pvcircuit

pvcircuit contains objects that are building blocks for PV modeling and interactive data fitting.

Based on publications:

2 terminal multijunction model: - J. F. Geisz, et al., IEEE Journal of Photovoltaics 5, p. 1827 (2015).

3 terminal tandem model: - J. F. Geisz, et al., Cell Reports Physical Science 2, p. 100677 (2021).

Luminescent coupling correction of EQE: - M. A. Steiner, et al., IEEE Journal of Photovoltaics 3, p. 879 (2013).

Energy yield: - J. M. Ripalda, et al., Nature Communications 9, p. 5126 (2018).

Installation

• install GitHub Desktop
• on this https://github.com/NREL/PVcircuit page click the green "Code" button and "Open with GitHub Desktop"
• cd to the PVcircuit directory in terminal and type "pip install -e ."

Packages needed

• pandas
• numpy
• matplotlib
• scipy
• ipywidgets
• ipympl
• parse
• num2words
• tandems > for fast energy yield calculations: > > git clone https://github.com/Ripalda/Tandems ~/Documents/GitHub/Tandems > > cd ~/Documents/GitHub/Tandems > > pip install -e .

Junction( ) Class

A single Junction that can be combined together to form tandem and multijunction solar cells.

The Junction is modeled by an arbitrary number of parallel diodes such as n=1, n=2, n=2/3, or any other ideality factor. The temperature dependance of each diode saturation current J0n is determined relative to the Jdb(Eg,T).

Junction.attributes

Junction.name

Name of this junction string

Junction.Eg

Junction bandgap (eV)

Junction.TC

Junction temperature (C)

Junction.Gsh

Junction shunt conductance (S/cm2)

Junction.Rser

Junction series resistance (Ohm cm2)

Junction.lightarea

Illuminated area (cm2)

Junction.totalarea

Total area including any dark area (cm2) always greater than or equal to .lightarea

Junction.Jext

External photocurrent (A/cm2)

Junction.JLC

Luminescent coupling photocurrent (A/cm2)

Junction.beta

Electro luminescent coupling factor (unitless) Amount of light collected as current in an adjectent junction relative to Jdb

Junction.gamma

Photo luminescent coupling factor (unitless)

Junction.pn

Direction of pn diode -1 n-on-p 0 simple resistor 1 p-on-n

Junction.n[n0, n1, etc]

Ideality factor numpy.array

Junction.J0ratio[J0ratio0, J0ratio1, etc]

Ratio of saturation current densities relative to Jdb numpy.array

J0ratio = J0 / (Jdb)^(1/n)

Junction.ui

User interface created by .controls() Consists of plot, widget controls, and other outputs

Junction.debugout

Hidden output of for debugging

Junction.properties

Junction.TK

Temperature (K) from TC

Junction.Vth

Thermal voltage = kT/q from TK

Junction.Jdb

Detailed balance current (A/cm2). If J0ratio for n=1 only this is Shockley-Quiesser limit

Junction.Jphoto

Total photocurrent = Jext+JLC

Junction.J0[J00, J01, etc]

Reverse saturation currents (A/cm2) corresponding to each n[] and J0ratio[] numpy.array

Junction.methods()

Junction.copy()

Create a copy of a Junction

Junction.set(**kwargs)

Controlled change of Junction attributes

Junction.controls()

Create a user interface Junction.ui which consists of widget controls that are mainly used to build up Multi2T.ui and Tandem3T.ui

Junction.update()

Update Junction attributes into Junction.ui controls if they exist

Junction._J0init(J0ref)

initialize self.J0ratio from J0ref

returns np.ndarray [J0(n0), J0(n1), etc]

Junction.Jem(Vmid)

Light emitted from junction by reciprocity, quantified as current density (A/cm2)

Junction.notdiode()

boolean whether the junction is a junction (False) or resistor (True)

Junction.Jmultidiodes(Vdiode)

Calculate total recombination current density from multiple parallel diodes: n[], J0[] given voltage across diode without series resistance.

