def load_nidata(self):
nidata = loadmat(os.path.join(self.raw_dir, "nidata.mat"), squeeze_me=True)
self.time_ni = nidata["t"]
self.sr_ni = (1.0/(self.time_ni[1] - self.time_ni[0]))
if "carriage_pos" in nidata:
self.lin_enc = True
self.carriage_pos = nidata["carriage_pos"]
self.tow_speed_ni = fdiff.second_order_diff(self.carriage_pos, self.time_ni)
self.tow_speed_ni = ts.smooth(self.tow_speed_ni, 8)
self.tow_speed_ref = self.tow_speed_ni
else:
self.lin_enc = False
self.tow_speed_ref = self.tow_speed_nom
self.torque = nidata["torque_trans"]
self.torque_arm = nidata["torque_arm"]
self.drag = nidata["drag_left"] + nidata["drag_right"]
# Remove offsets from drag, not torque
t0 = 2
self.drag = self.drag - np.mean(self.drag[0:self.sr_ni*t0])
# Compute RPM and omega
self.angle = nidata["turbine_angle"]
self.rpm_ni = fdiff.second_order_diff(self.angle, self.time_ni)/6.0
self.rpm_ni = ts.smooth(self.rpm_ni, 8)
self.omega_ni = self.rpm_ni*2*np.pi/60.0
self.omega = self.omega_ni
self.tow_speed = self.tow_speed_ref
def EveryNCallback_py(taskHandle, everyNsamplesEventType, nSamples,
callbackData_ptr):
"""Function called every N samples"""
callbackdata = daqmx.get_callbackdata_from_id(callbackData_ptr)
data, npoints = daqmx.ReadAnalogF64(taskHandle, self.nsamps,
10.0, daqmx.Val_GroupByChannel, self.nsamps,
len(self.analogchans))
callbackdata.extend(data.tolist())
self.data["torque_trans"] = np.append(self.data["torque_trans"],
data[:,0], axis=0)
self.data["torque_arm"] = np.append(self.data["torque_arm"],
data[:,1], axis=0)
self.data["drag_left"] = np.append(self.data["drag_left"],
data[:,2], axis=0)
self.data["drag_right"] = np.append(self.data["drag_right"],
data[:,3], axis=0)
self.data["time"] = np.arange(len(self.data["torque_trans"]),
dtype=float)/self.sr
carpos, cpoints = daqmx.ReadCounterF64(self.carpostask,
self.nsamps, 10.0,
self.nsamps)
self.data["carriage_pos"] = np.append(self.data["carriage_pos"],
carpos)
turbang, cpoints = daqmx.ReadCounterF64(self.turbangtask,
self.nsamps, 10.0,
self.nsamps)
self.data["turbine_angle"] = np.append(self.data["turbine_angle"],
turbang)
self.data["turbine_rpm"] \
= ts.smooth(fdiff.second_order_diff(self.data["turbine_angle"],
self.data["time"])/6.0, 8)
return 0 # The function should return an integer
def process_tare_torque(nrun, plot=False):
"""Process a single tare torque run."""
