ltbams 0.9.9__py3-none-any.whl → 1.0.2__py3-none-any.whl

ams/pypower/routines/cpf.py

@@ -0,0 +1,513 @@
+"""
+Runs a full AC continuation power flow.
+"""
+# --- run cpf ---
+import logging
+
+import numpy as np
+from numpy import flatnonzero as find
+
+import scipy.sparse as sp
+
+from andes.shared import deg2rad
+from andes.utils.misc import elapsed
+
+import ams.pypower.utils as putil
+import ams.pypower.io as pio
+import ams.pypower.routines.opffcns as opfcn
+from ams.pypower.idx import IDX
+from ams.pypower.make import (makeSbus, makeYbus, dSbus_dV)
+from ams.pypower.routines.pflow import newtonpf, pfsoln
+from ams.pypower.core import ppoption
+import ams.pypower.routines.cpf_callbacks as cpf_callbacks
+
+from ams.shared import inf, nan
+
+logger = logging.getLogger(__name__)
+
+
+def runcpf(casedata, ppopt=None, scale=1.2):
+    """
+    Runs a full AC continuation power flow.
+    """
+    ppopt = ppoption(ppopt)
+
+    # options
+    verbose = ppopt["VERBOSE"]
+    step = ppopt["CPF_STEP"]
+    parameterization = ppopt["CPF_PARAMETERIZATION"]
+    adapt_step = ppopt["CPF_ADAPT_STEP"]
+    cb_args = ppopt["CPF_USER_CALLBACK_ARGS"]
+
+    # set up callbacks
+    callback_names = ["cpf_default_callback"]
+    if len(ppopt["CPF_USER_CALLBACK"]) > 0:
+        if isinstance(ppopt["CPF_USER_CALLBACK"], list):
+            callback_names = np.r_[callback_names, ppopt["CPF_USER_CALLBACK"]]
+        else:
+            callback_names.append(ppopt["CPF_USER_CALLBACK"])
+    callbacks = []
+    for callback_name in callback_names:
+        callbacks.append(getattr(cpf_callbacks, callback_name))
+
+    # read base case data
+    ppcbase = pio.loadcase(casedata)
+    nb = ppcbase["bus"].shape[0]
+
+    # add zero columns to branch for flows if needed
+    if ppcbase["branch"].shape[1] < IDX.branch.QT:
+        ppcbase["branch"] = np.c_[ppcbase["branch"],
+                                  np.zeros((ppcbase["branch"].shape[0],
+                                            IDX.branch.QT - ppcbase["branch"].shape[1] + 1))]
+
+    # convert to internal indexing
+    ppcbase = opfcn.ext2int(ppcbase)
+    baseMVAb, busb, genb, branchb = \
+        ppcbase["baseMVA"], ppcbase["bus"], ppcbase["gen"], ppcbase["branch"]
+
+    # get bus index lists of each type of bus
+    ref, pv, pq = putil.bustypes(busb, genb)
+
+    # generator info
+    onb = find(genb[:, IDX.gen.GEN_STATUS] > 0)  # which generators are on?
+    gbusb = genb[onb, IDX.gen.GEN_BUS].astype(int)  # what buses are they at?
+
+    # scale the load and generation of base case as target case
+    ppctarget = pio.loadcase(casedata)
+    ppctarget["bus"][:, IDX.bus.PD] *= scale
+    ppctarget["bus"][:, IDX.bus.QD] *= scale
+    ppctarget["gen"][:, IDX.gen.PG] *= scale
+    ppctarget["gen"][:, IDX.gen.QG] *= scale
+
+    # add zero columns to branch for flows if needed
+    if ppctarget["branch"].shape[1] < IDX.branch.QT:
+        ppctarget["branch"] = np.c_[ppctarget["branch"],
+                                    np.zeros((ppctarget["branch"].shape[0],
+                                              IDX.branch.QT - ppctarget["branch"].shape[1] + 1))]
+
+    # convert to internal indexing
+    ppctarget = opfcn.ext2int(ppctarget)
+    baseMVAt, bust, gent, brancht = \
+        ppctarget["baseMVA"], ppctarget["bus"], ppctarget["gen"], ppctarget["branch"]
+
+    # get bus index lists of each type of bus
+    # ref, pv, pq = putil.bustypes(bust, gent)
+
+    # generator info
+    ont = find(gent[:, IDX.gen.GEN_STATUS] > 0)  # which generators are on?