Junction.JshuntRBB(Vdiode)

returns shunt + reverse-bias breakdown current

RBB_dict={'method':None}
RBB_dict={'method':'JFG', mrb'':10., 'J0rb':1., 'Vrb':0.}
RBB_dict={'method':'bishop','mrb'':3.28, 'avalanche':1, 'Vrb':-5.5}
RBB_dict={'method':'pvmismatch','ARBD':arbd,'BRBD':brbd,'VRBD':vrb,'NRBD':nrbd}

Junction.Jparallel(Vdiode,Jtot)

Circuit equation to be zeroed to solve for Vi for voltage across parallel diodes with shunt and reverse breakdown.

Junction.Vdiode(Jdiode)

Calculate voltage across diode without series resistance as a function current density through the diode.

Junction._dV(Vmid, Vtot)

Circuit equation to be zeroed (returns voltage difference) to solve for Vmid. Single junction circuit with series resistance and parallel diodes. internal use only

Junction.Vmid(Vtot)

Find intermediate voltage in a single junction diode with series resistance given Vtot including series resistance

Vtot = Vparallel + Rser * Jparallel

Multi2T( ) Class

Two terminal multijunction device composed of any number of series connected Junctions. The sum of all Rser is an attribute of the Multi2T object and the Rser attributes of each sub Junction are ignored

Multi2T.attributes

Multi2T.name

Name of this junction string

Multi2T.Rs2T

Total series resistance of Multi2T. (Junction.Rser are ignored)

Multi2T.njunc

Number of junctions contained in Multi2T

Multi2T.j

List of Junction objects series-connected within Multi2T object

Multi2T.Vmid[ ]

List of intermediate voltages between series-connected junctions

Multi2T.ui

User interface created by Multi2T.controls() which consists of plot, widget controls, and other outputs

Multi2T.debugout

Hidden output of for debugging

Multi2T.properties

Multi2T.TC

Maximum of the contained Junction.TCs

Multi2T.lightarea

Maximum of the contained Junction.lightareas

Multi2T.totalarea

Maximum of the contained Junction.totalareas

Multi2T.methods()

Multi2T.copy()

Create a copy of this Multi2T object

Multi2T.copy3T()

Create a Multi2T object from a Tandem3T object

Multi2T.single(junc, copy=True)

Create a Multi2T object from a Junction object

Multi2T.set(**kwargs)

Controlled change of Multi2T and its Junction's attributes

Multi2T.proplist(key)

Create a list of the scalar attributes or properties of the Junctions within a Multi2T object

Multi2T.controls()

Create a user interface Multi2T.ui which consists of plot, widget controls, and other outputs

Multi2T.update()

Update Multi2T and its Junction's attributes into Multi2T.ui controls if they exist

Multi2T.V2T(I)

Calculates the total Multi2T voltage as a function of the series-connected current

Also sets Multi2T.Vmid[]

Inputs scalar and outputs scalar.

Can be vectorized by

 V2Tvect = np.vectorize(self.V2T)

Multi2T.I2T(V)

Calculates the series-connected current as a function of total current Multi2T voltage.

Also sets Multi2T.Vmid[]

Inputs scalar and outputs scalar.

Can be vectorized by

 I2Tvect = np.vectorize(self.I2T)

Multi2T.Imaxrev()

maximum reverse-bias current (A) without Gsh or breakdown

Multi2T.Voc()

Open-circuit voltate (V) of Muli2T object

Multi2T.Isc()

Short-circuit current (A) of Multi2T object

Multi2T.MPP(pnts=11, bplot=False, timer=False)

Fast method to calculate the maximum power point of Multi2T object. Outputs dictionary:

{"Voc":Voc, "Isc":Isc, "Vmp":Vmp, "Imp":Imp, "Pmp":Pmp,  "FF":FF}

Multi2T.calcDark(hilog = 3, pdec = 5, timer=False)

Calculate a dark IV curve outputs: (Idark, Vdark, Vdarkmid)

Multi2T.calcLight(pnts=21, Vmin=-0.5, timer=False, fast=False)

Calculate a light IV curve outputs: (Vlight, Ilight, Plight, MPP)

Multi2T.plot(title='', pplot=False, dark=None, pnts=21, Vmin= -0.5, lolog = -8, hilog = 7, pdec = 5)

Create a light or dark plot modeled from Multi2T parameters. Outputs: (mpl.Figure, mpl.Axes)

Tandem3T( ) Class

Three terminal (3T) tandem composed of two Junctions. Four terminal (4T) tandems can be modeled as 3T tandems with no resistive coupling (Rz=0) but still require luminescent coupling. The 4T shunt (or breakdown) between the subcells is not treated but could become important for large voltage differences.

