A general purpose CFD code named DPC has been employed in the multidimensional simulation of the hybrid stores.

DPC is a library written by the CTTC for the resolution of combined heat and mass trans­fer problems by means of computational fluid dynamics. The Navier-Stokes equations are solved together with the energy equation and the species equations (when a mixture is modelled) using finite volume techniques. The problem domain can be discretised in one or several blocks [1] using structured staggered or co-located grids [11]. Turbulence is mod­elled by means of low-Reynolds number two equation models [11]. Different methods can be used for the modelling of the radiative heat transfer: radiosity-irradiosity method, discrete or­dinate method [4]. Solid/liquid phase change is modelled by means of enthalpy-like methods [5].

Continuity and momentum equations are solved with a coupled multigrid algebraic solver, or using pressure based SIMPLE-like methods and solvers multigrid [10, 7]. The scalar variables transport equations (energy, species, …) are segregatelly solved by means of multigrid solvers. DPC allows parallel processing when the multiblock technique is used [2].

Some of the most illustrative results obtained with DPC concerning multidimensional nu­merical simulation of storage devices and solid/liquid phase change phenomena can be found in [12, 2, 3, 6, 5].

Prediction models

Simplified models based on global or one-dimensional mass and energy analysis, have been developed in order to be used in the store thermal performance description (ENV 12977-3), and in long term thermal solar systems simulation codes.

The well-known multinode model [8] has been adjusted to numerically predict the thermal behaviour of hybrid latent/sensible stores with PCMs. The PCM modules have been mod — elized as internal tank elements characterised by an overall heat transfer coefficient. The PCM modules at each node are assumed to be at the same temperature, and their energy balance equation take into account the energy stored by latent heat.

We have considered of interest to look into this model in detail. Reviewing multinode mathematical formulation, for each i-t/г tank node, energy balance can be written as follows:

where, t is time, jV is the number of nodes, p is the fluid store density, cp is the specific heat, У, is the effective store volume, Ті is the temperature of the i-t/г node, and mheat are the temperature and the mass flow rate of the fluid from the heat source, Tload and mioaii are the temperature and the mass flow rate of the fluid from the load loop, Tem, is the environment temperature, TpemA the temperature of the PCM modules int the i-t/г node, (UA)em, is the overall heat loss coefficient, is the overall heat transfer coefficient

between the sensible heat store and the PCM modules, and Д are the direction controllers (1 if the fluid enters node i, 0 otherwise), Ae// is the effective thermal conductivity, and S’ is the store cross section.

On the other hand, an energy balance about the i-t/г PCM modules node can be written as:

Ррет I Cp, pcm ^

where, rpcTO is the effective ratio of PCM modules volume in the overall store volume, L is the PCM latent heat and FpcmA is the fraction of PCM in liquid state (i. e. FpcmA = 1 means that the PCM in the i-t/г node is liquid while, FpcmA = 0 means that it is in solid state).

Equations 1 and 2, assume that overall heat transfer coefficients are the same for each node of the tank. This hypothesis implies that PCM modules are located homogeneously throughout the tank.

In a numerical simulation, equations 1 and 2 are solved iteratively at each time increment. The fraction of liquid in the PCM modules is evaluated from equation 2 fixing ТрстЛ to the PCM melting temperature Tpemm and isolating FpcmA.

Illustrative results

As an illustrative example of the use of the numerical infrastructure commented above, preliminary results obtained in the design/optimisation studies of hybrid latent/sensible stores are hereafter presented.

Hybrid stores are numerically simulated by means of the CFD code. CFD simulations are used as a virtual design tool. The detailed numerical data obtained is then employed to identify store parameters on the basis of EN12977-3. In this task, the prediction codes explained above are used. The thermal performance of the store is determined by the following parameters: the effective store volume У,, the effective PCM ratio rpcm, the overall heat transfer coefficients (UA),,,pcm and (UA)env, and the effective thermal conductivity Ae//.