+    # gbust = gent[ont, IDX.gen.GEN_BUS].astype(int)  # what buses are they at?
+
+    # -----  run the power flow  -----
+    t0, _ = elapsed()  # start timer
+
+    # initial state
+    # V0    = np.ones((bus.shape[0])            ## flat start
+    V0 = busb[:, IDX.bus.VM] * np.exp((1j * deg2rad * busb[:, IDX.bus.VA]))
+    vcb = np.ones(V0.shape)    # create mask of voltage-controlled buses
+    vcb[pq] = 0     # exclude PQ buses
+    k = find(vcb[gbusb])    # in-service gens at v-c buses
+    V0[gbusb[k]] = genb[onb[k], IDX.gen.VG] / abs(V0[gbusb[k]]) * V0[gbusb[k]]
+
+    # build admittance matrices
+    Ybus, Yf, Yt = makeYbus(baseMVAb, busb, branchb)
+
+    # compute base case complex bus power injections (generation - load)
+    Sbusb = makeSbus(baseMVAb, busb, genb)
+    # compute target case complex bus power injections (generation - load)
+    Sbust = makeSbus(baseMVAt, bust, gent)
+
+    # scheduled transfer
+    Sxfr = Sbust - Sbusb
+
+    # Run the base case power flow solution
+    lam = 0
+    logger.info('Solve base case power flow')
+    V, success, iterations = newtonpf(Ybus, Sbusb, V0, ref, pv, pq, ppopt)
+
+    logger.info('Start full AC CPF')
+    logger.info('%2d: lambda = %6.3f, %2d Newton steps' % (0, 0, iterations))
+
+    lamprv = lam    # lam at previous step
+    Vprv = V    # V at previous step
+    continuation = 1
+    cont_steps = 0
+
+    # input args for callbacks
+    cb_data = dict(ppc_base=ppcbase, ppc_target=ppctarget,
+                   Sxfr=Sxfr, Ybus=Ybus, Yf=Yf, Yt=Yt,
+                   ref=ref, pv=pv, pq=pq, ppopt=ppopt)
+    cb_state = {}
+
+    # invoke callbacks
+    for k in range(len(callbacks)):
+        cb_state, _ = callbacks[k](cont_steps, V, lam, V, lam,
+                                   cb_data, cb_state, cb_args)
+
+    if np.linalg.norm(Sxfr) == 0:
+        logger.error('base case and target case have identical load and generation!')
