def create_model(Bias_layer,
low,
up,
F,
beta,
gamma,
Co,
m,
a0RNN,
batch_input_shape,
selectbias,
selectidx,
selectdk,
selectwk,
myDtype,
return_sequences=False,
unroll=False):
batch_adjusted_shape = (batch_input_shape[2] + 1, ) #Adding state
placeHolder = Input(shape=(batch_input_shape[2] + 1, )) #Adding state
filterBias = inputsSelection(batch_adjusted_shape, selectbias)(placeHolder)
filterSig = inputsSelection(batch_adjusted_shape, selectidx)(placeHolder)
filterdK = inputsSelection(batch_adjusted_shape, selectdk)(placeHolder)
filterda = inputsSelection(batch_adjusted_shape, selectwk)(placeHolder)
MLP_min = low
MLP_range = up - low
Bias_layer = Bias_layer(filterBias)
MLP = Lambda(lambda x: ((x * MLP_range) + MLP_min))(Bias_layer)
Filter = Lambda(lambda x: sign(x))(filterSig)
Bias_filtered_layer = Multiply()([MLP, Filter])
dk_input_shape = filterdK.get_shape()
dkLayer = StressIntensityRange(input_shape=dk_input_shape,
dtype=myDtype,
trainable=False)
dkLayer.build(input_shape=dk_input_shape)
dkLayer.set_weights([np.asarray([F], dtype=dkLayer.dtype)])
dkLayer = dkLayer(filterdK)
wmInput = Concatenate(axis=-1)([dkLayer, filterda])
wm_input_shape = wmInput.get_shape()
wmLayer = WalkerModel(input_shape=wm_input_shape,
dtype=myDtype,
trainable=False)
wmLayer.build(input_shape=wm_input_shape)
wmLayer.set_weights(
[np.asarray([beta, gamma, Co, m], dtype=wmLayer.dtype)])
wmLayer = wmLayer(wmInput)
da_layer = Add()([Bias_filtered_layer, wmLayer])
functionalModel = Model(inputs=[placeHolder], outputs=[da_layer])
"-------------------------------------------------------------------------"
CDMCellHybrid = CumulativeDamageCell(model=functionalModel,
batch_input_shape=batch_input_shape,
dtype=myDtype,
initial_damage=a0RNN)
CDMRNNhybrid = RNN(cell=CDMCellHybrid,
return_sequences=return_sequences,
return_state=False,
batch_input_shape=batch_input_shape,
unroll=unroll)
model = Sequential()
model.add(CDMRNNhybrid)
model.compile(loss=mape, optimizer=RMSprop(1e-11), metrics=['mse'])
return model
def create_pinn_model(a,
b,
Pu,
grid_array_aSKF,
bounds_aSKF,
table_shape_aSKF,
grid_array_kappa,
bounds_kappa,
table_shape_kappa,
grid_array_etac,
bounds_etac,
table_shape_etac,
d0RNN,
batch_input_shape,
selectdKappa,
selectCycle,
selectLoad,
selectBTemp,
myDtype,
return_sequences=False,
unroll=False):
batch_adjusted_shape = (batch_input_shape[2] + 1, ) #Adding states
placeHolder = Input(shape=(batch_input_shape[2] + 1, )) #Adding states
filterdKappaLayer = inputsSelection(batch_adjusted_shape,
selectdKappa)(placeHolder)
filterCycleLayer = inputsSelection(batch_adjusted_shape,
selectCycle)(placeHolder)
filterLoadLayer = inputsSelection(batch_adjusted_shape,
selectLoad)(placeHolder)
filterBTempLayer = inputsSelection(batch_adjusted_shape,
selectBTemp)(placeHolder)
physicalSpaceInvLoadLayer = Lambda(lambda x: (1 / (10**x)))(
filterLoadLayer)
xvalKappaLayer = Concatenate(axis=-1)(
[filterBTempLayer, filterdKappaLayer])
kappaLayer = TableInterpolation(table_shape=table_shape_kappa,
dtype=myDtype,
trainable=False)
kappaLayer.build(input_shape=xvalKappaLayer.shape)
kappaLayer.set_weights([grid_array_kappa, bounds_kappa])
kappaLayer = kappaLayer(xvalKappaLayer)
xvalEtacLayer = Concatenate(axis=-1)([kappaLayer, filterdKappaLayer])