Tandem3T.attributes

Tandem3T.name

Name of this junction string

Tandem3T.Rz

Common series resistance of Tandem3T z-terminal

Tandem3T.top

Top Junction object of Tandem3T object

Tandem3T.bot

Bottom Junction object of Tandem3T object

Tandem3T.ui

User interface created by Tandem3T.controls() which consists of plot, widget controls, and other outputs

Tandem3T.debugout

Hidden output of for debugging

Tandem3T.properties

Tandem3T.TC

Maximum of the contained Junction.TCs

Tandem3T.lightarea

Maximum of the contained Junction.lightareas

Tandem3T.totalarea

Maximum of the contained Junction.totalareas

Tandem3T.methods()

Tandem3T.copy()

Create a copy of this Tandem3T object

Tandem3T.set(**kwargs)

Controlled change of Tandem3T and its Junction's attributes

Tandem3T.controls()

Create a user interface Tandem3T.ui which consists of plot, widget controls, and other outputs

Tandem3T.update()

Update Tandem3T and its Junction's attributes into Tandem3T.ui controls if they exist

Tandem3T.V3T(iv3T)

Calcuate iv3T.(Vzt,Vrz,Vtr) from iv3T.(Iro,Izo,Ito)

input/output IV3T object

Tandem3T.J3Tabs(iv3T)

Calcuate (Jro,Jzo,Jto) mapped -> iv3T.(Iro,Izo,Ito) from ABSOLUTE (Vz,Vr,Vt) mapped <- iv3T.(Vzt,Vrz,Vtr)

input/output IV3T object

Tandem3T.I3Trel(iv3T)

calcuate (Jro,Jzo,Jto) mapped -> iv3T.(Iro,Izo,Ito) from RELATIVE iv3T.(Vzt,Vrz,Vtr) ignoring Vtr

input/output IV3T object

Tandem3T._dI(Vz,Vzt,Vrz,temp3T)

Return dI = Iro + Izo + Ito function solved for dI(Vz)=0 in I3rel

input Vzt, Vrz, IV3T object

Tandem3T.Voc3(meastype='CZ')

Triple Voc of 3T tandem returns IV3T object.

returns IV3T object with one point

    (Vzt, Vrz, Vtr) of (Iro = 0, Izo = 0, Ito = 0)

Tandem3T.Isc3(meastype='CZ')

Triple Isc of 3T tandem returns IV3T object.

returns IV3T object with one point

    (Iro, Izo, Ito ) of (Vzt = 0, Vrz = 0, Vtr = 0)

Tandem3T.MPP(pnts=31, VorI= 'I', less = 2., bplot=False)

Iteratively find unconstrained MPP from lines as experimentally done. Varying I is faster than varying V but initial guess is not as good.

returns IV3T object with one point

'less' must be > 1.0. If FF is really bad, may need larger 'less'

Use 'bplot' for debugging information

Tandem3T.VM(self, bot, top, pnts=11)

Create voltage matched (VM) constrained line for tandem3T. Focus iteratively on the MPP of the constrained line.

'bot' bottom subcells in parallel with 'top' top subcells

returns two IV3T objects - the constrained line and MPP

Tandem3T.CM(self, pnts=11)

Create current matched (CM) constrained line for tandem3T. Focus iteratively on the MPP of the constrained line.

returns two IV3T objects - the constrained line and MPP

Tandem3T.VI0(VIname, meastype='CZ')

Solve for mixed (V=0, I=0) zero power points using separate diodes for quick solutions

returns IV3T object with one point

Tandem3T.VIpoint(zerokey, varykey, crosskey, meastype='CZ', pnts=11, bplot=False)

Solve for mixed (V=0, I=0) zero power points does not work well

Tandem3T.specialpoints(meastype = 'CZ', bplot=False, fast=False)

Compile all the special zero power points into a labeled IV3T object

Tandem3T.plot(pnts=31, meastype='CZ', oper = 'load2dev', cmap='terrain')

Calculate and plot Tandem3T device

IV3T( ) Class

Structure to contain all the information about the operational state of a 3T tandem.