print("Processing tare torque run", nrun)
times = {0 : (35, 86),
1 : (12, 52),
2 : (11, 32),
3 : (7, 30)}
nidata = loadhdf("Data/Raw/Tare-torque/" + str(nrun) + "/nidata.h5")
# Compute RPM
time_ni = nidata["time"]
angle = nidata["turbine_angle"]
rpm_ni = fdiff.second_order_diff(angle, time_ni)/6.0
rpm_ni = ts.smooth(rpm_ni, 8)
try:
t1, t2 = times[nrun]
except KeyError:
t1, t2 = times[3]
meanrpm, _ = ts.calcstats(rpm_ni, t1, t2, 2000)
torque = nidata["torque_trans"]
meantorque, _ = ts.calcstats(torque, t1, t2, 2000)
print("Tare torque =", meantorque, "Nm at", meanrpm, "RPM")
if plot:
plt.figure()
plt.plot(time_ni, torque)
plt.xlabel("Time (s)")
plt.ylabel("Torque (Nm)")
plt.tight_layout()
plt.show()
return meanrpm, -meantorque
def calc_perf(theta_0=360,
plot=False,
verbose=True,
inertial=False,
export_csv=True):
t, torque, drag = foampy.load_all_torque_drag()
_t, theta, omega = foampy.load_theta_omega(t_interp=t)
reached_theta_0 = True
if theta.max() < theta_0:
theta_0 = 1
reached_theta_0 = False
# Compute tip speed ratio
tsr = omega * R / U_infty
# Compute mean TSR
meantsr = np.mean(tsr[theta >= theta_0])
if inertial:
inertia = 3 # guess from SolidWorks model
inertial_torque = inertia * fdiff.second_order_diff(omega, t)
torque -= inertial_torque
# Compute power coefficient
power = torque * omega
cp = power / (0.5 * rho * area * U_infty**3)
meancp = np.mean(cp[theta >= theta_0])
# Compute drag coefficient
cd = drag / (0.5 * rho * area * U_infty**2)
meancd = np.mean(cd[theta >= theta_0])
if export_csv:
df = pd.DataFrame()
df["theta_deg"] = theta
df["tsr"] = tsr
df["cp"] = cp
df["cd"] = cd
df.to_csv("processed/perf.csv", index=False)
if verbose:
print("Performance from {:.1f}--{:.1f} degrees:".format(
theta_0, theta.max()))
print("Mean TSR = {:.3f}".format(meantsr))
print("Mean C_P = {:.3f}".format(meancp))
print("Mean C_D = {:.3f}".format(meancd))
if plot:
plt.close('all')
plt.plot(theta[5:], cp[5:])
plt.title(r"$\lambda = %1.1f$" % meantsr)
plt.xlabel(r"$\theta$ (degrees)")
plt.ylabel(r"$C_P$")
#plt.ylim((0, 1.0))
plt.tight_layout()
plt.show()
if reached_theta_0:
return {"C_P": meancp, "C_D": meancd, "TSR": meantsr}
else:
return {"C_P": "nan", "C_D": "nan", "TSR": "nan"}
def load_nidata(self):
nidata = loadhdf(os.path.join(self.raw_dir, "nidata.h5"))
self.time_ni = nidata["time"]
self.sr_ni = (1.0/(self.time_ni[1] - self.time_ni[0]))
self.carriage_pos = nidata["carriage_pos"]
self.tow_speed_ni = fdiff.second_order_diff(self.carriage_pos, self.time_ni)
self.tow_speed_ni = ts.smooth(self.tow_speed_ni, 100)
self.tow_speed_ref = self.tow_speed_ni.copy()
self.tow_speed_ref[np.abs(self.tow_speed_ref) < 0.01] = np.nan
self.torque = nidata["torque_trans"]
self.torque_arm = nidata["torque_arm"]
self.drag = nidata["drag_left"] + nidata["drag_right"]
# Remove offsets from drag, not torque
t0 = 2
self.drag = self.drag - np.mean(self.drag[0:self.sr_ni*t0])
# Compute RPM and omega
self.angle = nidata["turbine_angle"]
self.rpm_ni = fdiff.second_order_diff(self.angle, self.time_ni)/6.0
self.rpm_ni = ts.smooth(self.rpm_ni, 8)
self.omega_ni = self.rpm_ni*2*np.pi/60.0
self.omega = self.omega_ni
self.tow_speed = self.tow_speed_ref
def process_strut_torque(nrun, zero_torque=0.0, plot=False, covers=False,
verbose=False):
"""Process a single strut torque run."""
