This model implements an ice tank model whose performance is computed based on performance curves.
The model is based on the implementation of Guowen et al., 2020 and similar to the detailed EnergyPlus ice tank model ThermalStorage:Ice:Detailed.
The governing equations are as follows:
The mass of ice in the storage mice is calculated as
d SOC/dt = Q̇/(Hf mice,max)
mice = SOC mice,max
where SOC is state of charge, Q̇ is the heat transfer rate of the ice tank, positive for charging and negative for discharging, Hf is the fusion of heat of ice and mice,max is the nominal mass of ice in the storage tank.
The heat transfer rate of the ice tank Q̇ is computed using
Q̇ = Qsto,nom q*,
where Qsto,nom is the storage capacity and q* is a normalized heat flow rate. The storage capacity is
Qsto,nom = Hf mice,max,
where Hf is the latent heat of fusion of ice and mice,max is the maximum ice storage capacity.
The normalized heat flow rate is computed using performance curves for charging (freezing) or discharging (melting). For charging, the heat transfer rate q* between the chilled water and the ice in the thermal storage tank is calculated using
q* Δt = C1 + C2x + C3 x2 + [C4 + C5x + C6 x2]ΔTlmtd*
where Δt is the time step of the data samples used for the curve fitting, C1-6 are the curve fit coefficients, x is the fraction of charging, also known as the state-of-charge, and Tlmtd* is the normalized LMTD calculated using Buildings.Fluid.Storage.Ice.BaseClasses.calculateLMTDStar. Similarly, for discharging, the heat transfer rate q* between the chilled water and the ice in the thermal storage tank is
- q* Δt = D1 + D2(1-x) + D3 (1-x)2 + [D4 + D5(1-x) + D6 (1-x)2]ΔTlmtd*
where Δt is the time step of the data samples used for the curve fitting, D1-6 are the curve fit coefficients.
The normalized LMTD ΔTlmtd* uses a nominal temperature difference of 10 Kelvin. This value must be used when obtaining the curve fit coefficients.
The log mean temperature difference is calculated using
ΔTlmtd* = ΔTlmtd/Tnom
ΔTlmtd = (Tin - Tout)/ln((Tin - Tfre)/(Tout - Tfre))
where Tin is the inlet temperature, Tout is the outlet temperature, Tfre is the freezing temperature and Tnom is a nominal temperature difference of 10 Kelvin.
This model requires the fluid to flow from port_a
to port_b
.
Otherwise, the simulation stops with an error.
Strand, R.K. 1992. “Indirect Ice Storage System Simulation,” M.S. Thesis, Department of Mechanical and Industrial Engineering, University of Illinois at Urbana-Champaign.
Guowen Li, Yangyang Fu, Amanda Pertzborn, Jin Wen and Zheng O'Neill. An Ice Storage Tank Modelica Model: Implementation and Validation. Modelica Conferences. 2021. doi:10.3384/ecp21181177.