Device parameters calculated for a 'Tandem3T' object.

(Iro, Izo, Ito) <-> (Vzt, Vrz, Vtr)

Device paramters converted to load parameters for given measurement configuration (CZ, CR, CT).

(Iro, Izo, Ito) <-> (IA, IB)

(Vzt, Vrz, Vtr) <-> (VA, VB)

Hexagonal representation of 3 device parameters in 2 dimensions.

(Iro, Izo, Ito) <-> (Ixhex, Iyhex)

(Vzt, Vrz, Vtr) <-> (Vxhex, Vyhex)

IV3T.attributes

IV3T.name

string Name of this IV3T

IV3T.meastype

string Measurement type of IV3T object 'CZ', 'CR', 'CT', 'CF', 'CRo', 'CTo', 'CZo', or 'CFo'

IV3T.shape

Shape of numpy.array that contains array points of IV3T object. tuple either 1D ie. (npts,) or 2D (xpnts, ypnts)

IV3T.area

float cell area used to calculate current density or power density from current or power

IV3T.xkey

string Name of the arraykey that is systematically varied.

IV3T.ykey

string Name of the orthoganal arraykey that is systematically varied by IV3T.box() or constrained by IV3T.line()

IV3T.x

1D ndarray of xkey values of box or line

IV3T.y

1D ndarray of ykey values of box or line

IV3T.names[ ]

List of labels of each point within IV3T using a string (optional but used by special points)

IV3T.Iro[ ], IV3T.Izo[ ], IV3T.Ito[ ], IV3T.IA[ ], IV3T.IB[ ]

1D or 2D ndarray of current (A) values for each Tandem3T operating point.

[Iro,Izo,Ito] are device currents.

[IA, IB] are load currents relative to meastype

IV3T.Vzt[ ], IV3T.Vrz[ ], IV3T.Vtr[ ], IV3T.VA[ ], IV3T.VB[ ]

1D or 2D ndarray of device voltage (V) values for each Tandem3T operating point

[Vzt,Vrz,Vtr] are device voltages.

[VA, VB] are load voltage relative to meastype

IV3T.Ixhex[ ], IV3T.Iyhex[ ], IV3T.Vxhex[ ], IV3T.Vyhex[ ]

1D or 2D ndarray of device voltage (V) or current (A) values mapped onto a 2D isometric hexagonal representation.

IV3T.Ptot[ ]

1D or 2D ndarray of total Tandem3T powers

IV3T.methods()

IV3T.copy()

Create a separate complete copy of a IV3T

IV3T.set()

IV3T.line(xkey, x0, x1, xn, ykey, yconstraint)

Create a 1D ndarray on xkey with evenly spaced values in IV3T..x

ykey is constrained to xkey with eval expression using 'x'

IV3T.box(xkey, x0, x1, xn, ykey, y0, y1, yn)

Create a 2D ndarray for xkey and ykey with shape (xn, yn) with evenly spaced values. IV3T.x are values in one dimention. IV3T.y are values in orthogonal dimention

IV3T.hexgrid(ax, VorI, step, xn=2, maxlines=10)

Add hexagonal grid lines to mpl.Axes. Range determined from self box IV3T object

IV3T.nanpnt(index)

Make indexed point in each keyarray an nan where index is a tuple i = (i, ) or (i,j)

IV3T.MPP(name='')

Find the MPP from within points of IV3T and return a new IV3T object with one point.

