File: polynomial_utils.py

package info (click to toggle)
python-fluids 1.0.27-1
  • links: PTS, VCS
  • area: main
  • in suites: forky, sid, trixie
  • size: 13,384 kB
  • sloc: python: 59,459; f90: 1,033; javascript: 49; makefile: 47
file content (384 lines) | stat: -rw-r--r-- 11,887 bytes parent folder | download
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
# type: ignore
"""Chemical Engineering Design Library (ChEDL). Utilities for process modeling.
Copyright (C) 2018, 2019, 2020, 2021, 2022, 2023 Caleb Bell <Caleb.Andrew.Bell@gmail.com>

Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicensse, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.
"""
from math import log, sqrt

__all__ = ['polyint', 'polyint_over_x', 'polyder', 'quadratic_from_points',
'deflate_cubic_real_roots',  'exp_poly_ln_tau_coeffs3', 'exp_poly_ln_tau_coeffs2',
'polynomial_offset_scale', 'stable_poly_to_unstable',
'polyint_stable',
'polyint_over_x_stable', 'poly_convert',
]
from fluids.numerics.special import comb


# def stable_poly_to_unstable(coeffs, low, high):
#     if len(coeffs) == 0:
#         return coeffs
#     if high != low:
#         from numpy.polynomial import Polynomial
#         # Handle the case of no transformation, no limits
#         my_poly = Polynomial([-0.5*(high + low)*2.0/(high - low), 2.0/(high - low)])
#         def horner(coeffs, x):
#             # Keep this copy here
#             tot = 0.0
#             for c in coeffs:
#                 tot = tot*x + c
#             return tot
#         coeffs = horner(coeffs, my_poly).coef[::-1].tolist()
#     return coeffs

def poly_add(p1, p2):
    # Adds two polynomials p1 and p2
    max_len = max(len(p1), len(p2))
    result = [0.0] * max_len
    for i in range(max_len):
        coeff1 = p1[i] if i < len(p1) else 0.0
        coeff2 = p2[i] if i < len(p2) else 0.0
        result[i] = coeff1 + coeff2
    return result

def poly_mul(p1, p2):
    # Multiplies two polynomials p1 and p2
    result = [0.0] * (len(p1) + len(p2) - 1)
    for i in range(len(p1)):
        for j in range(len(p2)):
            result[i + j] += p1[i] * p2[j]
    return result


def stable_poly_to_unstable(coeffs, low, high):
    if len(coeffs) == 0:
        return coeffs
    if high != low:
        a = 2.0 / (high - low)
        b = - (high + low) / (high - low)
        x_poly = [b, a]  # Represents the polynomial b + a*x
        # Initialize the result with the first coefficient of the polynomial
        result = [coeffs[0]]
        for c in coeffs[1:]:
            # Multiply the current result by x_poly
            result = poly_mul(result, x_poly)
            # Add the current coefficient c
            result = poly_add(result, [c])
        coeffs = result[::-1]  # Reverse to match the original coefficient order
        # Removes leading zeros from a polynomial
    for i in range(len(coeffs)):
        if coeffs[i] != 0.0:
            return coeffs[i:]
    return coeffs


def polyint_stable(coeffs, xmin, xmax):
    offset, scale = polynomial_offset_scale(xmin, xmax)
    scale = 1.0/scale
    N = len(coeffs)
    out = [0.0]*(N+1)
    for i in range(N):
        out[i] = scale*coeffs[i]/(N-i)
    return out

    # from numpy.polynomial import Polynomial
    # numpy_to_int = Polynomial(coeffs[::-1], domain=(xmin, xmax))
    # new_thing = numpy_to_int.integ()
    # coeffs_from_numpy = new_thing.coef
    # coeffs_from_numpy = coeffs_from_numpy[::-1].tolist()
    # return coeffs_from_numpy



def polynomial_offset_scale(xmin, xmax):
    range_inv = 1.0/(xmax - xmin)
    offset = (-xmax - xmin)*range_inv
    scale = 2.0*range_inv
    return offset, scale


def polyint(coeffs):
    """not quite a copy of numpy's version because this was faster to
    implement.
    Tried out a bunch of optimizations, and this hits a good balance
    between CPython and pypy speed.
    """
#    return ([0.0] + [c/(i+1) for i, c in enumerate(coeffs[::-1])])[::-1]
    N = len(coeffs)
    out = [0.0]*(N+1)
    for i in range(N):
        out[i] = coeffs[i]/(N-i)
    return out