etacLayer = TableInterpolation(table_shape=table_shape_etac,
dtype=myDtype,
trainable=False)
etacLayer.build(input_shape=xvalEtacLayer.shape)
etacLayer.set_weights([grid_array_etac, bounds_etac])
etacLayer = etacLayer(xvalEtacLayer)
xvalLayer1 = Lambda(lambda x: Pu * x)(etacLayer)
xvalLayer2 = Multiply()([xvalLayer1, physicalSpaceInvLoadLayer])
xvalLayer = Concatenate(axis=-1)([xvalLayer2, kappaLayer])
aSKFLayer = TableInterpolation(table_shape=table_shape_aSKF,
dtype=myDtype,
trainable=False)
aSKFLayer.build(input_shape=xvalLayer.shape)
aSKFLayer.set_weights([grid_array_aSKF, bounds_aSKF])
aSKFLayer = aSKFLayer(xvalLayer)
inverseaSKFLayer = Lambda(lambda x: (1 / x))(aSKFLayer)
sn_input_shape = (batch_input_shape[0], batch_input_shape[2])
SNLayer = SNCurve(input_shape=sn_input_shape,
dtype=myDtype,
trainable=False)
SNLayer.build(input_shape=sn_input_shape)
SNLayer.set_weights([np.asarray([a, b], dtype=SNLayer.dtype)])
SNLayer = SNLayer(filterLoadLayer)
multiplyLayer1 = Multiply()([SNLayer, filterCycleLayer])
multiplyLayer2 = Multiply()([multiplyLayer1, inverseaSKFLayer])
functionalModel = Model(inputs=[placeHolder], outputs=[multiplyLayer2])
"-------------------------------------------------------------------------"
CDMCellHybrid = CumulativeDamageCell(model=functionalModel,
batch_input_shape=batch_input_shape,
dtype=myDtype,
initial_damage=d0RNN)
CDMRNNhybrid = RNN(cell=CDMCellHybrid,
return_sequences=return_sequences,
return_state=False,
batch_input_shape=batch_input_shape,
unroll=unroll)
model = Sequential()
model.add(CDMRNNhybrid)
model.compile(loss='mse', optimizer=RMSprop(5e-4), metrics=['mae'])
return model
def create_physics_model(radius, length, thickness, poisson, inertia,
storage_data, storage_bounds, storage_table_shape,
loss_data, loss_bounds, loss_table_shape, select_freq,
select_temp, batch_input_shape, myDtype):
batch_adjusted_shape = (batch_input_shape[1], )
inputLayer = Input(shape=(batch_input_shape[1], ))
freqLayer = inputsSelection(batch_adjusted_shape, select_freq)(inputLayer)
tempLayer = inputsSelection(batch_adjusted_shape, select_temp)(inputLayer)
omegaLayer = Lambda(lambda x: 2 * np.pi * x)(freqLayer)
moduliInputLayer = Concatenate(axis=-1)([tempLayer, freqLayer])
storageModulusLayer = TableInterpolation(table_shape=storage_table_shape,
dtype=myDtype,
trainable=False)
storageModulusLayer.build(input_shape=moduliInputLayer.shape)
storageModulusLayer.set_weights([storage_data, storage_bounds])
storageModulusLayer = storageModulusLayer(moduliInputLayer)
lossModulusLayer = TableInterpolation(table_shape=loss_table_shape,
dtype=myDtype,
trainable=False)
lossModulusLayer.build(input_shape=moduliInputLayer.shape)
lossModulusLayer.set_weights([loss_data, loss_bounds])
lossModulusLayer = lossModulusLayer(moduliInputLayer)
stiffnessLayer = Stiffness(input_shape=storageModulusLayer.shape,
dtype=myDtype,
trainable=False)
stiffnessLayer.build(input_shape=storageModulusLayer.shape)
stiffnessLayer.set_weights(
[np.asarray([poisson], dtype=stiffnessLayer.dtype)])
stiffnessLayer = stiffnessLayer(storageModulusLayer)
artificialDelta1Layer = Lambda(lambda x: x[0] - x[1] * 10 + x[0] / x[1])(
[freqLayer, tempLayer])
modifiedStiffnessLayer = Lambda(lambda x: x[0] + x[1])(