+
+        continuation = 0
+        V0 = V
+        lam0 = lam
+
+    # tangent predictor z = [dx;dlam]
+    z = np.zeros(2*len(V)+1)
+    z[-1] = 1.0
+    while continuation:
+        cont_steps = cont_steps + 1
+        # prediction for next step
+        V0, lam0, z = cpf_predictor(V, lam, Ybus, Sxfr, pv, pq, step, z,
+                                    Vprv, lamprv, parameterization)
+
+        # save previous voltage, lambda before updating
+        Vprv = V
+        lamprv = lam
+
+        # correction
+        V, success, i, lam = cpf_corrector(Ybus, Sbusb, V0, ref, pv, pq,
+                                           lam0, Sxfr, Vprv, lamprv, z,
+                                           step, parameterization, ppopt)
+
+        if not success:
+            continuation = 0
+            logger.warning('%2d: lambda = %6.3f, corrector did not converge in %d iterations' % (
+                cont_steps, lam, i))
+            break
+
+        logger.info('%2d: lambda = %6.3f, %2d corrector Newton steps' %
+                    (cont_steps, lam, i))
+
+        # invoke callbacks
+        for k in range(len(callbacks)):
+            cb_state, _ = callbacks[k](cont_steps, V, lam, V0, lam0,
+                                       cb_data, cb_state, cb_args)
+
+        if isinstance(ppopt["CPF_STOP_AT"], str):
+            if ppopt["CPF_STOP_AT"].upper() == "FULL":
+                if abs(lam) < 1e-8:     # traced the full continuation curve
+                    logger.info('Reached steady state loading limit in %d continuation steps' % cont_steps)
+                    continuation = 0
+                elif lam < lamprv and lam - step < 0:    # next step will overshoot
+                    step = lam      # modify step-size
+                    parameterization = 1    # change to natural parameterization
+                    adapt_step = False      # disable step-adaptivity
+
+            else:   # == 'NOSE'
+                if lam < lamprv:    # reached the nose point
+                    logger.info('Reached steady state loading limit in %d continuation steps' % cont_steps)
+                    continuation = 0
+
+        else:
+            if lam < lamprv:
+                logger.info('Reached steady state loading limit in %d continuation steps' % cont_steps)
+                continuation = 0
+            elif abs(ppopt["CPF_STOP_AT"] - lam) < 1e-8:     # reached desired lambda
+                logger.info('Reached desired lambda %3.2f in %d continuation steps' % (
+                    ppopt["CPF_STOP_AT"], cont_steps))
+                continuation = 0
+            # will reach desired lambda in next step
+            elif lam + step > ppopt["CPF_STOP_AT"]:
+                step = ppopt["CPF_STOP_AT"] - lam   # modify step-size
+                parameterization = 1    # change to natural parameterization
+                adapt_step = False      # disable step-adaptivity
+
+        if adapt_step and continuation:
+            pvpq = np.r_[pv, pq]
+            # Adapt stepsize
+            cpf_error = np.linalg.norm(np.r_[np.angle(V[pq]), abs(
+                V[pvpq]), lam] - np.r_[np.angle(V0[pq]), abs(V0[pvpq]), lam0], inf)
+            if cpf_error < ppopt["CPF_ERROR_TOL"]:
+                # Increase stepsize
+                step = step * ppopt["CPF_ERROR_TOL"] / cpf_error
+                if step > ppopt["CPF_STEP_MAX"]:
+                    step = ppopt["CPF_STEP_MAX"]
