def gen_poly(self, n):
self.order = n
# zeroth polynomial
self.poly = [nppoly.polyone]
alpha = [self.get_alpha(self.poly[0])]
beta = [1.]
# first polynomial
self.poly.append(nppoly.polymulx(self.poly[0]))
self.poly[1] = nppoly.polyadd(self.poly[1], -alpha[0] * self.poly[0])
alpha.append(self.get_alpha(self.poly[1]))
beta.append(self.get_beta(self.poly[1], self.poly[0]))
# reccurence relation for other polynomials
for i in range(2, n + 1):
p_i = nppoly.polymulx(self.poly[i - 1])
p_i = nppoly.polyadd(p_i, -alpha[i - 1] * self.poly[i - 1])
p_i = nppoly.polyadd(p_i, -beta[i - 1] * self.poly[i - 2])
self.poly.append(p_i)
alpha.append(self.get_alpha(self.poly[i]))
beta.append(self.get_beta(self.poly[i], self.poly[i - 1]))
# normalise polynomials
for i in range(len(self.poly)):
self.poly[i] = self.poly[i] / np.prod(beta[:i])
# create Jacobi matrix
self.jacobi = (np.diag(np.sqrt(beta[1:]), -1) + np.diag(alpha, 0) +
np.diag(np.sqrt(beta[1:]), 1))
def getEpsrNumDen(Mat_pol, N_true_poles):
## create num and den
list_num, list_den = [], []
for k in range(2 * N_true_poles):
list_num.append((Mat_pol[1, k]))
list_den.append((-Mat_pol[2, k], 1.))
den = (1)
for k in range(2 * N_true_poles):
den = poly.polymul(den, list_den[k])
inter = []
pinter = (1)
for k1 in range(2 * N_true_poles):
pinter = list_num[k1]
for k2 in range(2 * N_true_poles):
if (k1 != k2):
pinter = poly.polymul(pinter, list_den[k2])
inter.append(pinter)
num = inter[0]
for k in range(1, 2 * N_true_poles):
num = poly.polyadd(num, inter[k])
for k in range(len(num)):
num[k] /= (1j)**k
for k in range(len(den)):
den[k] /= (1j)**k
num, den = num.real, den.real
# np.savetxt('text_num.out', num, delimiter=',')
# np.savetxt('text_den.out', den, delimiter=',')
num = poly.polyadd(num, den)
w2 = (0., 0., -1.)
num2 = poly.polymul(w2, num)
# print_w2_chi(num2,den)
# dnum2, dden = get_der_rational(num2,den)
return (num, den, num2)
def poly_lagrange(px, py):
tab_q = base_lagrange(px)
rt = len(px)
res = nppol.polyzero
for h in range(rt):
res = nppol.polyadd(res, py[h] * tab_q[h])
return res
def plot(self, U, V, ctr):
X1 = arange(min(U) - 2, max(U) + 2, 0.1)
x = [1]
for i in range(len(U)):
x = P.polymul(x, [-1 * U[i], 1])
plt.axis([-10, 10, min(V) - 1, max(V) + 1])
b = [0]
for i in range(len(U)):
a = P.polydiv(x, [-1 * U[i], 1])
b = P.polyadd(
P.polymul((P.polydiv(a[0], P.polyval(U[i], a[0])))[0], [V[i]]),
b)
Y = P.polyval(
X1,
P.polymul((P.polydiv(a[0], P.polyval(U[i], a[0])))[0], [V[i]]))
plt.plot(X1, Y, 'y')
plt.plot(U, V, 'ro')
X1 = arange(-5, 5, 0.1)
Y = P.polyval(X1, b)
plt.plot(X1, Y, 'b', label='Required Polynomial')
plt.plot((-8, 8), (0, 0), 'k')
plt.grid(b=True, which='both', color='0.65', linestyle='-')
Y = list(Y)
plt.plot((0, 0), (-max(Y) - 5, max(Y) + 5), 'k')
plt.xlabel('x-axis')
plt.ylabel('y-axis')
plt.legend(loc=1)
filename = "this_plot" + str(ctr) + ".png"
path = "C:\\Users\HP\Anaconda3\static"
fullpath = os.path.join(path, filename)
plt.savefig(fullpath)
def hermit_interpol_polynomial(nodes, f):
def divided_difference(row, col):
# row = j, col = i
# Если знаменатель не ноль
if nodes[row + col] != nodes[row]:
return (dif_matrix[row + 1, col - 1] -
dif_matrix[row, col - 1]) / (nodes[row + col] - nodes[row])
# Если знаменатель ноль
else:
deriv = derivative(f, col)
return deriv(nodes[row]) / factorial(col)
n = len(nodes)
dif_matrix = np.zeros((n, n))
for i in range(n):
dif_matrix[i, 0] = f(nodes[i])
for i in range(1, n):
for j in range(0, n - i):
dif_matrix[j, i] = divided_difference(j, i)
result = Polynomial(dif_matrix[0, 0])
for i in range(n - 1):
result = pl.polyadd(
result,
Polynomial(np.poly(nodes[0:i + 1])[::-1]) *
dif_matrix[0, i + 1])[0]
return result, dif_matrix
def newton_interpol_polynomial(nodes, f):
n = len(nodes)
dif_matrix = np.zeros((n, n))
for i in range(n):
dif_matrix[i, 0] = f(nodes[i])
# Считаем разделённые разности. Храним их в верхнетреугольной матрице
for i in range(1, n):
for j in range(0, n - i):
dif_matrix[j,
i] = (dif_matrix[j + 1, i - 1] -
dif_matrix[j, i - 1]) / (nodes[j + i] - nodes[j])
