{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Convergence"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"BioSTEAM iteratively solves for phase equilibria and mass and energy balances until they converge below a specified tolerance. Here we discuss ways to adjust the convergence criteria, issues that might affect convergence, and ways to customize convergence algorithms."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Recycle systems"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Adjusting convergence criteria"
]
},
{
"cell_type": "raw",
"metadata": {
"raw_mimetype": "text/restructuredtext"
},
"source": [
"A :class:`~biosteam.System` is composed of a path of units and subsytems and a recycle loop (if any). To solve for the molar flow rates and temperature of the recycle stream, BioSTEAM employs accelerated fixed-point convergence algorithms such as Aitken-Steffensen's method (the default method). Here is a list of defaults recycle systems use:"
]
},
{
"cell_type": "code",
"execution_count": 1,
"metadata": {
"raw_mimetype": "text/markdown",
"tags": [
"nbval-ignore-output"
]
},
"outputs": [],
"source": [
"from biosteam import System\n",
"System.default_maxiter # -> 200\n",
"System.default_molar_tolerance # -> 1. [kmol/hr] for each component \n",
"System.default_relative_molar_tolerance # -> 0.01 \n",
"System.default_temperature_tolerance # -> 0.10 [K]\n",
"System.default_relative_temperature_tolerance # -> 0.001\n",
"System.default_methods['Sequential modular']; # -> 'Aitken'\n",
"# Changing any of these would change the default for \n",
"# all new recycle systems you create."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"For a recycle system to converge, the difference between molar flow rate of the final and previous iteration must below **either** the relative or absolute molar tolerance (not both). Because the default molar tolarances are as wide as 1% or 0.01 kmol/hr, resimulating a system may give slightly different results in some cases. In the following example, we create a simple recycle loop consisting of a flash, a mixer, and a splitter:"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {
"scrolled": false,
"tags": [
"nbval-ignore-output"
]
},
"outputs": [
{
"data": {
"text/html": [
"\n",
" \n",
" "
],
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"System: sys\n",
"Highest convergence error among components in recycle\n",
"stream splitter-0 after 2 loops:\n",
"- flow rate 5.71e-04 kmol/hr (9.1%)\n",
"- temperature 0.00e+00 K (0%)\n",
"ins...\n",
"[0] feed \n",
" phase: 'l', T: 298.15 K, P: 101325 Pa\n",
" flow (kmol/hr): Water 1\n",
"outs...\n",
"[0] vapor_product \n",
" phase: 'g', T: 373.12 K, P: 101325 Pa\n",
" flow (kmol/hr): Water 0.943\n",
"[1] liquid_product \n",
" phase: 'l', T: 373.12 K, P: 101325 Pa\n",
" flow (kmol/hr): Water 0.0566\n"
]
}
],
"source": [
"import biosteam as bst\n",
"bst.nbtutorial() # Light-mode html diagrams and filter warnings\n",
"bst.settings.set_thermo(['Water'])\n",
"feed = bst.Stream('feed', Water=1)\n",
"recycle = bst.Stream('recycle')\n",
"liquid_product = bst.Stream('liquid_product')\n",
"mixer = bst.Mixer('mixer', [feed, recycle])\n",
"flash = bst.Flash('flash', mixer-0, ['vapor_product', ''], Q=feed.Hvap, P=101325)\n",
"splitter = bst.Splitter('splitter', flash.liquid, outs=[recycle, liquid_product], split=0.1) # flash.liquid is flash.outs[1]\n",
"sys = bst.main_flowsheet.create_system('sys')\n",
"sys.simulate()\n",
"sys.diagram()\n",
"sys.show()"
]
},
{
"cell_type": "raw",
"metadata": {
"raw_mimetype": "text/restructuredtext"
},
"source": [
"Note that a 9.1% convergence error is rather large. This is because the absolute flow rate error (6.35e-4 kmol/hr) is less than 1 kmol/hr, which is quite small for an industrial production process. We can update tolerances using the :meth:`~biosteam.System.set_tolerance` method before simulation to prevent this issue:"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {
"tags": [
"nbval-ignore-output"
]
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"System: sys\n",
"Highest convergence error among components in recycle\n",