IV3T.sizes(klist)

Return a range of sizes of the array attributes in klist. Returns tuple (nmin,nmax)

IV3T.resize(self, shape, fillname = '')

Resize IV3T arrays and clears the values

IV3T.append(iv3T)

Appends one IV3T object onto another

IV3T.init(inlist,outlist)

Initialize output arrays to nan if input arrays are consistent

IV3T.kirchhoff(two)

Apply or check Kirchoff's law on [Iro,Izo,Ito] or [Vzt,Vrz,Vtr]

Input a list of 2 or 3 of the device input keys:

        2 -> calculate the third device value from other two knowns
        3 -> check the validity of 3 device parameters

IV3T.Pcalc(oper='dev2load', meastype=None)

Calculate Ptot after converting using oper = 'dev2load' or 'load2dev'

IV3T.loadlabel(load, meastype=None)

Return descriptive axis label for load variables. Add an extra character to swap the loads: 'CRo','CTo','CZo', 'CFo'

IV3T.convert(VorI, oper, meastype=None)

Calculate some array values from other array values. Can optionally set the meastype here.

VorI: 'V' or 'I'

oper: 'load2dev', 'dev2load', 'dev2hex', 'hex2dev' (not developed yet)

meastype: 'CR','CT','CZ','CF' or swap the loads: 'CRo','CTo','CZo', 'CFo'

IV3T.loadcsv(name, path, fileA, fileB, VorI, meastype, Iscale=1000.)

import csv file as data table into iv3T object

two 2D arrays with x and y index on top and left

load variables:
VA(IA,IB) & VB(IA,IB) .......... VorI='I'
or
IA(VA,VB) & IB(VA,VB) .......... VorI='V'
Iscale converts current mA -> A or mA/cm2-> A

IV3T.plot(xkey = None, ykey = None, zkey = None, inplot = None, cmap='terrain', ccont = 'black', bar = True)

Plot 2D IV3T object zkey(xkey,ykey) as image if evenly spaced or randomly spaced with contours

IV3T.addpoints(ax, xkey, ykey, **kwargs)

Plot IV3T points or lines onto existing axes

QE Analysis Functions

qe.EQE( ) Class

Object to contain all EQE information

EQE.attributes

EQE.name

string Name of this EQE object

EQE.rawEQE

numpy.array 2D(lambda)(junction) raw input rawEQE (not LC corrected)

EQE.xEQE

numpy.array wavelengths [nm] for rawEQE data

EQE.njuncs

int number of junctions

EQE.sjuncs

str names of junctions used in plot legend

EQE.nQlams

int number of wavelengths in rawEQE data

EQE.corrEQE

numpy.array luminescent coupling corrected EQE same size as rawEQE

EQE.etas

numpy.array LC factor for next three junctions

EQE.methods()

EQE.LCcorr(self)

Calculate LC corrected EQE using procedure from Steiner et al., IEEE PV, v3, p879 (2013)

Creates self.corrEQE

EQE.Jdb(self, TC, Eguess = 1.0, kTfilter=3, dbug=False)

Calculate Jscs and Egs from self.corrEQE

EQE.Jint(self, Pspec='global', xspec=wvl)

Integrate junction currents = integrate (spectra * lambda * EQE(lambda))

EQE.plot(self, Pspec='global', ispec=0, specname=None, xspec=wvl)

Plot self.rawEQE and self.corrEQE on top of a spectrum

other qe functions

qe.JintMD(EQE, xEQE, Pspec, xspec=wvl)

obsolete use EQE.Jint calculate total power of spectra and Jsc of each junction from multi-dimentional EQE - integrate multidimentional QE(lambda)(junction) times MD reference spectra Pspec(lambda)(ispec) - external quantum efficiency QE[unitless] x-units = nm, - reference spectra Pspec[W/m2/nm] x-units = nm - optionally Pspec as string 'space', 'global', or 'direct' - xEQE in nm, can optionally use (start, step) for equally spaced data - default x values for Pspec from wvl

output: - Jsc[junc,spectrum] = int(Pspec[spectrum]EQE[junc]lambda) - total power=int(Pspec) if EQE is None

qe.JdbMD(EQE, xEQE, TC, Eguess = 1.0, kTfilter=3, bplot=False)

obsolte use EQE.Jdb calculate detailed-balance reverse saturation current from EQE vs xEQE - xEQE[=]nm - can optionally use (start, step) for equally spaced data - debug on bplot

qe.JdbFromEg(TC,Eg,dbsides=1.,method=None)

return the detailed balance dark current (see Geisz EL&LC paper, King EUPVSEC) assuming a square EQE - Eg[=]eV - TC[=]C - returns Jdb[=]A/cm2

optional parameters - method: 'gamma' - dbsides: single-sided->1. bifacial->2.

qe.EgFromJdb(TC, Jdb, Eg=1.0, eps=1e-6, itermax=100, dbsides=1.):

return the bandgap from the Jdb - assuming a square EQE - iterates using gammaInc(3,x)=2exp(-x)(1+x+x^2/2)

optional parameters: - Eg=1.0 eV #initial guess - eps=0.001 #tolerance - itermax=100 #maximum iterations - dbsides=1. #bifacial->2.