# Строим интерпол. полином Ньютона, испоьзуя полученные разности
result = Polynomial(dif_matrix[0, 0])
for i in range(n - 1):
result = pl.polyadd(
result,
Polynomial(np.poly(nodes[0:i + 1])[::-1]) *
dif_matrix[0, i + 1])[0]
# Помимо полинома возвращаем ещё и матрицу разностей, т.к. её, вообще говоря, можно использовать,
# если мы захотим добавить ещё узлов интерполяции
return result, dif_matrix
def negPNX(n):
if n == 1:
return (1)
if n < 1:
return
if n in negPNXDict:
return negPNXDict[n]
first = P.polymul((3, 2 * n - 4), negPNX(n - 1))
second = P.polymul((0, 4), (1, -1))
deriv = P.polyder(negPNX(n - 1))
second = P.polymul(second, deriv)
third = posPNX(n - 1)
ret = P.polyadd(first, second)
ret = P.polyadd(ret, third)
negPNXDict[n] = ret
return ret
def posPNX(n):
if n == 1:
return (1)
if n < 1:
return
if n in posPNXDict:
return posPNXDict[n]
first = P.polymul((1, 2 * n - 2), posPNX(n - 1))
second = P.polymul((0, 4), (1, -1))
deriv = P.polyder(posPNX(n - 1))
second = P.polymul(second, deriv)
third = P.polymul((0, 1), (negPNX(n - 1)))
ret = P.polyadd(first, second)
ret = P.polyadd(ret, third)
posPNXDict[n] = ret
return ret
def plot(self,U,V,ctr):
X1=arange(min(U)-2,max(U)+2,0.1)
x=[1]
for i in range(len(U)):
x=P.polymul(x,[-1*U[i],1])
plt.axis([-10,10,min(V)-1,max(V)+1])
b=[0]
for i in range(len(U)):
a=P.polydiv(x,[-1*U[i],1])
b=P.polyadd(P.polymul((P.polydiv(a[0],P.polyval(U[i],a[0])))[0],[V[i]]),b)
Y=P.polyval(X1,P.polymul((P.polydiv(a[0],P.polyval(U[i],a[0])))[0],[V[i]]))
plt.plot(X1,Y,'y')
plt.plot(U,V,'ro')
X1=arange(-5,5,0.1)
Y=P.polyval(X1,b)
plt.plot(X1,Y,'b',label='Required Polynomial')
plt.plot((-8,8),(0,0),'k')
plt.grid(b=True, which='both', color='0.65',linestyle='-')
Y=list(Y)
plt.plot((0,0),(-max(Y)-5,max(Y)+5),'k')
plt.xlabel('x-axis')
plt.ylabel('y-axis')
plt.legend(loc=1)
filename = "this_plot"+str(ctr)+".png"
path = "C:\\Users\HP\Anaconda3\static"
fullpath = os.path.join(path, filename)
plt.savefig(fullpath)
def get_extended_cycle(p, n, extender):
newp = p**n - 1
a = [0, 1]
curr = [0, 1]
alphas = {0: [1], 1: a}
for x in range(2, n):
curr = pol.polymul(curr, a)
alphas[x] = list(map(int, curr))
curr = list(map(lambda x: -x % p, extender[:-1]))
curr = trim(curr)
alphas[n] = curr
for x in range(n + 1, newp):
curr = pol.polymul(curr, a)
while (len(curr) > n):
nam = int(curr[-1] % p)
curr = list(
map(
lambda x: int(x) % p,
pol.polyadd(
list(map(lambda x: x * nam, alphas[len(curr) - 1])),
curr[:-1])))
curr = list(map(lambda x: int(x) % p, curr))
alphas[x] = list(map(int, curr))
return (alphas)
def reshare(PORT, n, t, secret, polynom):
HOST = ''
s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
s.bind((HOST, PORT))
s.listen(n)
i = 1
prime = 7103
coeffs = []
for j in range(t):
coeffs.append(random.randint(1, prime))
coeffs.append(0)
newpolynom = numpy.poly1d(polynomial.polyadd(polynom.c, coeffs))
print "Old polynomial is: {0}x + {1}".format(int(polynom.c[0]),
int(polynom.c[1]))
print "New polynomial is: {0}x + {1}".format(
int(polynom.c[0]) + coeffs[0],
int(polynom.c[1]) + coeffs[1])
while i <= n:
conn, addr = s.accept()
conn.sendall(str(i))
time.sleep(0.1)
conn.sendall(str(int(newpolynom(i))))
time.sleep(0.1)
conn.close()
i = i + 1
s.close()
def polyadd(a, b):
"""
Add polynomial a by polynomial b.
a and b are lists from highest order term to lowest.
"""
val = polynomial.polyadd(a[::-1], b[::-1])
val = val[::-1]
return val
def lagrange(nodes):
result = Polynomial([0])
w = Polynomial(np.poly(nodes)[::-1])
deriv = w.deriv(1)
for i in range(len(nodes)):
result = pl.polyadd(result,
make_l_k(i, nodes, w, deriv) * f(nodes[i]))[0]
return result
def polynomials_sum(x, y, mod, poly_mod):
return np.int64(
np.round(
poly.polydiv(
poly.polyadd(x, y) % mod,
poly_mod
)[1] % mod
)
)
def test_polyadd(self) :
for i in range(5) :
for j in range(5) :
msg = "At i=%d, j=%d" % (i,j)
tgt = np.zeros(max(i,j) + 1)
tgt[i] += 1
tgt[j] += 1
res = poly.polyadd([0]*i + [1], [0]*j + [1])
assert_equal(trim(res), trim(tgt), err_msg=msg)
def polyadd_wm(x, y, poly_mod):
"""Add two polynomials
Args:
x, y: two polynomials to be added.
poly_mod: polynomial modulus.