"stream splitter-0 after 3 loops:\n",
"- flow rate 1.13e-17 kmol/hr (0%)\n",
"- temperature 0.00e+00 K (0%)\n",
"ins...\n",
"[0] feed \n",
" phase: 'l', T: 298.15 K, P: 101325 Pa\n",
" flow (kmol/hr): Water 1\n",
"outs...\n",
"[0] vapor_product \n",
" phase: 'g', T: 373.12 K, P: 101325 Pa\n",
" flow (kmol/hr): Water 0.943\n",
"[1] liquid_product \n",
" phase: 'l', T: 373.12 K, P: 101325 Pa\n",
" flow (kmol/hr): Water 0.0571\n"
]
}
],
"source": [
"sys.empty_recycles() # Restart simulation\n",
"sys.set_tolerance(mol=1e-9, rmol=1e-9) # Set absolute and relative tolerances\n",
"sys.simulate()\n",
"sys.show()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Convergence methods"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"It may be possible that a system may be more efficient to converge using another solver. Here is a list of solvers available in BioSTEAM:"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"['aitken',\n",
" 'wegstein',\n",
" 'fixedpoint',\n",
" 'anderson',\n",
" 'diagbroyden',\n",
" 'excitingmixing',\n",
" 'linearmixing',\n",
" 'broyden1',\n",
" 'broyden2']"
]
},
"execution_count": 4,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"list(bst.System.available_methods)"
]
},
{
"cell_type": "raw",
"metadata": {
"raw_mimetype": "text/restructuredtext"
},
"source": [
"Aitken, Wegstein, and fixed-point methods are in-house solvers from BioSTEAM. The other methods are imported from `scipy `_. \n",
"\n",
"Let's try solving the recycle system with the :func:`~scipy.optimize.diagbroyden` method:"
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {
"tags": [
"nbval-ignore-output"
]
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"System: sys\n",
"Highest convergence error among components in recycle\n",
"stream splitter-0 after 4 loops:\n",
"- flow rate 1.56e-11 kmol/hr (2.5e-07%)\n",
"- temperature 0.00e+00 K (0%)\n",
"ins...\n",
"[0] feed \n",
" phase: 'l', T: 298.15 K, P: 101325 Pa\n",
" flow (kmol/hr): Water 1\n",
"outs...\n",
"[0] vapor_product \n",
" phase: 'g', T: 373.12 K, P: 101325 Pa\n",
" flow (kmol/hr): Water 0.943\n",
"[1] liquid_product \n",
" phase: 'l', T: 373.12 K, P: 101325 Pa\n",
" flow (kmol/hr): Water 0.0571\n"
]
}
],
"source": [
"sys.set_tolerance(method='diagbroyden')\n",
"sys.empty_recycles()\n",
"sys.simulate()\n",
"sys.show()"
]
},
{
"cell_type": "raw",
"metadata": {
"raw_mimetype": "text/restructuredtext"
},
"source": [
"Notice how only 6 iterations was necessary using the diagonal Broyden methods. We can also register new methods such as `scipy's hybr-modified Powell solver `_ using :meth:`~biosteam.System.register_method`:"
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {
"tags": [
"nbval-ignore-output"
]
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"System: sys\n",
"Highest convergence error among components in recycle\n",
"stream splitter-0 after 7 loops:\n",
"- flow rate 4.57e-12 kmol/hr (7.2e-08%)\n",
"- temperature 0.00e+00 K (0%)\n",
"ins...\n",
"[0] feed \n",
" phase: 'l', T: 298.15 K, P: 101325 Pa\n",
" flow (kmol/hr): Water 1\n",
"outs...\n",
"[0] vapor_product \n",
" phase: 'g', T: 373.12 K, P: 101325 Pa\n",
" flow (kmol/hr): Water 0.943\n",
"[1] liquid_product \n",
" phase: 'l', T: 373.12 K, P: 101325 Pa\n",
" flow (kmol/hr): Water 0.0571\n"
]
}
],
"source": [
"from scipy.optimize import root\n",
"bst.System.register_method(\n",
" # Set solver tolerances extreemly high so that it does not stop early and\n",
" # only stops when BioSTEAM's tolerances are satisfied.\n",
" name='hybr', solver=root, options=dict(xtol=1e-24, maxfev=int(1e6), method='hybr')\n",
")\n",
"sys.set_tolerance(method='hybr')\n",
"sys.empty_recycles()\n",
"sys.simulate()\n",
"sys.show()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Importance of thermodynamic property package"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"The choice of thermodynamic property package may have a large impact in the results, regardless of convergence. For example, in the liquefaction of nitrogen, it is important to account for excess thermodynamic energies due to high pressures."
