SIMULATION INTEGRATED OPTIMIZATION-BASED FEATURE DESIGN OF HYDRAULIC VANE MACHINE

<p dir="ltr">This thesis presents a simulation-integrated, feasibility-guided optimization framework</p><p dir="ltr">for the performance enhancement of a dual-lip balanced vane pump. Traditional hydraulic</p><p dir="ltr">pump design relies heavily on intuition-driven prototyping and manual iteration, which can</p><p dir="ltr">be time-consuming and restrict exploration of non-intuitive design configurations. To address</p><p dir="ltr">this, the study introduces a computationally efficient methodology that integrates parametric</p><p dir="ltr">geometry definition, staged design space filtering, and multi-objective optimization using</p><p dir="ltr">evolutionary algorithms.</p><p dir="ltr">The proposed design framework begins with a structured feasibility analysis of the camring</p><p dir="ltr">profile, employing constraint-based filtering over radial and angular design parameters</p><p dir="ltr">to eliminate infeasible candidates early in the workflow. Three iterations of increasing complexity</p><p dir="ltr">were used to isolate a viable subspace of the full design domain, significantly reducing</p><p dir="ltr">simulation costs while preserving design diversity. A similar constraint-based bound refinement</p><p dir="ltr">was implemented for the delivery groove length, allowing meaningful variation within</p><p dir="ltr">operational limits.</p><p dir="ltr">Two optimization case studies are presented. The first investigates performance trade-offs</p><p dir="ltr">under a single high-pressure, medium-speed operating condition, demonstrating improvements</p><p dir="ltr">in volumetric efficiency and flow ripple through simultaneous optimization of the</p><p dir="ltr">cam-ring and groove geometries. The second case generalizes the design for multiple operating</p><p dir="ltr">conditions, producing cam-ring profiles with consistent improvements across a range</p><p dir="ltr">of load cases. Selected Pareto-optimal designs were validated using high-fidelity full-model</p><p dir="ltr">simulations, showing favorable alignment with optimization predictions. Through this study,</p><p dir="ltr">a decrease in volumetric loses of 8% as well as a decrease in flow disturbance of 25.7% was</p><p dir="ltr">achieved for an existing industry standard design.</p><p dir="ltr">Overall, the study demonstrates the effectiveness of combining geometric parameterization,</p><p dir="ltr">feasibility-guided sampling, and multi-objective optimization in enabling high-performance,</p><p dir="ltr">low-cost design workflows for positive displacement machines.</p>