In this work, the effect of various parameters involved in the design of a natural convection micropump using density-based topology optimization are studied. The motion of fluid in a natural convection micropump is caused by the differential heating of the walls. The problem is modeled as a conjugate heat transfer problem. The goal is to determine the fluid flow path within design domain to maximize the flow rate in the system. A steady-state, laminar, incompressible flow is assumed, with the coupled Navier-Stokes equations and convection-diffusion equations solved using the Boussinesq approximation. Various interpolation schemes for material properties in density-based topology optimization methods, such as RAMP, SIMP, and sigmoid functions, are tested and compared for the micropump problem. Use of projection schemes, along with the parameters used in them, are examined. The problem is analyzed for varying Grashof numbers, Prandtl numbers and channel width, with particular attention given to the challenges associated with higher Grashof numbers

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Parametric Study on Designing Natural Convection-Based Micropump for Maximizing Flow Rate Using Density-Based Topology Optimization

  • Om Venkata Bhargava Rama Reddy Karri,
  • Sourav Rakshit

摘要

In this work, the effect of various parameters involved in the design of a natural convection micropump using density-based topology optimization are studied. The motion of fluid in a natural convection micropump is caused by the differential heating of the walls. The problem is modeled as a conjugate heat transfer problem. The goal is to determine the fluid flow path within design domain to maximize the flow rate in the system. A steady-state, laminar, incompressible flow is assumed, with the coupled Navier-Stokes equations and convection-diffusion equations solved using the Boussinesq approximation. Various interpolation schemes for material properties in density-based topology optimization methods, such as RAMP, SIMP, and sigmoid functions, are tested and compared for the micropump problem. Use of projection schemes, along with the parameters used in them, are examined. The problem is analyzed for varying Grashof numbers, Prandtl numbers and channel width, with particular attention given to the challenges associated with higher Grashof numbers