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {
"tags": [
"nbval-ignore-output"
]
},
"outputs": [
{
"data": {
"text/html": [
"\n",
" \n",
" "
],
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Results without excess free energies\n",
"------------------------------------\n",
"System: N2_liquefaction\n",
"Highest convergence error among components in recycle\n",
"stream flash-0 after 2 loops:\n",
"- flow rate 0.00e+00 kmol/hr (0%)\n",
"- temperature 0.00e+00 K (0%)\n",
"ins...\n",
"[0] inlet \n",
" phase: 'g', T: 280 K, P: 2e+07 Pa\n",
" flow (kmol/hr): N2 353\n",
"outs...\n",
"[0] out \n",
" phase: 'g', T: 280 K, P: 100000 Pa\n",
" flow (kmol/hr): N2 353\n",
"[1] flash_liquid \n",
" phase: 'l', T: 280 K, P: 100000 Pa\n",
" flow: 0\n",
"\n",
"Results with excess free energies\n",
"------------------------------------\n",
"System: N2_liquefaction\n",
"Highest convergence error among components in recycle\n",
"stream flash-0 after 11 loops:\n",
"- flow rate 1.53e-12 kmol/hr (4.7e-13%)\n",
"- temperature 0.00e+00 K (0%)\n",
"ins...\n",
"[0] inlet \n",
" phase: 'g', T: 280 K, P: 2e+07 Pa\n",
" flow (kmol/hr): N2 353\n",
"outs...\n",
"[0] out \n",
" phase: 'g', T: 270.45 K, P: 100000 Pa\n",
" flow (kmol/hr): N2 325\n",
"[1] flash_liquid \n",
" phase: 'l', T: 77.244 K, P: 100000 Pa\n",
" flow (kmol/hr): N2 28.2\n"
]
}
],
"source": [
"import biosteam as bst\n",
"bst.main_flowsheet.clear()\n",
"\n",
"# Setup thermodynamic backend\n",
"N2 = bst.Chemical('N2')\n",
"bst.settings.set_thermo([N2])\n",
"\n",
"# Heat integration between inlet and recycle\n",
"inlet = bst.Stream(\"inlet\", N2=2.75, units=\"kg/s\", T=6.85+273.15, P=200e5, phase=\"g\")\n",
"flash_gas = bst.Stream(\"flash_gas\")\n",
"regenerator = bst.units.HXprocess(\"regenerator\", ins=(inlet, flash_gas), outs=(\"throttle_in\", 'out'), dT=9.55)\n",
"throttle_in = regenerator.outs[0]\n",
"\n",
"# Throttling\n",
"valve = bst.units.IsenthalpicValve(\"expansion\", ins=throttle_in, outs=\"valve_out\", P=1e5, vle=True)\n",
"valve_out = valve.outs[0]\n",
"\n",
"# Flash drum\n",
"flash = bst.units.Flash(\"flash\", ins=valve_out, outs=(flash_gas, \"flash_liquid\"), Q=0, P=1e5)\n",
"\n",
"# Connect flash gas to regenerator\n",
"regenerator.ins[1] = flash.vapor\n",
" \n",
"sys = bst.main_flowsheet.create_system('N2_liquefaction')\n",
"sys.set_tolerance(mol=1e-9, rmol=1e-9, T=1e-5, rT=1e-5)\n",
"bst.settings.mixture.include_excess_energies = False\n",
"sys.simulate()\n",
"sys.diagram()\n",
"print('Results without excess free energies')\n",
"print('------------------------------------')\n",
"sys.show()\n",
"print()\n",
"print('Results with excess free energies')\n",
"print('------------------------------------')\n",
"sys.reset_cache() # Resets cached chemical/mixture properties\n",
"bst.settings.mixture.include_excess_energies = True\n",
"sys.empty_recycles()\n",
"sys.simulate()\n",
"sys.show()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"It may be preferable to use other methods for computing thermodynamic properties. For example, we can change the heat capacity methods to [CoolProp](http://www.coolprop.org/), which is known to be very rigorous:"
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {
"tags": [
"nbval-skip"
]
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"System: N2_liquefaction\n",
"Highest convergence error among components in recycle\n",
"stream flash-0 after 11 loops:\n",