Returns:
A polynomial in Z[X]/(poly_mod).
"""
return poly.polydiv(poly.polyadd(x, y), poly_mod)[1]
def test_polyadd(self):
for i in range(5):
for j in range(5):
msg = "At i=%d, j=%d" % (i, j)
tgt = np.zeros(max(i, j) + 1)
tgt[i] += 1
tgt[j] += 1
res = poly.polyadd([0] * i + [1], [0] * j + [1])
assert_equal(trim(res), trim(tgt), err_msg=msg)
def interpolation_polynomial(xi, eta, e, field):
coeflist = 0
for i in range(e):
xi_j_list = [xi[j] for j in range(e) if j != i]
xi_i = xi[i]
xi_minus_xi_j = construct_polynomial(xi_j_list, field)
lambnda_i = eta[i] * inv(evaluation(xi_minus_xi_j, xi_i, field), field)
a = P.polymul(lambnda_i, xi_minus_xi_j)
coeflist = P.polyadd(coeflist, a) % field
return [int(i) for i in coeflist]
def add_poly_matrices(matrix_one, matrix_two):
size = len(matrix_one)
output_matrix = []
for i in range(size):
row = []
for j in range(size):
sum = P.polyadd(matrix_one[i][j], matrix_two[i][j])
row.append(sum)
output_matrix.append(row)
return output_matrix
def lagrange(nodes):
result = Polynomial([0])
# np.poly строит полином по набору корней. Но порядок коэффициетов противоположный тому, который
# принимает конструктор класса Polynomial
w = Polynomial(np.poly(nodes)[::-1])
deriv = w.deriv(1)
for i in range(len(nodes)):
result = pl.polyadd(result,
make_l_k(i, nodes, w, deriv) * f(nodes[i]))[0]
# возвращается не только сумма
return result
def test_polydiv(self) :
# check zero division
assert_raises(ZeroDivisionError, poly.polydiv, [1], [0])
# check scalar division
quo, rem = poly.polydiv([2],[2])
assert_equal((quo, rem), (1, 0))
quo, rem = poly.polydiv([2,2],[2])
assert_equal((quo, rem), ((1,1), 0))
# check rest.
for i in range(5) :
for j in range(5) :
msg = "At i=%d, j=%d" % (i,j)
ci = [0]*i + [1,2]
cj = [0]*j + [1,2]
tgt = poly.polyadd(ci, cj)
quo, rem = poly.polydiv(tgt, ci)
res = poly.polyadd(poly.polymul(quo, ci), rem)
assert_equal(res, tgt, err_msg=msg)
def test_polydiv(self):
# check zero division
assert_raises(ZeroDivisionError, poly.polydiv, [1], [0])
# check scalar division
quo, rem = poly.polydiv([2], [2])
assert_equal((quo, rem), (1, 0))
quo, rem = poly.polydiv([2, 2], [2])
assert_equal((quo, rem), ((1, 1), 0))
# check rest.
for i in range(5):
for j in range(5):
msg = "At i=%d, j=%d" % (i, j)
ci = [0] * i + [1, 2]
cj = [0] * j + [1, 2]
tgt = poly.polyadd(ci, cj)
quo, rem = poly.polydiv(tgt, ci)
res = poly.polyadd(poly.polymul(quo, ci), rem)
assert_equal(res, tgt, err_msg=msg)
def _polytustin(a, T):
n = len(a)
if sum(a[1:]) == 0:
return a
a_new = 0
num = [(T - 1) / (T + 1), 1]
den = [1, (T - 1) / (T + 1)]
for i in range(n):
pnum = polypow(num, i)
pden = polypow(den, n - 1 - i)
a_new = polyadd(a_new, a[i] * polymul(pnum, pden))
return a_new
def polyadd(x, y, modulus, poly_mod):
"""Multiply two polynoms
Args:
x, y: two polynoms to be multiplied.
modulus: coefficient modulus.
poly_mod: polynomial modulus.
Returns:
A polynomial in Z_modulus[X]/(poly_mod).
"""
return np.int64(
np.round(
poly.polydiv(poly.polyadd(x, y) % modulus, poly_mod)[1] % modulus))
def polycomp(c1, c2):
"""
Compose two polynomials : c1 o c2 = c1(c2(x))
The arguments are sequences of coefficients, from lowest order term to highest, e.g., [1,2,3] represents the polynomial 1 + 2*x + 3*x**2.
"""
# TODO: Polynomial(Polynomial()) seems to do just that?
# using Horner's method to compute the result of a polynomial
cr = [c1[-1]]
for a in reversed(c1[:-1]):
# cr = cr * c2 + a
cr = polynomial.polyadd(polynomial.polymul(cr, c2), [a])
return cr
def polycomp(c1, c2):
"""
Compose two polynomials : c1 o c2 = c1(c2(x))
The arguments are sequences of coefficients, from lowest order term to highest, e.g., [1,2,3] represents the polynomial 1 + 2*x + 3*x**2.
