Add new Complementarity formualtion for VLE with cubic EoSs

docs/explanations/components/property_package/general/pe/smooth_flash.rst

@@ -1,9 +1,12 @@
-Smooth Vapor-Liquid Equilibrium Formulation (``smooth_VLE``)
-============================================================
+Smooth Vapor-Liquid Equilibrium Formulation (SmoothVLE)
+=======================================================
 
 .. contents:: Contents 
     :depth: 2
 
+.. note::
+  For property packages using cubic Equations of State, there is an alternative :ref:`CubicComplementarityVLE <explanations/components/property_package/general/pe/smooth_vle2:Cubic Smooth Vapor-Liquid Equilibrium Formulation (CubicComplementarityVLE)>` class that may give better performance.
+
 Source
 ------
 

docs/explanations/components/property_package/general/pe/smooth_vle2.rst

@@ -0,0 +1,67 @@
+Cubic Smooth Vapor-Liquid Equilibrium Formulation (CubicComplementarityVLE)
+===========================================================================
+
+.. contents:: Contents 
+    :depth: 2
+
+.. note::
+  This formulation for vapor-liquid equilibrium is only valid if using a cubic Equation of State. For other equations of state, use :ref:`SmoothVLE <explanations/components/property_package/general/pe/smooth_flash:Smooth Vapor-Liquid Equilibrium Formulation (SmoothVLE)>`.
+
+Source
+------
+
+Dabadghao, V., Ghouse, J., Eslick, J., Lee, A., Burgard, A., Miller, D., and Biegler, L., A Complementarity-based Vapor-Liquid Equilibrium Formulation for Equation-Oriented Simulation and Optimization, AIChE Journal, 2023, Volume 69(4), e18029. https://doi.org/10.1002/aic.18029
+
+Introduction
+------------
+
+Often, a user may not know whether a state corresponds to a liquid, gas, or coexisting mixture. Even if a user knows the phase composition of a problem's initial condition, optimization may push the stream into or out of the two-phase region. Therefore, it is necessary to formulate phase equilibrium equations that are well-behaved for both one-phase and two-phase streams.
+
+To address this, the cubic smooth vapor-liquid equilibrium (VLE) formulation always solves the equilibrium equations at a condition where a valid two-phase solution exists. In situations where only a single phase is present, the phase equilibrium is solved at the either the bubble or dew point, where the second phase is just beginning to form; in this way, a non-trivial solution is guaranteed. Rather than explicitly calculate the bubble and dew points (as is done in the :ref:`non-cubic formulation <explanations/components/property_package/general/pe/smooth_flash:Smooth Vapor-Liquid Equilibrium Formulation (SmoothVLE)>`), this formulation leverages properties of the  cubic equation of state to identify the "equilibrium temperature".
+
+Formulation
+-----------
+
+.. note::
+  For the full derivation of the cubic smooth VLE formulation, see the reference above.
+
+.. note::
+  For consistency of naming between the cubic and non-cubic formulations, :math:`\bar{T}` is referred to as :math:`T_{eq}` in this document and the resulting model.
+
+The approach used by the smooth VLE formulation is to define an "equilibrium temperature" (:math:`T_{eq}`) at which the equilibrium calculations will be performed. The equilibrium temperature is defined such that:
+
+.. math:: T = T_{eq} - s_{vap} + s_{liq}
+
+where :math:`T` is the actual state temperature, and :math:`s_{liq}` and :math:`s_{vap}` are non-negative slack variables. For systems existing the the liquid-only region, :math:`s_{liq}` will be non-zero whilst :math:`s_{vap}=0` (indicating that the system is below the bubble point and thus :math:`T_{eq}>T`). Similarly, for systems in the vapor-only region, :math:`s_{vap}` will be non-zero whilst :math:`s_{liq}=0`. Finally, in the two-phase region, :math:`s_{liq}=s_{vap}=0`, indicating that :math:`T_{eq}=T`.
+
+In order to determine the values of :math:`s_{liq}` and :math:`s_{vap}`, the following complementarity constraints are written:
+
+.. math:: 0 = \min(s_{liq}, F_{liq})
+.. math:: 0 = \min(s_{vap}, F_{vap})
+
+where :math:`F_{p}` is the flow rate of each phase :math:`p`. That is, for each phase (liquid and vapor), if there is any flowrate associated with that phase (i.e., the phase exists), its slack variable must be equal to zero.
+
+Additionally, the follow complementarities are written to constraint the roots of the cubic equation of state.
+
+.. math:: 0 = \min(g^{+}_{liq}, F_{liq})
+.. math:: 0 = \min(g^{-}_{vap}, F_{vap})
+
+where :math:`g^{+}_p` and :math:`g^{-}_p` are another pair of non-negative slack variables associated with each phase :math:`p`. These slack variables are defined such that:
+
+.. math:: f''(Z_p) = g^{+}_{p} - g^{-}_{p}
+
+where :math:`f''(Z_p)` is the second derivative of the cubic equation of state written in terms of the compressibility factor :math:`Z_p` for each phase :math:`p`.
+
+Smooth Approximation
+''''''''''''''''''''
+
+In order to express the minimum operators in a tractable form, these equations are reformulated using the IDAES `smooth_min` function:
+
+.. math:: \min(a, b) =  0.5{\left[a + b - \sqrt{(a-b)^2 + \epsilon^2}\right]}
+
+Each complementarity requires a smoothing parameter, named :math:`\epsilon_T` and :math:`\epsilon_Z` for the temperature and cubic root constraints respectively. Within the IDAES model, these are rendered as ``eps_t_phase1_phase2`` and ``eps_z_phase1_phase2``, where ``phase1`` and ``phase2`` are the names assigned to the liquid and vapor phases in the property package (order will depend on the order these are declared).
+
+The tractability of the VLE problem depends heavily upon the values chosen for :math:`\epsilon_T` and :math:`\epsilon_Z`, with larger values resulting in smoother transitions at the phase boundaries (and thus increased tractability) at the expense of decreased accuracy near these points. It is recommended that users employ a 2-stage approach to solving these problems, starting with a larger value of :math:`\epsilon_T` and :math:`\epsilon_Z` initially to determine which region the solution lies in, followed by a second solve using smaller values to refine the solution.
+
+As a rule of thumb, the values of :math:`\epsilon_T` and :math:`\epsilon_Z` should be between 2 and 4 orders of magnitude smaller than the largest quantify involved in the smooth maximum operation. This means the value of :math:`\epsilon_T` should be based on the larger of :math:`T` and :math:`F_p`, whilst :math:`\epsilon_Z` should be based on the larger of :math:`f''(Z_p)` and :math:`F_p`. The value of :math:`f''(Z_p)` may be difficult to determine *a priori*, however :math:`F_p` is likely to dominate in most cases unless :math:`F_p` is small or :math:`P` is large.
+

docs/explanations/components/property_package/general/phase_equilibrium.rst

@@ -40,6 +40,7 @@ Phase Equilibrium State Libraries
     :maxdepth: 1
 
     pe/smooth_flash
+    pe/smooth_vle2
 
 Necessary Properties
 --------------------