+            else:
+                # decrese stepsize
+                step = step * ppopt["CPF_ERROR_TOL"] / cpf_error
+                if step < ppopt["CPF_STEP_MIN"]:
+                    step = ppopt["CPF_STEP_MIN"]
+
+    # invoke callbacks
+    if success:
+        cpf_results = {}
+        for k in range(len(callbacks)):
+            cb_state, cpf_results = callbacks[k](cont_steps, V, lam, V0, lam0,
+                                                 cb_data, cb_state, cb_args, results=cpf_results, is_final=True)
+    else:
+        cpf_results = {}
+        cpf_results["iterations"] = i
+
+    # update bus and gen matrices to reflect the loading and generation
+    # at the noise point
+    bust[:, IDX.bus.PD] = busb[:, IDX.bus.PD] + lam * (bust[:, IDX.bus.PD] - busb[:, IDX.bus.PD])
+    bust[:, IDX.bus.QD] = busb[:, IDX.bus.QD] + lam * (bust[:, IDX.bus.QD] - busb[:, IDX.bus.QD])
+    gent[:, IDX.gen.PG] = genb[:, IDX.gen.PG] + lam * (gent[:, IDX.gen.PG] - genb[:, IDX.gen.PG])
+
+    # update data matrices with solution
+    bust, gent, brancht = pfsoln(baseMVAt, bust, gent, brancht,
+                                 Ybus, Yf, Yt, V, ref, pv, pq)
+
+    ppctarget["et"] = elapsed(t0)
+    ppctarget["success"] = success
+
+    # -----  output results  -----
+    # convert back to original bus numbering & print results
+    ppctarget["bus"], ppctarget["gen"], ppctarget["branch"] = bust, gent, brancht
+    if success:
+        n = cpf_results["iterations"] + 1
+        cpf_results["V_p"] = opfcn.i2e_data(
+            ppctarget, cpf_results["V_p"], np.full((nb, n), nan), "bus", 0)
+        cpf_results["V_c"] = opfcn.i2e_data(
+            ppctarget, cpf_results["V_c"], np.full((nb, n), nan), "bus", 0)
+    results = opfcn.int2ext(ppctarget)
+    results["cpf"] = cpf_results
+
+    # zero out result fields of out-of-service gens & branches
+    if len(results["order"]["gen"]["status"]["off"]) > 0:
+        results["gen"][np.ix_(results["order"]["gen"]
+                              ["status"]["off"], [IDX.gen.QF, IDX.gen.QG])] = 0
+
+    if len(results["order"]["branch"]["status"]["off"]) > 0:
+        results["branch"][np.ix_(results["order"]["branch"]
+                                 ["status"]["off"], [IDX.branch.PF, IDX.branch.QF, IDX.branch.PT, IDX.branch.QT])] = 0
+
+    if ppopt["CPF_PLOT_LEVEL"]:
+        import matplotlib.pyplot as plt
+        plt.show()
+    sstats = dict(solver_name='PYPOWER', num_iters=cpf_results["iterations"])
+    return results, success, sstats
+
+
+def cpf_predictor(V, lam, Ybus, Sxfr, pv, pq,
+                  step, z, Vprv, lamprv, parameterization):
+    # sizes
+    pvpq = np.r_[pv, pq]
+    nb = len(V)
+    npv = len(pv)
+    npq = len(pq)
+
+    # compute Jacobian for the power flow equations
+    dSbus_dVm, dSbus_dVa = dSbus_dV(Ybus, V)
+
+    j11 = dSbus_dVa[np.array([pvpq]).T, pvpq].real
+    j12 = dSbus_dVm[np.array([pvpq]).T, pq].real
+    j21 = dSbus_dVa[np.array([pq]).T, pvpq].imag
+    j22 = dSbus_dVm[np.array([pq]).T, pq].imag
+
+    J = sp.vstack([
+        sp.hstack([j11, j12]),
+        sp.hstack([j21, j22])
+    ], format="csr")
+
+    dF_dlam = -np.r_[Sxfr[pvpq].real, Sxfr[pq].imag].reshape((-1, 1))
+    dP_dV, dP_dlam = cpf_p_jac(parameterization, z, V, lam, Vprv, lamprv, pv, pq)
+
+    # linear operator for computing the tangent predictor
+    J = sp.vstack([
+        sp.hstack([J, dF_dlam]),