"""
# TODO: Polynomial(Polynomial()) seems to do just that?
# using Horner's method to compute the result of a polynomial
cr = [c1[-1]]
for a in reversed(c1[:-1]):
# cr = cr * c2 + a
cr = polynomial.polyadd(polynomial.polymul(cr, c2), [a])
return cr
def get_B(z, a):
N = z.shape[0]
left = np.array([1])
middle = np.array([1])
for i in range(1, N):
c1 = np.array([-a[i] * z[i - 1], a[i]])
cc = P.polymul(c1, left)
right = P.polyadd(middle, cc)
left = middle
middle = right
return right
def Newton(self, L, M): # Newton function takes two lists as arguments
# L contains the list of x values
# M contains the list of f(x) values
# e.g.-
#L=[1,2,3] , M=[0,-1,0]
# i.e., f(1)=0, f(2)=-1, f(3)=0
self.x_values = L
self.fx_values = M
n = len(self.x_values) # n=length of L, i.e., the number of points
mat = [[0.0 for i in range(n)]
for j in range(n)] # initialising an n*n matrix
for i in range(n): # filling 1st column of matrix with f(x) values
mat[i][0] = self.fx_values[i]
for i in range(1, n): # calculating entries of matrix
for j in range(n - i):
mat[j][i] = (mat[j + 1][i - 1] - mat[j][i - 1]) / (
self.x_values[j + i] - self.x_values[j])
# The matrix is of the form (for 4 points - x,y,z,w)
# f(x) f(x,y) f(x,y,z) f(x,y,z,w)
# f(y) f(y,z) f(y,z,w) 0
# f(z) f(z,w) 0 0
# f(w) 0 0 0
result = array((mat[0][0], )) # initialising result array
for i in range(1, n):
prod = (-1 * self.x_values[0], 1
) # initialising prod polynomial which is to be multiplied
# with corresponding element of matrix mat
for j in range(1, i):
prod = P.polymul(
prod, (-1 * self.x_values[j], 1)) # calculating prod
result = P.polyadd(result,
array(prod) * mat[0][i]) # calculating result
result = list(result) # list of co-efficients
self.polynomial = "" # string to store final polynomial
for i in range(len(result) - 1, 0, -1): # building up the string
if (result[i] != 0):
if (result[i] > 0 and i != (len(result) - 1)):
self.polynomial += " + " + str(
result[i]) + "x^" + str(i) + " "
elif (result[i] > 0 and i == (len(result) - 1)):
self.polynomial += str(result[i]) + "x^" + str(i) + " "
else:
self.polynomial += " - " + str(
-1 * result[i]) + "x^" + str(i) + " "
if (result[0] != 0):
self.polynomial += " + " + str(
result[0]) if result[0] > 0 else " - " + str(-1 * result[0])
return (self.polynomial) # return result
def plot(points):
s, X, F = points[:], [], []
while (s != ''):
i = s.index('(')
j = s.index(',', i)
X.append(float(s[i + 1:j]))
k = s.index(')', j)
F.append(float(s[j + 1:k]))
s = s[k + 2:]
fig = plt.figure()
plt.clf()
sub = fig.add_subplot(111)
X1 = np.arange(min(X) - 2, max(X) + 2, 0.1)
num_plots = len(X)
colormap = plt.cm.gist_ncar
plt.gca().set_color_cycle(
[colormap(i) for i in np.linspace(0.2, 0.9, num_plots)])
x = [1.0]
for i in range(len(X)):
x = P.polymul(x, [-1 * X[i], 1])
b = [0.0]
for i in range(len(X)):
a = P.polydiv(x, [-1 * X[i], 1])
b = P.polyadd(
P.polymul((P.polydiv(a[0], P.polyval(X[i], a[0])))[0], [F[i]]), b)
Y = P.polyval(
X1, P.polymul((P.polydiv(a[0], P.polyval(X[i], a[0])))[0], [F[i]]))
sub.plot(X1, Y)
Y = P.polyval(X1, b)
Y1 = P.polyval(np.arange(min(X), max(X) + 0.1, 0.1), b)
interpol_obj = sub.plot(X1, Y, 'k', linewidth=2)
sub.plot(X, F, 'ro', markersize=8)
plt.grid(True)
fig.legend(interpol_obj, ['Interpolating Polynomial'],
fancybox=True,
shadow=True,
loc='upper left')
plt.axis([min(X) - 3, max(X) + 3, min(Y1) - 2, max(Y1) + 2])
plt.xlabel('x axis')
plt.ylabel('y axis')
plt.title('Interpolate')
canvas = FigureCanvas(fig)
output = StringIO.StringIO()
canvas.print_png(output)
response = make_response(output.getvalue())
response.mimetype = 'image/png'
return response
def get_poly(n):
if n > len(Bessel.polynomials):
if n == 1:
Bessel.polynomials.append(Polynomial(coef=[1, 1]))
elif n == 2:
Bessel.polynomials.append(Polynomial(coef=[3, 3, 1]))
else:
b_n_2 = poly.polymul(Polynomial(coef=[0, 0, 1]),
Bessel.get_poly(n - 2))
b_n_1 = poly.polymul(Polynomial(coef=[2 * n - 1]),
Bessel.get_poly(n - 1))
b_n = poly.polyadd(b_n_1, b_n_2)
Bessel.polynomials.append(b_n[0])
return Bessel.polynomials[n - 1]
def add_poly(x, y, mod_coeff, mod_poly):
"""
Takes:
x, y: the two polynoms to be multiplied.
mod_coeff: coefficient modulus.
mod_poly: polynomial modulus.
Returns:
A polynomial in Z_modulus[X]/(poly_mod).
Adds the two polynoms x and y.
"""
return np.int64(
np.round(poly.polydiv(poly.polyadd(x, y) % mod_coeff, mod_poly)[1] % mod_coeff)
)
def funcionNewton( f, a ):
Ai = 0
polyDef = []
for i in range( len(a) ):
Ai = AiNewt( f, i, a, Ai)
b = []
for j in range( len(a) ):
b.append(a[i]) if j!=i
lista = np.poly(b)
pol = [ Ai*x for x in lista]
polyDef = P.polyadd(poly, polyDef);
return polyDef
def poly_coeff(n):
if 0 <= n < len(polys_start_coeff):
return polys_start_coeff[n]
elif n >= len(polys_start_coeff):
coeff = np.array([0.])
for i, func in recursive_poly_functions:
if i < 0 or i >= n:
raise ValueError(
"Can't apply on not yet calculated polynomial! i={}, n={}"
.format(i, n))
coeff = npoly.polyadd(coeff, func(n, poly_coeff(n - i - 1)))
return coeff
else:
raise ValueError(
"Illegal degree n={} for polynomial coefficients.".format(n))
def __init__(self, pol1, pol2):
"""
Composes two polynomials (i.e., `pol1'(`pol2')) with distinct covariance
matrices.