def polyint_over_x(coeffs):
    N = len(coeffs)
    Nm1 = N - 1
    poly_terms = [0.0]*N
    for i in range(Nm1):
        poly_terms[i] = coeffs[i]/(Nm1-i)
    if N:
        log_coef = coeffs[-1]
        return poly_terms, log_coef
    else:
        return poly_terms, 0.0
#    N = len(coeffs)
#    log_coef = coeffs[-1]
#    Nm1 = N - 1
#    poly_terms = [coeffs[Nm1-i]/i for i in range(N-1, 0, -1)]
#    poly_terms.append(0.0)
#    return poly_terms, log_coef
#    coeffs = coeffs[::-1]
#    log_coef = coeffs[0]
#    poly_terms = [0.0]
#    for i in range(1, len(coeffs)):
#        poly_terms.append(coeffs[i]/i)
#    return list(reversed(poly_terms)), log_coef

def polyder(c, m=1):
    """not quite a copy of numpy's version because this was faster to
    implement.
    """
    cnt = m

    if cnt == 0:
        return c

    n = len(c)
    if cnt >= n:
        c = []
    else:
        der = [0.0]*n
        for i in range(cnt): # normally only happens once
            n -= 1
            for j in range(n, 0, -1):
                der[j - 1] = j*c[j]
            c = der[0:n]
    return c

def quadratic_from_points(x0, x1, x2, f0, f1, f2):
    '''
    from sympy import *
    f, a, b, c, x, x0, x1, x2, f0, f1, f2 = symbols('f, a, b, c, x, x0, x1, x2, f0, f1, f2')

    func = a*x**2 + b*x + c
    Eq0 = Eq(func.subs(x, x0), f0)
    Eq1 = Eq(func.subs(x, x1), f1)
    Eq2 = Eq(func.subs(x, x2), f2)
    sln = solve([Eq0, Eq1, Eq2], [a, b, c])
    cse([sln[a], sln[b], sln[c]], optimizations='basic', symbols=utilities.iterables.numbered_symbols(prefix='v'))
    '''
    v0 = -x2
    v1 = f0*(v0 + x1)
    v2 = f2*(x0 - x1)
    v3 = f1*(v0 + x0)
    v4 = x2*x2
    v5 = x0*x0
    v6 = x1*x1
    v7 = 1.0/(v4*x0 + v5*x1 + v6*x2 - (v4*x1  + v5*x2 + v6*x0))
    v8 = -v4
    a = v7*(v1 + v2 - v3)
    b = -v7*(f0*(v6 + v8) - f1*(v5 + v8) + f2*(v5 - v6))
    c = v7*(v1*x1*x2 + v2*x0*x1 - v3*x0*x2)
    return (a, b, c)

def quadratic_from_f_ders(x, v, d1, d2):
    '''from sympy import *
    f, a, b, c, x, v, d1, d2 = symbols('f, a, b, c, x, v, d1, d2')

    f0 = a*x**2 + b*x + c
    f1 = diff(f0, x)
    f2 = diff(f0, x, 2)

    solve([Eq(f0, v), Eq(f1, d1), Eq(f2, d2)], [a, b, c])
    '''
    a = d2*0.5
    b = d1 - d2*x
    c = -d1*x + d2*x*x*0.5 + v
    return (a, b, c)


def exp_poly_ln_tau_coeffs2(T, Tc, val, der):
    '''
    from sympy import *
    T, Tc, T0, T1, T2, sigma0, sigma1, sigma2 = symbols('T, Tc, T0, T1, T2, sigma0, sigma1, sigma2')
    val, der = symbols('val, der')
    from sympy.abc import a, b, c
    from fluids.numerics import horner
    coeffs = [a, b]
    lntau = log(1 - T/Tc)
    sigma = exp(horner(coeffs, lntau))
    d0 = diff(sigma, T)
    Eq0 = Eq(sigma,val)
    Eq1 = Eq(d0, der)
    s = solve([Eq0, Eq1], [a, b])
    '''
    x0 = 1.0/val
    x1 = T - Tc
    x2 = der*log(-x1/Tc)
    c0 = der*x0*x1
    c1 = x0*(-T*x2 + Tc*x2 + val*log(val))
    return (c0, c1)

def exp_poly_ln_tau_coeffs3(T, Tc, val, der, der2):
    '''
    from sympy import *
    T, Tc, T0, T1, T2, sigma0, sigma1, sigma2 = symbols('T, Tc, T0, T1, T2, sigma0, sigma1, sigma2')
    val, der, der2 = symbols('val, der, der2')
    from sympy.abc import a, b, c
    from fluids.numerics import horner
    coeffs = [a, b, c]
    lntau = log(1 - T/Tc)
    sigma = exp(horner(coeffs, lntau))
    d0 = diff(sigma, T)

    Eq0 = Eq(sigma,val)
    Eq1 = Eq(d0, der)
    Eq2 = Eq(diff(d0, T), der2)