+        sp.hstack([dP_dV, dP_dlam])
+    ], format="csr")
+
+    Vaprv = np.angle(V)
+    Vmprv = abs(V)
+
+    # compute normalized tangent predictor
+    s = np.zeros(npv+2*npq+1)
+    s[-1] = 1
+    z[np.r_[pvpq, nb+pq, 2*nb]] = sp.linalg.spsolve(J, s)
+    z = z / np.linalg.norm(z)
+
+    Va0 = Vaprv
+    Vm0 = Vmprv
+    lam0 = lam
+
+    # prediction for next step
+    Va0[pvpq] = Vaprv[pvpq] + step * z[pvpq]
+    Vm0[pq] = Vmprv[pq] + step * z[nb+pq]
+    lam0 = lam + step * z[2*nb]
+    V0 = Vm0 * np.exp((1j * Va0))
+
+    return V0, lam0, z
+
+
+def cpf_corrector(Ybus, Sbus, V0, ref, pv, pq,
+                  lam0, Sxfr, Vprv, lamprv, z,
+                  step, parameterization, ppopt):
+
+    # default arguments
+    if ppopt is None:
+        ppopt = ppoption(ppopt)
+
+    # options
+    verbose = ppopt["VERBOSE"]
+    tol = ppopt["PF_TOL"]
+    max_it = ppopt["PF_MAX_IT"]
+
+    # initialize
+    converged = 0
+    i = 0
+    V = V0
+    Va = np.angle(V)
+    Vm = abs(V)
+    lam = lam0
+
+    # set up indexing for updating V
+    pvpq = np.r_[pv, pq]
+    npv = len(pv)
+    npq = len(pq)
+    nb = len(V)
+    j1 = 0
+    j2 = npv    # j1:j2 - V angle of pv buses
+    j3 = j2
+    j4 = j2 + npq   # j1:j2 - V angle of pv buses
+    j5 = j4
+    j6 = j4 + npq   # j5:j6 - V mag of pq buses
+    j7 = j6
+    j8 = j6 + 1    # j7:j8 - lambda
+
+    # evaluate F(x0, lam0), including Sxfr transfer/loading
+
+    mis = V * np.conj(Ybus.dot(V)) - Sbus - lam*Sxfr
+    F = np.r_[mis[pvpq].real, mis[pq].imag]
+
+    # evaluate P(x0, lambda0)
+    P = cpf_p(parameterization, step, z, V, lam, Vprv, lamprv, pv, pq)
+
+    # augment F(x, lambda) with P(x, lambda)
+    F = np.r_[F, P]
+
+    # check tolerance
+    normF = np.linalg.norm(F, inf)
+    logger.debug('CPF correction')
+    logger.debug('%2d: |F(x)| = %.10g', i, normF)
+    if normF < tol:
+        converged = 1
+
+    # do Newton iterations
+    while not converged and i < max_it:
+        # update iteration counter
+        i = i + 1
+
+        # evaluate Jacobian
+        dSbus_dVm, dSbus_dVa = dSbus_dV(Ybus, V)
+
+        j11 = dSbus_dVa[np.array([pvpq]).T, pvpq].real
+        j12 = dSbus_dVm[np.array([pvpq]).T, pq].real
+        j21 = dSbus_dVa[np.array([pq]).T, pvpq].imag
+        j22 = dSbus_dVm[np.array([pq]).T, pq].imag
+
+        J = sp.vstack([
+            sp.hstack([j11, j12]),
+            sp.hstack([j21, j22])
+        ], format="csr")
+
+        dF_dlam = -np.r_[Sxfr[pvpq].real, Sxfr[pq].imag].reshape((-1, 1))
+        dP_dV, dP_dlam = cpf_p_jac(parameterization, z, V, lam, Vprv, lamprv, pv, pq)
+
+        # augment J with real/imag -Sxfr and z^T
+        J = sp.vstack([
+            sp.hstack([J, dF_dlam]),
+            sp.hstack([dP_dV, dP_dlam])
+        ], format="csr")
+
+        # compute update step
+        dx = -1 * sp.linalg.spsolve(J, F)
+
+        # update voltage
+        if npv:
+            Va[pv] = Va[pv] + dx[j1:j2]
+        if npq:
+            Va[pq] = Va[pq] + dx[j3:j4]
+            Vm[pq] = Vm[pq] + dx[j5:j6]
+        V = Vm * np.exp((1j * Va))
+        Vm = abs(V)
+        Va = np.angle(V)
+
+        # update lambda
+        lam = lam + dx[j7:j8]
+
+        # evalute F(x, lam)
+        mis = V * np.conj(Ybus.dot(V)) - Sbus - lam*Sxfr
+        F = np.r_[mis[pv].real, mis[pq].real, mis[pq].imag]
+
+        # evaluate P(x, lambda)
+        P = cpf_p(parameterization, step, z, V, lam, Vprv, lamprv, pv, pq)
