fast design, analysis and optimization for the multistage axial compressor. Finally, the total pressure ratio and adiabatic efficiency at the design point and the surge margin at the design rotate speed are 5.057, 88.16% and 17.25%, which are 0.377%, 1.043% and 1.173% relatively higher than those before the 3D DFFD optimization respectively.