[stiffnessLayer, artificialDelta1Layer])
dampingInputLayer = Concatenate(axis=-1)(
[modifiedStiffnessLayer, storageModulusLayer, lossModulusLayer])
dampingLayer = Damping(input_shape=dampingInputLayer.shape,
dtype=myDtype,
trainable=False)
dampingLayer.build(input_shape=dampingInputLayer.shape)
dampingLayer.set_weights([np.asarray([inertia], dtype=dampingLayer.dtype)])
dampingLayer = dampingLayer(dampingInputLayer)
artificialDelta2Layer = Lambda(lambda x: x[0] / 250 - x[1] / 50 + x[0] /
(10 * x[1]))([freqLayer, tempLayer])
modifiedDampingLayer = Lambda(lambda x: x[0] + x[1])(
[dampingLayer, artificialDelta2Layer])
FRFAmpInputLayer = Concatenate(axis=-1)(
[omegaLayer, modifiedStiffnessLayer, modifiedDampingLayer])
FRFAmpLayer = FRFAmplitude(input_shape=FRFAmpInputLayer.shape,
dtype=myDtype,
trainable=False)
FRFAmpLayer.build(input_shape=FRFAmpInputLayer.shape)
FRFAmpLayer.set_weights([np.asarray([inertia], dtype=FRFAmpLayer.dtype)])
FRFAmpLayer = FRFAmpLayer(FRFAmpInputLayer)
functionalModel = Model(inputs=[inputLayer], outputs=[FRFAmpLayer])
functionalModel.compile(
loss='mean_squared_error',
optimizer=Adam(1e-2),
metrics=['mean_absolute_error', 'mean_squared_error'])
return functionalModel
def create_calibration_model(
storage_coefs, loss_coefs, radius, length, thickness, poisson, inertia,
storage_data, storage_bounds, storage_table_shape, loss_data,
loss_bounds, loss_table_shape, delta_stiffness_mlp, delta_damping_mlp,
stiffness_low, stiffness_up, damping_low, damping_up, input_min,
input_range, select_freq, select_temp, batch_input_shape, myDtype):
batch_adjusted_shape = (batch_input_shape[1], )
inputLayer = Input(shape=(batch_input_shape[1], ))
normalizedInputLayer = Lambda(
lambda x, input_min=input_min, input_range=input_range:
(x - input_min) / input_range)(inputLayer)
freqLayer = inputsSelection(batch_adjusted_shape, select_freq)(inputLayer)
tempLayer = inputsSelection(batch_adjusted_shape, select_temp)(inputLayer)
omegaLayer = Lambda(lambda x: 2 * np.pi * x)(freqLayer)
moduliInputLayer = Concatenate(axis=-1)([tempLayer, freqLayer])
storageModulusLayer = TableInterpolation(table_shape=storage_table_shape,
dtype=myDtype,
trainable=False)
storageModulusLayer.build(input_shape=moduliInputLayer.shape)
storageModulusLayer.set_weights([storage_data, storage_bounds])
storageModulusLayer = storageModulusLayer(moduliInputLayer)
lossModulusLayer = TableInterpolation(table_shape=loss_table_shape,
dtype=myDtype,
trainable=False)
lossModulusLayer.build(input_shape=moduliInputLayer.shape)
lossModulusLayer.set_weights([loss_data, loss_bounds])
lossModulusLayer = lossModulusLayer(moduliInputLayer)
stiffnessLayer = Stiffness(input_shape=storageModulusLayer.shape,
dtype=myDtype,
trainable=True,
kernel_constraint=MinMaxNorm(min_value=-1.0,
max_value=0.5,
rate=1.0))
stiffnessLayer.build(input_shape=storageModulusLayer.shape)
stiffnessLayer.set_weights(
[np.asarray([poisson], dtype=stiffnessLayer.dtype)])
stiffnessLayer = stiffnessLayer(storageModulusLayer)
deltaStiffnessLayer = delta_stiffness_mlp(normalizedInputLayer)
scaledDeltaStiffnessLayer = Lambda(
lambda x, stiffness_low=stiffness_low, stiffness_up=stiffness_up: x *
(stiffness_up - stiffness_low) + stiffness_low)(deltaStiffnessLayer)