testplan = pd.read_csv("Config/Test plan/Strut-torque.csv",
index_col="run")
ref_speed = testplan.ref_speed.iloc[nrun]
tsr_nom = testplan.tsr.iloc[nrun]
revs = testplan.revs.iloc[nrun]
rpm_nom = tsr_nom*ref_speed/R/(2*np.pi)*60
dur = revs/rpm_nom*60
if covers:
if verbose:
print("Processing strut torque with covers run", nrun)
nidata = loadhdf("Data/Raw/Strut-torque-covers/" + str(nrun) + \
"/nidata.h5")
else:
if verbose:
print("Processing strut torque run", nrun)
nidata = loadhdf("Data/Raw/Strut-torque/" + str(nrun) + "/nidata.h5")
# Compute RPM
time_ni = nidata["time"]
angle = nidata["turbine_angle"]
rpm_ni = fdiff.second_order_diff(angle, time_ni)/6.0
rpm_ni = ts.smooth(rpm_ni, 8)
t1, t2 = 9, dur
meanrpm, _ = ts.calcstats(rpm_ni, t1, t2, 2000)
torque = nidata["torque_trans"]
torque += calc_tare_torque(rpm_ni)
meantorque, _ = ts.calcstats(torque, t1, t2, 2000)
tsr_ref = meanrpm/60.0*2*np.pi*R/ref_speed
if verbose:
print("Reference TSR =", np.round(tsr_ref, decimals=4))
print("Strut torque =", meantorque, "Nm at", meanrpm, "RPM")
if plot:
plt.figure()
plt.plot(time_ni, torque)
plt.xlabel("Time (s)")
plt.ylabel("Torque (Nm)")
plt.tight_layout()
plt.show()
meantorque -= zero_torque
ct = meantorque/(0.5*rho*A*R*ref_speed**2)
cp = ct*tsr_ref
summary = pd.Series()
summary["run"] = nrun
summary["tsr_ref"] = tsr_ref
summary["cp"] = cp
summary["mean_torque"] = meantorque
summary["mean_rpm"] = meanrpm
return summary
def calc_perf(theta_0=360, plot=False, verbose=True, inertial=False):
t, torque, drag = foampy.load_all_torque_drag()
_t, theta, omega = foampy.load_theta_omega(t_interp=t)
reached_theta_0 = True
if theta.max() < theta_0:
theta_0 = 1
reached_theta_0 = False
# Compute tip speed ratio
tsr = omega*R/U_infty
# Compute mean TSR
meantsr = np.mean(tsr[theta >= theta_0])
if inertial:
inertia = 3 # guess from SolidWorks model
inertial_torque = inertia*fdiff.second_order_diff(omega, t)
torque -= inertial_torque
# Compute power coefficient
power = torque*omega
cp = power/(0.5*rho*area*U_infty**3)
meancp = np.mean(cp[theta >= theta_0])
# Compute drag coefficient
cd = drag/(0.5*rho*area*U_infty**2)
meancd = np.mean(cd[theta >= theta_0])
if verbose:
print("Performance from {:.1f}--{:.1f} degrees:".format(theta_0,
theta.max()))
print("Mean TSR = {:.3f}".format(meantsr))
print("Mean C_P = {:.3f}".format(meancp))
print("Mean C_D = {:.3f}".format(meancd))
if plot:
plt.close('all')
plt.plot(theta[5:], cp[5:])
plt.title(r"$\lambda = %1.1f$" %meantsr)
plt.xlabel(r"$\theta$ (degrees)")
plt.ylabel(r"$C_P$")
#plt.ylim((0, 1.0))
plt.tight_layout()
plt.show()
if reached_theta_0:
return {"C_P" : meancp,
"C_D" : meancd,
"TSR" : meantsr}
else:
return {"C_P" : "nan",
"C_D" : "nan",
"TSR" : "nan"}
def EveryNCallback_py(taskHandle, everyNsamplesEventType, nSamples,
callbackData_ptr):
"""Function called every N samples"""
callbackdata = daqmx.get_callbackdata_from_id(callbackData_ptr)
data, npoints = daqmx.ReadAnalogF64(taskHandle, self.nsamps, 10.0,
daqmx.Val_GroupByChannel,
self.nsamps,
len(self.analogchans))
callbackdata.extend(data.tolist())
self.data["torque_trans"] = np.append(self.data["torque_trans"],
data[:, 0],
axis=0)
self.data["torque_arm"] = np.append(self.data["torque_arm"],
data[:, 1],
axis=0)
self.data["drag_left"] = np.append(self.data["drag_left"],
data[:, 2],
axis=0)
self.data["drag_right"] = np.append(self.data["drag_right"],
data[:, 3],
axis=0)
self.data["t"] = np.arange(len(self.data["torque_trans"]),
dtype=float) / self.sr
carpos, cpoints = daqmx.ReadCounterF64(self.carpostask,
self.nsamps, 10.0,
self.nsamps)
self.data["carriage_pos"] = np.append(self.data["carriage_pos"],
carpos)
turbang, cpoints = daqmx.ReadCounterF64(self.turbangtask,
self.nsamps, 10.0,
self.nsamps)
self.data["turbine_angle"] = np.append(self.data["turbine_angle"],
turbang)
self.data["turbine_rpm"] \
= ts.smooth(fdiff.second_order_diff(self.data["turbine_angle"],
self.data["t"])/6.0, 50)
return 0 # The function should return an integer