    # s = solve([Eq0, Eq1], [a, b])
    s = solve([Eq0, Eq1, Eq2], [a, b, c])
    '''
    x0 = der*val
    x1 = Tc*x0
    x2 = T*x0
    x3 = der2*val
    x4 = 2.0*T*Tc
    x5 = x3*x4
    x6 = T*T
    x7 = der*der
    x8 = x6*x7
    x9 = Tc*Tc
    x10 = x7*x9
    x11 = x4*x7
    x12 = x3*x6
    x13 = x3*x9
    x14 = val*val
    x15 = 1.0/x14
    x16 = x15*0.5
    x17 = log(-(T - Tc)/Tc)
    x18 = x1*x17
    x19 = x17*x2
    x20 = x17*x17
    a = -x16*(x1 + x10 - x11 - x12 - x13 - x2 + x5 + x8)
    b = x15*(-x1 + x10*x17 - x11*x17 - x12*x17 - x13*x17 + x17*x5 + x17*x8 + x18 - x19 + x2)
    c = x16*(-x1*x20 - x10*x20 + x11*x20 + x12*x20 + x13*x20 + 2*x14*log(val) + 2.0*x18 - 2.0*x19 + x2*x20 - x20*x5 - x20*x8)
    return (a, b, c)


def deflate_cubic_real_roots(b, c, d, x0):
    F = b + x0
    G = -d/x0

    D = F*F - 4.0*G
#     if D < 0.0:
#         D = (-D)**0.5
#         x1 = (-F + D*1.0j)*0.5
#         x2 = (-F - D*1.0j)*0.5
#     else:
    if D < 0.0:
        return (0.0, 0.0)
    D = sqrt(D)
    x1 = 0.5*(D - F)#(D - c)*0.5
    x2 = 0.5*(-F - D) #-(c + D)*0.5
    return (x1, x2)



def polyint_over_x_stable_helper(coeffs, i, n, scale, offset, scale_powers, offset_powers):
#     term = scale**(i)
    term = scale_powers[i]
    inner_term = 0.0
    for j in range(n):
        multiplier = comb(j, i)
#         delta = multiplier*coeffs[-j-1]*offset**(j-i)
        delta = multiplier*coeffs[-j-1]*offset_powers[j-i]
        inner_term += delta
    return term*inner_term/i

def polyint_over_x_stable(coeffs, xmin, xmax):
    '''Take a stable polynomial coefficient series as
    evaluated by horner_stable and the limits e.g. Tmin, Tmax
    and transform them into the integral over x.

    This has the unfortunate property of breaking the stability
    of the series. The impact of this is bad but nothing
    catastropic has been found yet.

    The output int_over_x_coeffs, log_coeff should
    be evaulated with horner_log.

    I tried using math.fsum for power accuracy in the coefficients but it
    did not help.

    The coefficients from this function can be converted
    to stable form (goes directly into horner_stable_log) as follows:

    from numpy.polynomial.polynomial import Polynomial
    stable_coeffs = Polynomial(terms[::-1]).convert(domain=(Tmin, Tmax)).coef.tolist()[::-1]

    However, the precision of the conversion is worse.
    '''
    offset, scale = polynomial_offset_scale(xmin, xmax)
    n = len(coeffs)
    scale_iter = 1.0
    scale_powers = [scale_iter]
    for i in range(n):
        scale_iter *= scale
        scale_powers.append(scale_iter)
    offset_iter = 1.0
    offset_powers = [offset_iter]
    for i in range(n):
        offset_iter *= offset
        offset_powers.append(offset_iter)

    log_coeff = 0.
    for i, coeff in enumerate(coeffs[::-1]):
        log_coeff += coeff*offset_powers[i]
    terms = [0.0]
    for i in range(1, n):
        term = polyint_over_x_stable_helper(coeffs, i, n, scale, offset, scale_powers, offset_powers)
        terms.append(term)
    terms.reverse()
    return terms, log_coeff

def poly_convert(coeffs, Tmin, Tmax):
    # from numpy.polynomial.polynomial import Polynomial
    # return Polynomial(coeffs).convert(domain=(Tmin, Tmax)).coef.tolist()
    off = 0.5*(Tmin + Tmax)
    scl = 0.5*(Tmax - Tmin)
    degree_P = len(coeffs) - 1
    Q_coeffs = [0.0] * (degree_P + 1)

    # Precompute powers of off and scl up to degree_P
    off_powers = [1.0]
    scl_powers = [1.0]
    for _ in range(degree_P):
        off_powers.append(off_powers[-1] * off)
        scl_powers.append(scl_powers[-1] * scl)

    for i in range(len(coeffs)):
        coeff_i = coeffs[i]
        binom = 1.0
        for k in range(i + 1):
            term = coeff_i * binom * off_powers[i - k] * scl_powers[k]
            Q_coeffs[k] += term
            binom *= (i - k) / (k + 1.0)

    return Q_coeffs