correctedStiffnessLayer = Lambda(lambda x: x[0] + x[1])(
[stiffnessLayer, scaledDeltaStiffnessLayer])
dampingInputLayer = Concatenate(axis=-1)(
[correctedStiffnessLayer, storageModulusLayer, lossModulusLayer])
dampingLayer = Damping(input_shape=dampingInputLayer.shape,
dtype=myDtype,
trainable=False)
dampingLayer.build(input_shape=dampingInputLayer.shape)
dampingLayer.set_weights([np.asarray([inertia], dtype=dampingLayer.dtype)])
dampingLayer = dampingLayer(dampingInputLayer)
deltaDampingLayer = delta_damping_mlp(normalizedInputLayer)
scaledDeltaDampingLayer = Lambda(
lambda x, damping_low=damping_low, damping_up=damping_up: x *
(damping_up - damping_low) + damping_low)(deltaDampingLayer)
correctedDampingLayer = Lambda(lambda x: x[0] + x[1])(
[dampingLayer, scaledDeltaDampingLayer])
FRFAmpInputLayer = Concatenate(axis=-1)(
[omegaLayer, correctedStiffnessLayer, correctedDampingLayer])
FRFAmpLayer = FRFAmplitude(input_shape=FRFAmpInputLayer.shape,
dtype=myDtype,
trainable=False)
FRFAmpLayer.build(input_shape=FRFAmpInputLayer.shape)
FRFAmpLayer.set_weights([np.asarray([inertia], dtype=FRFAmpLayer.dtype)])
FRFAmpLayer = FRFAmpLayer(FRFAmpInputLayer)
functionalModel = Model(inputs=[inputLayer], outputs=[FRFAmpLayer])
functionalModel.compile(
loss='mean_squared_error',
optimizer=Adam(5e-2),
metrics=['mean_absolute_error', 'mean_squared_error'])
return functionalModel
def create_model(MLP_C_layer,
MLP_m_layer,
low_C,
up_C,
low_m,
up_m,
F,
a0RNN,
batch_input_shape,
selectaux,
selectdk,
myDtype,
return_sequences=False,
unroll=False):
batch_adjusted_shape = (batch_input_shape[2] + 1, ) #Adding state
placeHolder = Input(shape=(batch_input_shape[2] + 1, )) #Adding state
filterLayer = inputsSelection(batch_adjusted_shape, selectaux)(placeHolder)
filterdkLayer = inputsSelection(batch_adjusted_shape,
selectdk)(placeHolder)
MLP_C_min = low_C
MLP_C_range = up_C - low_C
MLP_C_layer = MLP_C_layer(filterLayer)
C_layer = Lambda(lambda x: ((x * MLP_C_range) + MLP_C_min))(MLP_C_layer)
MLP_m_min = low_m
MLP_m_range = up_m - low_m
MLP_m_layer = MLP_m_layer(filterLayer)
MLP_scaled_m_layer = Lambda(lambda x: ((x * MLP_m_range) + MLP_m_min))(
MLP_m_layer)
dk_input_shape = filterdkLayer.get_shape()
dkLayer = StressIntensityRange(input_shape=dk_input_shape,
dtype=myDtype,
trainable=False)
dkLayer.build(input_shape=dk_input_shape)
dkLayer.set_weights([np.asarray([F], dtype=dkLayer.dtype)])
dkLayer = dkLayer(filterdkLayer)
ldK_layer = Lambda(lambda x: tf.math.log(x) /
(tf.math.log(tf.constant(10.))))(dkLayer)
dKm_layer = Multiply()([MLP_scaled_m_layer, ldK_layer])
aux_layer = Add()([C_layer, dKm_layer])
da_layer = Lambda(lambda x: 10**(x))(aux_layer)
functionalModel = Model(inputs=[placeHolder], outputs=[da_layer])
"-------------------------------------------------------------------------"
CDMCellHybrid = CumulativeDamageCell(model=functionalModel,
batch_input_shape=batch_input_shape,
dtype=myDtype,
initial_damage=a0RNN)
CDMRNNhybrid = RNN(cell=CDMCellHybrid,
return_sequences=return_sequences,
return_state=False,
batch_input_shape=batch_input_shape,
unroll=unroll)
model = Sequential()
model.add(CDMRNNhybrid)
model.compile(loss='mse',
optimizer=RMSprop(learning_rate=1e-6),
metrics=['mae'])
return model