Considering we have polynomials
f(x) = \\sum_i a_i x^i,
g(x) = \\sum_j b_j x^j,
with variances \\sigma_f and \\sigma_g when evaluated (see
coll_dyn_activem.maths.Polynomial), we compute
\\sigma_fg(x) = \\sigma_f(g(x))
+ \\sigma_g(x) \\times [ \\sum_i i a_i g(x)^{i-1} ]^2,
as the variance of the composed polynomial f(g) evaluated at x. We
stress that this considers no correlations between the coefficients of
the polynomials.
(see https://en.wikipedia.org/wiki/Propagation_of_uncertainty#Non-linear_combinations)
Parameters
----------
pol1 : coll_dyn_activem.maths.Polynomial
First polynomial.
pol2 : coll_dyn_activem.maths.Polynomial
Second polynomial.
Returns
-------
pol : coll_dyn_activem.maths.Polynomial
Composed polynomial.
"""
self._pol1, self._pol2 = pol1, pol2
self.deg = self._pol1.deg * self._pol2.deg # degree of composed polynomial
# WARNING: numpy.polynomial.polynomial.polyadd and polypow considers
# arrays as polynomials with lowest coefficient first,
# contrarily to polyval and polyfit.
_pol1, _pol2 = self._pol1.pol[::-1], self._pol2.pol[::-1]
self.pol = np.zeros((1, )) # composed polynomial
for i in range(pol1.deg + 1):
self.pol = polyadd(self.pol, _pol1[i] * polypow(_pol2, i))
self.pol = self.pol[::-1]
def plot(points):
s,X,F=points[:],[],[]
while(s!=''):
i=s.index('(')
j=s.index(',',i)
X.append(float(s[i+1:j]))
k=s.index(')',j)
F.append(float(s[j+1:k]))
s=s[k+2:]
fig=plt.figure()
plt.clf()
sub=fig.add_subplot(111)
X1=np.arange(min(X)-2,max(X)+2,0.1)
num_plots=len(X)
colormap = plt.cm.gist_ncar
plt.gca().set_color_cycle([colormap(i) for i in np.linspace(0.2, 0.9, num_plots)])
x=[1.0]
for i in range(len(X)):
x=P.polymul(x,[-1*X[i],1])
b=[0.0]
for i in range(len(X)):
a=P.polydiv(x,[-1*X[i],1])
b=P.polyadd(P.polymul((P.polydiv(a[0],P.polyval(X[i],a[0])))[0],[F[i]]),b)
Y=P.polyval(X1,P.polymul((P.polydiv(a[0],P.polyval(X[i],a[0])))[0],[F[i]]))
sub.plot(X1,Y)
Y=P.polyval(X1,b)
Y1=P.polyval(np.arange(min(X)-0.1,max(X)+0.1,0.1),b)
interpol_obj=sub.plot(X1,Y,'k',linewidth=2)
sub.plot(X,F,'ro',markersize=8)
plt.grid(True)
fig.legend(interpol_obj,['Interpolating Polynomial'],fancybox=True,shadow=True,loc='upper left')
plt.axis([min(X)-3,max(X)+3,min(Y1)-2,max(Y1)+2])
plt.xlabel('x axis')
plt.ylabel('y axis')
plt.title('Interpolate')
#plt.show()
canvas = FigureCanvas(fig)
output = StringIO.StringIO()
canvas.print_png(output)
response = make_response(output.getvalue())
response.mimetype = 'image/png'
return response
def interpolate(pointList):
size = len(pointList)
if len(pointList) == 1:
return (fr(pointList[0][1]),)
result = [fr(0)]
for index_i in range(0, size):
x_i = fr(pointList[index_i][0])
y_i = fr(pointList[index_i][1])
member = [y_i]
for index_j in range(0, size):
if index_i == index_j:
continue
x_j = fr(pointList[index_j][0])
member = polymul(member, [-x_j, fr(1)])
multiplier = [fr(1, x_i - x_j)]
member = polymul(member, multiplier)
result = polyadd(result, member)
return result
def Newton(self,A,B):
from numpy import array
from numpy.polynomial import polynomial as P
n=len(A)
mat=[[0.0 for i in range(n)] for j in range(n)]
for i in range(n):
mat[i][0]=B[i]
for i in range(1,n):
for j in range(n-i):
mat[j][i]=(mat[j+1][i-1]-mat[j][i-1])/(A[j+i]-A[j])
res=array((mat[0][0],))
for i in range(1,n):
prod=(-1*A[0],1)
for j in range(1,i):
prod=P.polymul(prod,(-1*A[j],1))
res=P.polyadd(res,array(prod)*mat[0][i])
s=self.plot(list(res),A,B)
return s
ax2.set_ylabel('volume (ml)')
ax2.grid(True)
MAX = 300
data = data[numpy.all(data<MAX,axis=1)]
length = data.shape[0]
volumes = volumes[-length:]
order = 3
sum = None
for column_index in range(0,data.shape[1]):
coefficients = polyfit(data[:,column_index],volumes,order)
if sum is None:
sum = coefficients
else:
sum = polyadd(sum,coefficients)
average = sum/data.shape[1]
round_digits = 8
average = [round(i,round_digits) for i in average]
poly = Polynomial(average)
ys = poly(data[:,-1])
ax2.plot(data[:,-1],ys,'r',linewidth=3)
ax2.text(5,7.5,r'$v = c_0 + c_1s + c_2s^2 + c_3s^3$',fontsize=20)
ax2.text(5,6.5,str(average),fontsize=18,color='r')
plot.show()
flow = False
else:
op = int(entrada)
if op == 1:
print "Valor en un punto:\n"
n = input("Grado del polinomio: ")
p1 = obj.insertPol(n,p1)
pto = input("Digite el punto a evaluar: ")
print "El resultado es: "+str(obj.evalPto(p1,pto))
elif op == 2:
print "Suma"
gr1 = input("Grado del polinomio 1")
p1 = obj.insertPol(gr1,p1)
gr1 = input("Grado del polinomio 2")