"- flow rate 8.53e-13 kmol/hr (0%)\n",
"- temperature 0.00e+00 K (0%)\n",
"ins...\n",
"[0] inlet \n",
" phase: 'g', T: 280 K, P: 2e+07 Pa\n",
" flow (kmol/hr): N2 353\n",
"outs...\n",
"[0] out \n",
" phase: 'g', T: 270.45 K, P: 100000 Pa\n",
" flow (kmol/hr): N2 325\n",
"[1] flash_liquid \n",
" phase: 'l', T: 77.244 K, P: 100000 Pa\n",
" flow (kmol/hr): N2 28.2\n"
]
}
],
"source": [
"N2.Cn.l.method = N2.Cn.g.method = \"COOLPROP\" # You may need to \"pip install coolprop\" for this\n",
"N2.reset_free_energies() # This is necessary to update enthalpy and entropy algorithms with new heat capacities\n",
"sys.reset_cache()\n",
"sys.empty_recycles()\n",
"sys.simulate()\n",
"sys.show()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"If you are working with gas mixtures at high pressures, it is important to account for excess energies of mixing. You can do so by using mixture equations of state. This is facilitated through EOSMixture objects like PRMixture or SRKMixture, which always account for excess energies:"
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"With SRK EOS\n",
"------------------\n",
"MultiStream: \n",
"phases: ('g', 'l'), T: 188.99 K, P: 1e+07 Pa\n",
"flow (kmol/hr): (g) N2 9.46\n",
" CO2 0.544\n",
" (l) N2 0.544\n",
" CO2 9.46\n",
"With ideal mixture\n",
"------------------\n",
"MultiStream: \n",
"phases: ('g', 'l'), T: 203.72 K, P: 1e+07 Pa\n",
"flow (kmol/hr): (g) N2 9.72\n",
" CO2 0.278\n",
" (l) N2 0.278\n",
" CO2 9.72\n"
]
}
],
"source": [
"chemicals = bst.Chemicals(['N2', 'CO2'])\n",
"mixture = bst.SRKMixture.from_chemicals(chemicals) # Soave-Redlich-Kuang EOS\n",
"bst.settings.set_thermo(chemicals, mixture=mixture, Phi=bst.SRKFugacityCoefficients)\n",
"s = bst.Stream(None, N2=10, CO2=10, phase='g')\n",
"s.vle(V=0.5, P=1e7)\n",
"print('With SRK EOS')\n",
"print('------------------')\n",
"s.show()\n",
"\n",
"bst.settings.set_thermo(chemicals)\n",
"s = bst.Stream(None, N2=10, CO2=10, phase='g')\n",
"s.vle(V=0.5, P=1e7)\n",
"print('With ideal mixture')\n",
"print('------------------')\n",
"s.show()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Process specification issues"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"At times, a process specification might involve having to rerun upstream unit operations. In such cases, it may be easier to set the system to run more than once than to code-in the simulation of upstream units. For demonstration purposes, let's specify a mixer's product flow rate of water to be constant by varying one of the feeds:"
]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"HXutility: heat_exchanger\n",
"ins...\n",
"[0] feed_a \n",
" phase: 'l', T: 298.15 K, P: 101325 Pa\n",
" flow (kmol/hr): Water 9.99e+05\n",
"outs...\n",
"[0] s1 to Mixer-mixer\n",
" phase: 'l', T: 298.15 K, P: 101325 Pa\n",
" flow: 0\n"
]
}
],
"source": [
"bst.main_flowsheet.clear()\n",
"bst.settings.set_thermo(['Water'])\n",
"feed_a = bst.Stream('feed_a')\n",
"heat_exchanger = bst.HXutility('heat_exchanger', feed_a, T=320, rigorous=True)\n",
"feed_b = bst.Stream('feed_b', Water=1e3)\n",
"mixer = bst.Mixer('mixer', [heat_exchanger-0, feed_b])\n",
"mixer_product_flow_specification = 1e6 # kmol/hr\n",
"@mixer.add_specification\n",
"def adjust_fresh_flow():\n",
" # Updates the feed, but doesn't run the upstream heat exchanger.\n",
" # A good solution would be to also run the heat exchanger, but here\n",
" # we do not for demonstration purposes\n",