+
+        # augment F(x, lambda) with P(x, lambda)
+        F = np.r_[F, P]
+
+        # check for convergence
+        normF = np.linalg.norm(F, inf)
+        logger.debug('%2d: |F(x)| = %.10g', i, normF)
+
+        if normF < tol:
+            converged = 1
+
+    if not converged:
+        logger.error('Newton''s method corrector did not converge in %d iterations.' % i)
+
+    return V, converged, i, lam
+
+
+def cpf_p_jac(parameterization, z, V, lam, Vprv, lamprv, pv, pq):
+    if parameterization == 1:
+        npv = len(pv)
+        npq = len(pq)
+        dP_dV = np.zeros(npv+2*npq)
+        if lam >= lamprv:
+            dP_dlam = 1.0
+        else:
+            dP_dlam = -1.0
+
+    elif parameterization == 2:
+        pvpq = np.r_[pv, pq]
+
+        Va = np.angle(V)
+        Vm = abs(V)
+        Vaprv = np.angle(Vprv)
+        Vmprv = abs(Vprv)
+        dP_dV = 2 * (np.r_[Va[pvpq], Vm[pq]] -
+                     np.r_[Vaprv[pvpq], Vmprv[pq]])
+        if lam == lamprv:
+            dP_dlam = 1.0
+        else:
+            dP_dlam = 2 * (lam - lamprv)
+
+    elif parameterization == 3:
+        nb = len(V)
+        dP_dV = z[np.r_[pv, pq, nb+pq]]
+        dP_dlam = z[2 * nb]
+
+    return dP_dV, dP_dlam
+
+
+def cpf_p(parameterization, step, z, V, lam, Vprv, lamprv, pv, pq):
+    # evaluate P(x0, lambda0)
+    if parameterization == 1:
+        if lam >= lamprv:
+            P = lam - lamprv - step
+        else:
+            P = lamprv - lam - step
+
+    elif parameterization == 2:
+        pvpq = np.r_[pv, pq]
+
+        Va = np.angle(V)
+        Vm = abs(V)
+        Vaprv = np.angle(Vprv)
+        Vmprv = abs(Vprv)
+        P = sum(np.square(np.r_[Va[pvpq], Vm[pq], lam] -
+                          np.r_[Vaprv[pvpq], Vmprv[pq], lamprv])) - np.square(step)
+
+    elif parameterization == 3:
+        pvpq = np.r_[pv, pq]
+
+        nb = len(V)
+        Va = np.angle(V)
+        Vm = abs(V)
+        Vaprv = np.angle(Vprv)
+        Vmprv = abs(Vprv)
+        P = np.dot(z[np.r_[pvpq, nb+pq, 2*nb]].reshape((1, -1)),
+                   np.array(np.r_[Va[pvpq], Vm[pq], lam] - np.r_[Vaprv[pvpq], Vmprv[pq], lamprv]).T) - step
+
+    return P

ams/pypower/routines/cpf_callbacks.py

@@ -0,0 +1,114 @@
+"""Callback functions for CPF
+"""
+
+from numpy import c_, r_, amax, argmax, array
+from scipy.sparse import issparse
+
+import sys
+import os
+
+
+def cpf_default_callback(k, V_c, lam_c, V_p, lam_p, cb_data, cb_state, cb_args, results=None, is_final=False):
+    """Default callback function for CPF
+    """
+
+    # initialize plotting options
+    plot_level = cb_data["ppopt"]["CPF_PLOT_LEVEL"]
+    plot_bus = cb_data["ppopt"]["CPF_PLOT_BUS"]
+
+    if plot_level:
+        try:
+            import matplotlib.pyplot as plt
+        except ImportError:
+            logger.debug(
+                "Error: Please install \"Matpotlib\" package for plotting function!\n")
+
+        if plot_bus == '':  # no bus specified
+            if len(cb_data["Sxfr"][cb_data["pq"]]) != 0:
+                idx = argmax(abs(cb_data["Sxfr"][cb_data["pq"]]), axis=0)
+                idx = cb_data["pq"][idx]
+            else:   # or bus 1 if there are none
+                idx = 0
+            idx_e = int(cb_data["ppc_target"]["order"]["bus"]["i2e"][idx])
+
+        else:
+            idx_e = plot_bus
+            if issparse(cb_data["ppc_target"]["order"]["bus"]["e2i"][idx_e]):