p2 = obj.insertPol(gr1,p2)
print "La suma es:"+str(npy.polyadd(p1,p2))
elif op == 3:
print "Resta"
gr1 = input("Grado del polinomio 1")
p1 = obj.insertPol(gr1,p1)
gr1 = input("Grado del polinomio 2")
p2 = obj.insertPol(gr1,p2)
print "La resta es:"+str(npy.polysub(p1,p2))
elif op == 4:
print "Multiplicacion"
gr1 = input("Grado del polinomio 1")
p1 = obj.insertPol(gr1,p1)
gr1 = input("Grado del polinomio 2")
p2 = obj.insertPol(gr1,p2)
print "La multiplicacion es:"+str(npy.polymul(p1,p2))
elif op == 5:
# marker = markers[column_index]
# ax2.plot(data_oa[:,column_index],volumes_oa,marker=marker,linestyle='--',color=color)
ax2.set_xlabel('volume (ml)')
ax2.set_ylabel('offset mean signals (ADC units)')
ax2.grid(True)
order = 3
sum_va = None
for column_index in range(0,data_va.shape[1]):
coefficients_va = polyfit(volumes_va,data_va[:,column_index],order)
if sum_va is None:
sum_va = coefficients_va
else:
sum_va = polyadd(sum_va,coefficients_va)
average_va = sum_va/data_va.shape[1]
round_digits = 8
average_va = [round(i,round_digits) for i in average_va]
with open('volume_to_adc_va.yaml', 'w') as f:
yaml.dump(average_va, f, default_flow_style=False)
poly_va = Polynomial(average_va)
ys_va = poly_va(volumes_va)
ax2.plot(volumes_va,ys_va,'r',linewidth=3)
ax2.text(0.5,110,r'$s = c_0 + c_1v + c_2v^2 + c_3v^3$',fontsize=20)
ax2.text(0.5,100,str(average_va),fontsize=18,color='r')
plot.show()
def longitudinal(velocity, density, S_gross_w, mac, Cm_q, Cz_alpha, mass, Cm_alpha, Iy, Cm_alpha_dot, Cz_u, Cz_alpha_dot, Cz_q, Cw, Theta, Cx_u, Cx_alpha):
""" output = SUAVE.Methods.Flight_Dynamics.Dynamic_Stablity.Full_Linearized_Equations.longitudinal(velocity, density, S_gross_w, mac, Cm_q, Cz_alpha, mass, Cm_alpha, Iy, Cm_alpha_dot, Cz_u, Cz_alpha_dot, Cz_q, Cw, Theta, Cx_u, Cx_alpha)
Calculate the natural frequency and damping ratio for the full linearized short period and phugoid modes
Inputs:
velocity - flight velocity at the condition being considered [meters/seconds]
density - flight density at condition being considered [kg/meters**3]
S_gross_w - area of the wing [meters**2]
mac - mean aerodynamic chord of the wing [meters]
Cm_q - coefficient for the change in pitching moment due to pitch rate [dimensionless] (2 * K * dC_m/di * lt/c where K is approximately 1.1)
Cz_alpha - coefficient for the change in Z force due to the angle of attack [dimensionless] (-C_D - dC_L/dalpha)
mass - mass of the aircraft [kilograms]
Cm_alpha - coefficient for the change in pitching moment due to angle of attack [dimensionless] (dC_m/dC_L * dCL/dalpha)
Iy - moment of interia about the body y axis [kg * meters**2]
Cm_alpha_dot - coefficient for the change in pitching moment due to rate of change of angle of attack [dimensionless] (2 * dC_m/di * depsilon/dalpha * lt/mac)
Cz_u - coefficient for the change in force in the Z direction due to change in forward velocity [dimensionless] (usually -2 C_L or -2C_L - U dC_L/du)
Cz_alpha_dot - coefficient for the change of angle of attack caused by w_dot on the Z force [dimensionless] (2 * dC_m/di * depsilon/dalpha)
Cz_q - coefficient for the change in Z force due to pitching velocity [dimensionless] (2 * K * dC_m/di where K is approximately 1.1)
Cw - coefficient to account for gravity [dimensionless] (-C_L)
Theta - angle between the horizontal axis and the body axis measured in the vertical plane [radians]
Cx_u - coefficient for the change in force in the X direction due to change in the forward velocity [dimensionless] (-2C_D)
Cx_alpha - coefficient for the change in force in the X direction due to the change in angle of attack caused by w [dimensionless] (C_L-dC_L/dalpha)
Outputs:
output - a data dictionary with fields:
short_w_n - natural frequency of the short period mode [radian/second]
short_zeta - damping ratio of the short period mode [dimensionless]
phugoid_w_n - natural frequency of the short period mode [radian/second]
phugoid_zeta - damping ratio of the short period mode [dimensionless]
Assumptions:
X-Z axis is plane of symmetry
Constant mass of aircraft
Origin of axis system at c.g. of aircraft
Aircraft is a rigid body
Earth is inertial reference frame
Perturbations from equilibrium are small
Flow is Quasisteady
Zero initial conditions
Cm_a = CF_z_a = CF_x_a = 0
Neglect Cx_alpha_dot, Cx_q and Cm_u
Source:
J.H. Blakelock, "Automatic Control of Aircraft and Missiles" Wiley & Sons, Inc. New York, 1991, p 26-41.
