def test_Min_max(param, expected):
'''
Testing instantiate Min_Max model and test it against test cases
'''
my_ALPHA = hs.Alpha()
A = hs.convert_matrix_to_cart(param[0]['A'])
weight_const = param[0]['weight_const']
A0 = [0]
# It is acceptable to enter either direct_matrix or A,B,U matrices
try:
direct_matrix = hs.convert_matrix_to_cart(param[0]['ALPHA'])
my_ALPHA.add(direct_matrix=direct_matrix)
except KeyError:
B = hs.convert_matrix_to_cart(param[0]['B'])
U = hs.convert_matrix_to_cart(param[0]['U'])
my_ALPHA.add(A=A, B=B, U=U)
try:
A0 = hs.convert_matrix_to_cart(param[0]['A0'])
except KeyError:
pass
expected_W = hs.convert_matrix_to_cart(expected)
my_model = hs.Min_max(A, my_ALPHA,
weight_const=weight_const,name='Min_max') # Setting the model almost with no constraints
W = my_model.solve()
print((expected))
print('Residual Vibration rmse calculated = ', my_model.rmse())
print('Residual Vibration rmse from test_case = ',
hs.rmse(hs.residual_vibration(my_ALPHA.value, expected_W, A)))
print('expected_residual_vibration',
hs.convert_matrix_to_math(my_model.expected_residual_vibration()))
print('Correction weights', hs.convert_cart_math(W))
# Constraint Minmax algorithm was slightly inefficient in CVXPY
# The rmse was marginally more than the author solution
np.testing.assert_allclose(W, expected_W, rtol=0.09) # allowance 9% error
def test_WLS(param, expected):
'''
Testing instantiate LMI model and test it against test cases
'''
my_ALPHA = hs.Alpha()
A = hs.convert_matrix_to_cart(param[0]['A'])
A0 = [0]
# It is acceptable to enter either direct_matrix or A,B,U matrices
try:
direct_matrix = hs.convert_matrix_to_cart(param[0]['ALPHA'])
my_ALPHA.add(direct_matrix=direct_matrix)
except KeyError:
B = hs.convert_matrix_to_cart(param[0]['B'])
U = hs.convert_matrix_to_cart(param[0]['U'])
my_ALPHA.add(A=A, B=B, U=U)
try:
A0 = hs.convert_matrix_to_cart(param[0]['A0'])
except KeyError:
pass
expected_W = hs.convert_matrix_to_cart(expected)
my_model = hs.LeastSquares(A - A0, my_ALPHA, name='simple_least_square')
W = my_model.solve(solver='WLS')
print('Residual Vibration rmse calculated = ', my_model.rmse())
print('Residual Vibration rmse from test_case = ',
hs.rmse(hs.residual_vibration(my_ALPHA.value, expected_W, A)))
print('expected_residual_vibration',
hs.convert_matrix_to_math(my_model.expected_residual_vibration()))
np.testing.assert_allclose(W, expected_W, rtol=0.05) # allowance 5% error
def test_big_matrix_LSQ(test_big_alpha, test_big_A, test_model_LSQ):
'''
Testing the perfromance of model
'''
start = time.time()
error = hs.residual_vibration(test_big_alpha.value, test_model_LSQ,
test_big_A)
end = time.time()
print(error.shape)
print('Time elapsed = ', end - start)
np.testing.assert_allclose(error, 0, atol=1e-5)
def test_performance(n):
alpha = hs.Alpha()
real = np.random.uniform(0, 10, [n, n])
imag = np.random.uniform(0, 10, [n, n])
alpha.add(real + imag * 1j)
real = np.random.uniform(0, 10, [n, 1])
imag = np.random.uniform(0, 10, [n, 1])
A = real + imag * 1j
start = time.time()
w = hs.LeastSquares(A, alpha).solve()
error = hs.residual_vibration(alpha.value, w, A)
t = time.time() - start
return round(t, 2)
def test_LMI(param, expected):
'''
Testing insantiate Least square model and test it against test cases
'''
my_ALPHA = hs.Alpha()
A = hs.convert_matrix_to_cart(param[0]['A'])
weight_const = param[0]['weight_const']
A0 = [0]
# It is acceptable to enter either direct_matrix or A,B,U matrices
try:
direct_matrix = hs.convert_matrix_to_cart(param[0]['ALPHA'])
my_ALPHA.add(direct_matrix=direct_matrix)
except KeyError:
B = hs.convert_matrix_to_cart(param[0]['B'])
U = hs.convert_matrix_to_cart(param[0]['U'])
my_ALPHA.add(A=A, B=B, U=U)
try:
A0 = hs.convert_matrix_to_cart(param[0]['A0'])
except KeyError:
pass
expected_W = hs.convert_matrix_to_cart(expected)
my_model = hs.LMI(A,
my_ALPHA,
weight_const=weight_const,
V_max=76,
critical_planes={1, 9},
name='LMI')
W = my_model.solve()
print('Residual Vibration rmse calculated = ', my_model.rmse())
print('Residual Vibration rmse from test_case = ',
hs.rmse(hs.residual_vibration(my_ALPHA.value, expected_W, A)))
print('expected_residual_vibration',
hs.convert_matrix_to_math(my_model.expected_residual_vibration()))
print('Correction weights', hs.convert_cart_math(W))
np.testing.assert_allclose(W, expected_W, rtol=0.05) # allowance 5% error
import hsbalance as hs
A = [['170@112'],
['53@78']] # --> Initial vibration conditions First Row in Table above.
B = [
[
'235@94', '185@115'
], # --> Vibration at sensor 1 when trial masses were added at plane 1&2 (First column for both trial runs)
['58@68', '77@104']
] # Vibration at sensor 2 when trial masses were added at plane 1&2 (Second column for both trial runs)
U = ['1.15@0', '1.15@0'] # Trial masses 2.5 g at plane 1 and 2 consequently
A = hs.convert_math_cart(A)
B = hs.convert_math_cart(B)
U = hs.convert_math_cart(U)
alpha = hs.Alpha() # Instantiate Alpha class
alpha.add(A=A, B=B, U=U)
model_LeastSquares = hs.LeastSquares(A=A, alpha=alpha)
w = model_LeastSquares.solve(
) # Solve the model and get the correction weights vector
# Calculate Residual vibration vector
residual_vibration = hs.residual_vibration(alpha.value, w, A)
# Caculate Root mean square error for model
RMSE = hs.rmse(residual_vibration)
# Convert w back into mathmatical expression
w = hs.convert_cart_math(w)
# print results
print(model_LeastSquares.info())