W8 - Flow Domain Methods Flashcards
Name 2 Flow Domain Methods
➤ Nodal Network
➤ CFD
List the stages/parts to the➤ Nodal Network Method
➤ BC ➤ Node Definition ➤ Buoyancy Effects ➤ Component Fluid Model ➤ Solver
Nodal Network Method - BC (4)
➤ Pressure coef - Wind tunnel testing
values then used in simulation to enable calculations
➤ P around body with local weather parameters
➤ Understanding better how P evolves over the surface
➤ typicall 6 angles (22.5) apart
Nodal Network Method - Node Definition (3)
➤ Assumed Discrete, homogeneous fluid characterised by single Temp/StaticP/Height
➤ Guidance to get PC to connect elements
➤ Typically orthogonal
Nodal Network Method - Buoyancy Effects (2)
➤ DeltaP is determined using Bernoulli Eq.
➤ 1d steady in-compressible flow
Nodal Network Method - Component Fluid Model (2)
➤ Applying nodes and connect to the changing factor (ie chair temp change due to window oppening)
➤ connect to the M(dot) of flow entering environment etc
Nodal Network Method - Solver (5)
➤ Itteritive solution
➤ Assuming IC then solve all nodes
➤ Progressively add detail and re calculate m(dot) and deltaP for the matrix
➤ Itterate until solution become realistic and small residuals are present
➤ id m(dot not ok then change P and re-do steps above.
List the stages/parts to the ➤ CFD Method
➤ Disretise - orthogonal with bias to BLayers
➤ Conservation Eqn’s - HMAM, Turbulance Model, linear parameters
➤ Initial BC - rho, u, delta T at t=0
➤ Solver - standard in CFD. Semi implicit available to assist in convergance for specific senarios.
➤ Result Analysis - usually graphically.
main factors determining air infiltration?
discuss factors related to the outdoor environment, the indoor environment, the building and occupants
➤ Outdoor environment:
➤ ➤ wind speed, wind direction, surrounding buildings, location type (exposed, urban, city centre etc.) and outdoor air temperature.
➤ Indoor environment:
➤ ➤ indoor air temperature and over/under pressure relative to outside due to mechanical ventilation system.
➤ Building:
➤ ➤ shape of the building, amount, type and location of envelope openings and leakages, and amount, type and location of inside partition openings and leakages.
➤ Occupants:
➤ ➤ opening/closing of building openings (doors, windows etc.), control of vents and control of mechanical ventilation system(s).
How does fluid flow get modeled by NN?
➤ BC are based on temp & wind pressure coef. &surface inclination dimentions etc.
➤ System split into nodes where each node is a point where the values are know or unknow to be claculated at
➤ Bouyancy effects are included is high enough resolution is present
➤ Flow models determine flow from node to node using know parameters (deltaP etc)
➤ itteritive mas(dot) is obtaned to get adequate mass ballance for the node pressures.
NN - Explain the steps involved in solving the equations
➤ A guessed value of pressure is assigned to each internal (unknown) node.
➤ The mass flow rate between nodes is evaluated by applying the pressure difference to the empirical equations representing connected components.
➤ A mass residual (imbalance) is determined for each node.
Node pressure corrections are determined by establishing the system Jacobian matrix from knowledge of the nodal residuals, their rate of change with respect to nodal pressure change and the rate of change of the branch mass flow rates with branch pressure drop (all of which are known at the end of an iteration step).
➤ The pressure corrections are applied to the nodal pressures and the next iterative step commenced.
➤ The process is terminated when the nodal residuals fall below a prescribed value.
Steps to modelling by the computational fluid dynamics (CFD)
➤ Boundary conditions are represented by the temperature of bounding surfaces and mass & momentum inputs.
➤ The system is represented by the discretised Navier-Stokes equations when applied to a finite volume lattice representing the flow domain.
➤ Buoyancy is represented by the Boussinesq approximation whereby the fluid density is held constant and the effects of buoyancy are included within the momentum equations.
➤ A turbulence transport model is normally used whereby the influence of turbulence on the time averaged motion of the fluid may be determined.
➤ An iterative solution procedure is used to solve the energy, momentum and mass equations to give the distribution of pressure, temperature and velocity (and perhaps other parameters such as humidity and contaminant concentration).
State a problem to NN & CFD
➤ NN: best suited to the calculation of fluid flow in a network representing building leakage distribution or plant.
➤ CFD: best suited to an assessment of room air quality and distribution of comfort parameters.
Why is it important, in an energy systems modelling context, to re-establish the parameters of the turbulence model at each time step? In your answer identify these parameters.
➤ Because energy systems are often non-steady, low Reynolds Number flows occur in which the flow regime may be characterised as weakly turbulent and be moving from laminar flow through the transition zone to fully turbulent.
➤ The parameters are the turbulent kinetic energy (k) and its rate of dissipation (ε) throughout the domain; and the shear stress and convective heat transfer at surfaces.