"""
#process
# constructing matrix of coefficients
A = (- Cx_u, mass * velocity / S_gross_w / (0.5*density*velocity**2.)) # X force U term
B = (-Cx_alpha) # X force alpha term
C = (-Cw * np.cos(Theta)) # X force theta term
D = (-Cz_u) # Z force U term
E = (-Cz_alpha, (mass*velocity/S_gross_w/(0.5*density*velocity**2.)-mac*0.5/velocity*Cz_alpha_dot)) # Z force alpha term
F = (- Cw * np.sin(Theta), (-mass*velocity/S_gross_w/(0.5*density*velocity**2.)-mac*0.5/velocity*Cz_q))# Z force theta term
G = (0.) # M moment U term
H = (-Cm_alpha, -mac*0.5/velocity*Cm_alpha_dot) # M moment alpha term
I = (0., - mac*0.5/velocity*Cm_q, Iy/S_gross_w/(0.5*density*velocity**2)/mac) # M moment theta term
# Taking the determinant of the matrix ([A, B, C],[D, E, F],[G, H, I])
EI = P.polymul(E,I)
FH = P.polymul(F,H)
part1 = P.polymul(A,P.polysub(EI,FH))
DI = P.polymul(D,I)
FG = P.polymul(F,G)
part2 = P.polymul(B,P.polysub(FG,DI))
DH = P.polymul(D,H)
GE = P.polymul(G,E)
part3 = P.polymul(C,P.polysub(DH,GE))
total = P.polyadd(part1,P.polyadd(part2,part3))
# Use Synthetic division to split polynomial into two quadratic factors
poly = total / total[4]
poly1 = poly * 1.
poly2 = poly * 1.
poly3 = poly * 1.
poly4 = poly * 1.
poly1[4] = poly[4] - poly[2]/poly[2]
poly1[3] = poly[3] - poly[1]/poly[2]
poly1[2] = poly[2] - poly[0]/poly[2]
poly2[4] = 0
poly2[3] = poly1[3] - poly[2]*poly1[3]/poly[2]
poly2[2] = poly1[2] - poly[1]*poly1[3]/poly[2]
poly2[1] = poly1[1] - poly[0]*poly1[3]/poly[2]
poly3[4] = poly[4] - poly2[2]/poly2[2]
poly3[3] = poly[3] - poly2[1]/poly2[2]
poly3[2] = poly[2] - poly2[0]/poly2[2]
poly4[3] = poly3[3] - poly2[2]*poly3[3]/poly2[2]
poly4[2] = poly3[2] - poly2[1]*poly3[3]/poly2[2]
poly4[1] = poly3[1] - poly2[0]*poly3[3]/poly2[2]
# Generate natural frequency and damping for Short Period and Phugoid modes
short_w_n = (poly4[2])**0.5
short_zeta = poly3[3]*0.5/short_w_n
phugoid_w_n = (poly2[0]/poly2[2])**0.5
phugoid_zeta = poly2[1]/poly2[2]*0.5/phugoid_w_n
output = Data()
output.short_natural_frequency = short_w_n
output.short_damping_ratio = short_zeta
output.phugoid_natural_frequency = phugoid_w_n
output.phugoid_damping_ratio = phugoid_zeta
return output
adc_data_points.append(adc_data_point)
print(adc_data_points)
volume_array = numpy.array(volume_data_points,dtype='float64')
adc_array = numpy.array(adc_data_points,dtype='int')
adc_array = adc_array[volume_array<=6]
volume_array = volume_array[volume_array<=6]
ax2.plot(volume_array,
adc_array,
linestyle='--',
linewidth=1,
color=color)
coefficients = polyfit(volume_array,adc_array,order)
if coefficients_sum is None:
coefficients_sum = coefficients
else:
coefficients_sum = polyadd(coefficients_sum,coefficients)
coefficients_average = coefficients_sum/run_count
poly_fit = Polynomial(coefficients_average)
adc_fit = poly_fit(volume_array)
ax2.plot(volume_array,
adc_fit,
linestyle='-',
linewidth=2,
label=cylinder,
color=color)
coefficients_list = [float(coefficient) for coefficient in coefficients_average]
output_data[cylinder]['volume_to_adc_low'] = coefficients_list
ax2.set_xlabel('volume (ml)')
ax2.set_ylabel('adc low value (adc units)')
ax2.legend(loc='best')
def __add__(self, other):
result = polynomial.polyadd(self._poly, other._poly)
result = self._normalizePoly(result, len(self._poly))
return Polynomial(poly=result)
def lateral_directional(velocity, Cn_Beta, S_gross_w, density, span, I_z, Cn_r, I_x, Cl_p, J_xz, Cl_r, Cl_Beta, Cn_p, Cy_phi, Cy_psi, Cy_Beta, mass):
""" output = SUAVE.Methods.Flight_Dynamics.Dynamic_Stablity.Full_Linearized_Equations.lateral_directional(velocity, Cn_Beta, S_gross_w, density, span, I_z, Cn_r, I_x, Cl_p, J_xz, Cl_r, Cl_Beta, Cn_p, Cy_phi, Cy_psi, Cy_Beta)
Calculate the natural frequency and damping ratio for the full linearized dutch roll mode along with the time constants for the roll and spiral modes
Inputs:
velocity - flight velocity at the condition being considered [meters/seconds]
Cn_Beta - coefficient for change in yawing moment due to sideslip [dimensionless] (no simple relation)
S_gross_w - area of the wing [meters**2]
density - flight density at condition being considered [kg/meters**3]
span - wing span of the aircraft [meters]
I_z - moment of interia about the body z axis [kg * meters**2]