EY.TMY( ) Class

Energy yield (EY) typical meterological year (TMY) at a specific location

Currently imports US proxy spectra from https://github.com/Ripalda/Tandems

Expects 'Tandems' github to be parallel to 'PVcircuit' github

TMY.attributes

TMY.name

string Name of this TMY

TMY.tilt

boolean tilt=True, axis=False

TMY.longitude

scalar

TMY.latitude

scalar

TMY.altitude

scalar

TMY.zone

scalar

TMY.DayTime

scalar

TMY.Irradiance

numpy.array Spectral irradiance each spectrum as a function of wavelength(wvl) 2D [nspecs][nlambda]

TMY.Temp

numpy.array Ambient temperature associate with each spectrum 1D [nspecs]

TMY.TempCell

numpy.array Cell temperature associate with each spectrum 1D [nspecs]

TMY.Wind

numpy.array Wind speed associate with each spectrum 1D [nspecs]

TMY.NTime

numpy.array Fractional time associate with each spectrum 1D [nspecs]

TMY.Angle

numpy.array Angle associate with each spectrum 1D [nspecs]

TMY.AngleMOD

numpy.array Angle modifier factor associate with each spectrum 1D [nspecs]

TMY.SpecPower

numpy.array Optical power associate with each spectrum 1D [nspecs]

TMY.RefPower

numpy.array Optical power associate with each reference spectrum 1D [3]

TMY.inPower

numpy.array (SpecPower * NTime) associate with each spectrum 1D [nspecs]

TMY.YearlyEnergy

scalar Yearly Optical Energy Resource [kWh/m2/yr]

TMY.methods()

TMY.cellcurrents(self,EQE, STC=False)

Calculate subcell currents under TMY for a given EQE class or standard test conditions (if STC=True)

Creates numpy.array for subsequent calculations - self.JscSTCs[refspec,junc] if STC=True - self.Jscs[spec,junc]

TMY.cellbandgaps(self,EQE,TC=25)

Calculate Egs and Jdbs under TMY for a given EQE class

Creates numpy.array for subsequent calculations - self.Jdbs[junc] - self.Eg[junc]

TMY.cellSTCeff(self,model,oper,iref=1)

Calculate efficiency of a cell under a reference spectrum self.JscSTCs and self.Egs must be calculate first.

Inputs - cell 'model' can be 'Multi2T' or 'Tandem3T' objects - 'oper' describes operation method unconstrained 'MPP', series-connected 'CM', parallel-configurations 'VM21', etc - iref = 0 -> space - iref = 1 -> global - iref = 2 -> direct

Outputs - STCeff efficiency of cell under reference spectrum (space,global,direct)

TMY.cellEYeff(self,model,oper)

Calculate efficiency of a cell under self (TMY). self.Jscs and self.Egs must be calculate first.

Inputs - cell 'model' can be 'Multi2T' or 'Tandem3T' - 'oper' describes operation method unconstrained 'MPP', series-connected 'CM', parallel-configurations 'VM'

Outputs - EY energy yield of cell [kWh/m2/yr] - EYeff energy yield efficiency = EY/YearlyEnergy

other EY functions

VMloss(type3T, bot, top, ncells):

Calculates approximate loss factor for VM strings of 3T tandems

VMlist(mmax):

generate a list of VM configurations + 'MPP'=4T and 'CM'=2T

mmax < 10 for formating reasons

cellmodeldesc(model,oper):

return description of model and operation - cell 'model' can be 'Multi2T' or 'Tandem3T' - 'oper' describes operation method unconstrained 'MPP', series-connected 'CM', parallel-configurations 'VM'

Outputs: (bot, top, ratio, type3T)