+                idx = int(cb_data["ppc_target"]["order"]
+                          ["bus"]["e2i"][idx_e].toarray())
+            else:
+                idx = int(cb_data["ppc_target"]["order"]["bus"]["e2i"][idx_e])
+
+            if idx == 0:
+                logger.debug(
+                    "cpf_default_callback: %d is not a valid bus number for PPOPT[\"CPF_PLOT_BUS\"]\n" % idx_e)
+
+    # -----  FINAL call  -----
+    if is_final:
+        # assemble results struct
+        results = cb_state
+        results["max_lam"] = amax(cb_state["lam_c"])
+        results["iterations"] = k
+
+        # finish final lambda-V nose curve plot
+        if plot_level:
+            plt.plot(cb_state["lam_c"],
+                     abs(cb_state["V_c"][idx, :]), '-', color=(0.25, 0.25, 1))
+            plt.axis([0, amax([1, amax(cb_state["lam_p"]), amax(cb_state["lam_c"])])*1.05, 0,
+                      amax([1, amax(abs(cb_state["V_p"][idx])), amax(abs(cb_state["V_c"][idx]))*1.05])])
+            plt.pause(sys.float_info.epsilon)
+            plt.ioff()
+
+    elif k == 0:
+        # initialize state
+        cb_state = {
+            "V_p": V_p,
+            "lam_p": lam_p,
+            "V_c": V_c,
+            "lam_c": lam_c,
+            "iterations": 0
+        }
+
+        # initialize lambda-V nose curve plot
+        if plot_level:
+            plt.ion()
+            plt.plot(cb_state["lam_p"],
+                     abs(cb_state["V_p"][idx]), '-', color=(0.25, 0.25, 1))
+            plt.title('Voltage at Bus %d' % idx_e)
+            plt.xlabel('$\\lambda$')
+            plt.ylabel('Voltage Magnitude')
+            plt.axis([0, amax([1, amax(cb_state["lam_p"]), amax(cb_state["lam_c"])])*1.05, 0,
+                      amax([1, amax(abs(cb_state["V_p"][idx])), amax(abs(cb_state["V_c"][idx]))*1.05])])
+            plt.pause(sys.float_info.epsilon)
+
+    # -----  ITERATION call  -----
+    else:
+        # update state
+        cb_state["V_p"] = c_[cb_state["V_p"], V_p]
+        cb_state["lam_p"] = r_[cb_state["lam_p"], lam_p]
+        cb_state["V_c"] = c_[cb_state["V_c"], V_c]
+        cb_state["lam_c"] = r_[cb_state["lam_c"], lam_c]
+        cb_state["iterations"] = k
+
+        # plot single step of the lambda-V nose curve
+        if plot_level > 1:
+            plt.plot(r_[cb_state["lam_c"][k-1], cb_state["lam_p"][k]],
+                     r_[abs(cb_state["V_c"][idx, k-1]),
+                        abs(cb_state["V_p"][idx, k])],
+                     '-', color=0.85*array([1, 0.75, 0.75]))
+            plt.plot(r_[cb_state["lam_p"][k], cb_state["lam_c"][k]],
+                     r_[abs(cb_state["V_p"][idx, k]),
+                        abs(cb_state["V_c"][idx, k])],
+                     '-', color=0.85*array([0.75, 1, 0.75]))
+            plt.plot(cb_state["lam_p"][k], abs(cb_state["V_p"][idx, k]),
+                     'x', color=0.85*array([1, 0.75, 0.75]))
+            plt.plot(cb_state["lam_c"][k], abs(
+                cb_state["V_c"][idx, k]), '-o', markerfacecolor='none', color=(0.25, 0.25, 1))
+            plt.axis([0, amax([1, amax(cb_state["lam_p"]), amax(cb_state["lam_c"])])*1.05, 0,
+                      amax([1, amax(abs(cb_state["V_p"][idx])), amax(abs(cb_state["V_c"][idx]))*1.05])])
+            plt.pause(sys.float_info.epsilon)
+            
+        if plot_level > 2:
+            os.system("pause")
+
+    return cb_state, results