Cn_r - coefficient for change in yawing moment due to yawing velocity [dimensionless] ( - C_D(wing)/4 - 2 * Sv/S * (l_v/b)**2 * (dC_L/dalpha)(vert) * eta(vert))
I_x - moment of interia about the body x axis [kg * meters**2]
Cl_p - change in rolling moment due to the rolling velocity [dimensionless] (no simple relation for calculation)
J_xz - products of inertia in the x-z direction [kg * meters**2] (if X and Z lie in a plane of symmetry then equal to zero)
Cl_r - coefficient for change in rolling moment due to yawing velocity [dimensionless] (Usually equals C_L(wing)/4)
Cl_Beta - coefficient for change in rolling moment due to sideslip [dimensionless]
Cn_p - coefficient for the change in yawing moment due to rolling velocity [dimensionless] (-C_L(wing)/8*(1 - depsilon/dalpha)) (depsilon/dalpha = 2/pi/e/AspectRatio dC_L(wing)/dalpha)
Cy_phi - coefficient for change in sideforce due to aircraft roll [dimensionless] (Usually equals C_L)
Cy_psi - coefficient to account for gravity [dimensionless] (C_L * tan(Theta))
Cy_Beta - coefficient for change in Y force due to sideslip [dimensionless] (no simple relation)
mass - mass of the aircraft [kilograms]
Outputs:
output - a data dictionary with fields:
dutch_w_n - natural frequency of the dutch roll mode [radian/second]
dutch_zeta - damping ratio of the dutch roll mode [dimensionless]
roll_tau - approximation of the time constant of the roll mode of an aircraft [seconds] (positive values are bad)
spiral_tau - time constant for the spiral mode [seconds] (positive values are bad)
Assumptions:
X-Z axis is plane of symmetry
Constant mass of aircraft
Origin of axis system at c.g. of aircraft
Aircraft is a rigid body
Earth is inertial reference frame
Perturbations from equilibrium are small
Flow is Quasisteady
Zero initial conditions
Neglect Cy_p and Cy_r
Source:
J.H. Blakelock, "Automatic Control of Aircraft and Missiles" Wiley & Sons, Inc. New York, 1991, p 118-124.
"""
#process
# constructing matrix of coefficients
A = (0, -span * 0.5 / velocity * Cl_p, I_x/S_gross_w/(0.5*density*velocity**2)/span ) # L moment phi term
B = (0, -span * 0.5 / velocity * Cl_r, -J_xz / S_gross_w / (0.5 * density * velocity ** 2.) / span) # L moment psi term
C = (-Cl_Beta) # L moment Beta term
D = (0, - span * 0.5 / velocity * Cn_p, -J_xz / S_gross_w / (0.5 * density * velocity ** 2.) / span ) # N moment phi term
E = (0, - span * 0.5 / velocity * Cn_r, I_z / S_gross_w / (0.5 * density * velocity ** 2.) / span ) # N moment psi term
F = (-Cn_Beta) # N moment Beta term
G = (-Cy_phi) # Y force phi term
H = (-Cy_psi, mass * velocity / S_gross_w / (0.5 * density * velocity ** 2.))
I = (-Cy_Beta, mass * velocity / S_gross_w / (0.5 * density * velocity ** 2.))
# Taking the determinant of the matrix ([A, B, C],[D, E, F],[G, H, I])
EI = P.polymul(E,I)
FH = P.polymul(F,H)
part1 = P.polymul(A,P.polysub(EI,FH))
DI = P.polymul(D,I)
FG = P.polymul(F,G)
part2 = P.polymul(B,P.polysub(FG,DI))
DH = P.polymul(D,H)
GE = P.polymul(G,E)
part3 = P.polymul(C,P.polysub(DH,GE))
total = P.polyadd(part1,P.polyadd(part2,part3))
poly = total / total[5]
# Generate the time constant for the spiral and roll modes along with the damping and natural frequency for the dutch roll mode
root = np.roots(poly)
root = sorted(root,reverse=True)
spiral_tau = 1 * root[0].real
w_n = (root[1].imag**2 + root[1].real**2)**(-0.5)
zeta = -2*root[1].real/w_n
roll_tau = 1 * root [3].real
output = Data()
output.dutch_natural_frequency = w_n
output.dutch_damping_ratio = zeta
output.spiral_tau = spiral_tau
output.roll_tau = roll_tau
return output
rand.seed(2)
print "TESING 10000 CASES..."
for k in range(10000):
order1=rand.randint(0,10)
order2=rand.randint(0,10)
c1=[0]*(order1+1)
c2=[0]*(order2+1)
for i in range(len(c1)):
c1[i]=rand.randint(-5,5)
for i in range(len(c2)):
c2[i]=rand.randint(-5,5)
# TESTING ADDITION FUNCTION
a= P.polyadd(c1,c2).tolist()
b= q.addPoly(c1,c2)
if(a!=b):
print "Failed on Case:"
print "a add b"
print "a",a
print "b",b
break
# TESTING SUBTRACTION FUNCTION
a=P.polysub(c1,c2).tolist()
b=q.subPoly(c1,c2)
if(a!=b):
print "Failed on Case:"
print "a subtract b"
print "a",a
def polyadd(c1, c2):
from numpy.polynomial.polynomial import polyadd
return polyadd(c1, c2)