" feed_a.imol['Water'] = mixer_product_flow_specification - feed_b.imol['Water']\n",
" mixer.run()\n",
"\n",
"sys = bst.main_flowsheet.create_system('sys')\n",
"sys.simulate()\n",
"heat_exchanger.show()"
]
},
{
"cell_type": "raw",
"metadata": {
"raw_mimetype": "text/restructuredtext"
},
"source": [
"Although we changed the feed upstream of the heat exchanger, notice how we will need to rerun the heat exchanger. We can do this by setting the number of times a system runs using :attr:`~biosteam.System.N_runs`:"
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"HXutility: heat_exchanger\n",
"ins...\n",
"[0] feed_a \n",
" phase: 'l', T: 298.15 K, P: 101325 Pa\n",
" flow (kmol/hr): Water 9.99e+05\n",
"outs...\n",
"[0] s1 to Mixer-mixer\n",
" phases: ('g', 'l'), T: 320 K, P: 101325 Pa\n",
" flow (kmol/hr): (l) Water 9.99e+05\n"
]
}
],
"source": [
"sys.N_runs = 2 # Run twice in simulate method\n",
"sys.empty_recycles()\n",
"sys.simulate()\n",
"heat_exchanger.show()"
]
},
{
"cell_type": "raw",
"metadata": {
"raw_mimetype": "text/restructuredtext"
},
"source": [
"In general, depending on :attr:`~biosteam.System.N_runs` is not recommended, but it may be useful when dealing with complex process specifications."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Phase equilibrium and energy balance"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Adjusting convergence criteria"
]
},
{
"cell_type": "raw",
"metadata": {
"raw_mimetype": "text/restructuredtext"
},
"source": [
"The tolerance on solving for temperature at an energy balance can by adjusted through the :class:`~thermosteam.mixture.Mixture` class:"
]
},
{
"cell_type": "code",
"execution_count": 12,
"metadata": {
"tags": [
"nbval-ignore-output"
]
},
"outputs": [],
"source": [
"from thermosteam import Mixture\n",
"Mixture.T_tol # -> 1e-6\n",
"Mixture.maxiter; # -> 20"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Thermodynamic phase equilibrium algorithms also have a number of tolerances that may be adjusted:"
]
},
{
"cell_type": "code",
"execution_count": 13,
"metadata": {
"tags": [
"nbval-ignore-output"
]
},
"outputs": [],
"source": [
"from thermosteam.equilibrium import VLE\n",
"VLE.maxiter # -> 20 [-]\n",
"VLE.T_tol # -> 5e-8 [K]\n",
"VLE.P_tol # -> 1. [Pa]\n",
"VLE.H_hat_tol # -> 1e-6 [J/g]\n",
"VLE.S_hat_tol # -> 1e-6 [J/g/K]\n",
"VLE.V_tol # -> 1e-6 [mol %] \n",
"VLE.x_tol # -> 1e-9 [mol %]\n",
"VLE.y_tol; # -> 1e-9 [mol %]"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"The tolerances thermodynamic convergence are purposefully low for stable system convergence. In rare occasions, these algorithms may fail to converge and increasing the maximum number of iterations, `maxiter`, may help. Increasing tolerances may also help convergence of a unit operation, but may cause instabilities to recycle systems."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Importance of thermodynamic property package"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"As an example to drive home how important it is to understand the impact of thermodynamic and phase equilibrium assumptions, let's compare the phase envelope of ethanol and water with and without activity coefficients:"
]
},
{
"cell_type": "code",
"execution_count": 14,
"metadata": {
"tags": [
"nbval-skip"
]
},
"outputs": [
{
"data": {
"image/png": "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",
"text/plain": [
"