关于「均匀为什么会失稳:图灵条件与扩散比门槛」,《图灵斑图与据点间距》给出的定量结论是什么?
结论:第一章推出门槛:以扩散系数之比计是 11.657,以作用半径之比计是 2.414,后者与衰减率无关;立项蓝图的「一定倍数」据此修正,并并陈马尔孔等人关于差速扩散并非普遍必要的反方结论。
数据口径:书中半径、衰减率、城市长度与数据均为示意;全部计算由书中程序生成;本书属非个人化经营方法,不构成投资、法律或选址决策建议。 书中以标注框区分已发表的证据、可运行的 Python 计算(17 段程序)、原典与作者的假说。
对照数据表
下表数字由本章 Python 程序原样打印,可重新运行复现。 程序只用 Python 标准库与结论吻合的打印值: 11.657, 2.414
| uniform state a* = 2.0 | h* = 2.0 | |||||
| Jacobian [[1.0, -1.0], [4.0, -2.0]] | trace -1.0 | det 2.0 | ||||
| critical Dh/Da = (3 + 2 sqrt 2) nu/mu = 11.657 | ||||||
| critical reach ratio lh/la = 1 + sqrt 2 = 2.414 | ||||||
| lh/la | lh km | Dh/Da | max rate | lam_fast km | lam_c km | band km |
| 1.50 | 0.90 | 4.50 | -0.5000 | none | 4.62 | none |
| 2.00 | 1.20 | 8.00 | -0.1877 | none | 5.33 | none |
| 2.30 | 1.38 | 10.58 | -0.0450 | none | 5.72 | none |
| 2.50 | 1.50 | 12.50 | 0.0311 | 5.94 | 5.96 | 5.10- 6.97 |
| 3.00 | 1.80 | 18.00 | 0.1782 | 6.37 | 6.53 | 4.39- 9.71 |
| 4.00 | 2.40 | 32.00 | 0.3669 | 7.13 | 7.54 | 4.05-14.02 |
| 5.00 | 3.00 | 50.00 | 0.4840 | 7.81 | 8.43 | 3.94-18.00 |
| 6.00 | 3.60 | 72.00 | 0.5642 | 8.42 | 9.23 | 3.88-21.89 |
核心公式
\[a_t = D_a a_{xx} + a^2/h - \mu a, \qquad h_t = D_h h_{xx} + a^2 - \nu h.\]
查看完整验算程序
# Chapter 1. p11: when does even spreading of outposts become unstable?
# Model: Gierer-Meinhardt activator-inhibitor (Gierer and Meinhardt
# 1972) on a line, read as a niche brand in a new city:
# a_t = Da a_xx + a^2 / h - mu a (a: word-of-mouth activity)
# h_t = Dh h_xx + a^2 - nu h (h: saturation, cannibalisation)
# Reaches are diffusion lengths la = sqrt(Da / mu), lh = sqrt(Dh / nu).
# Linear stability of the uniform state (Turing 1952) gives the
# threshold on the reach ratio lh / la and the band of growing
# wavelengths. mu = 1, nu = 2 and la = 0.6 km are illustrative; the
# book's default cannibalisation reach is lh = 2.4 km (ratio 4.0).
import math
MU, NU = 1.0, 2.0
LA = 0.6 # word-of-mouth reach, km (illustrative)
def jacobian():
a, h = NU / MU, NU / (MU * MU) # uniform steady state
return [[2.0 * a / h - MU, -a * a / (h * h)], [2.0 * a, -NU]]
def growth(k, da, dh, j):
a11, a22 = j[0][0] - da * k * k, j[1][1] - dh * k * k
tr, det = a11 + a22, a11 * a22 - j[0][1] * j[1][0]
disc = tr * tr - 4.0 * det
return (tr + math.sqrt(disc)) / 2.0 if disc >= 0 else tr / 2.0
def scan(lh):
j = jacobian()
da, dh = MU * LA * LA, NU * lh * lh
ks = [0.001 * i for i in range(1, 8001)]
rates = [growth(k, da, dh, j) for k in ks]
best = max(range(len(ks)), key=lambda i: rates[i])
band = [ks[i] for i in range(len(ks)) if rates[i] > 0.0]
return rates[best], ks[best], band
j = jacobian()
tr = j[0][0] + j[1][1]
det = j[0][0] * j[1][1] - j[0][1] * j[1][0]
print("uniform state a* = %.1f h* = %.1f" % (NU / MU, NU / MU ** 2))
print("Jacobian [[%.1f, %.1f], [%.1f, %.1f]] trace %.1f det %.1f"
% (j[0][0], j[0][1], j[1][0], j[1][1], tr, det))
dc = (3.0 + 2.0 * math.sqrt(2.0)) * NU / MU
print("critical Dh/Da = (3 + 2 sqrt 2) nu/mu = %.3f" % dc)
print("critical reach ratio lh/la = 1 + sqrt 2 = %.3f" % (1 + 2 ** 0.5))
print("lh/la lh km Dh/Da max rate lam_fast km lam_c km band km")
for ratio in (1.5, 2.0, 2.3, 2.5, 3.0, 4.0, 5.0, 6.0):
lh = ratio * LA
rate, kf, band = scan(lh)
lam_c = 2.0 * math.pi * math.sqrt(LA * lh)
d = NU * lh * lh / (MU * LA * LA)
if band:
txt = "%5.2f-%5.2f" % (2 * math.pi / band[-1],
2 * math.pi / band[0])
lf = "%7.2f" % (2.0 * math.pi / kf)
else:
txt, lf = " none ", " none"
print("%5.2f %6.2f %6.2f %8.4f %11s %8.2f %s"
% (ratio, lh, d, rate, lf, lam_c, txt))唯一数据源与引用
范雅福(Fan, Y.). (2026). 《图灵反应扩散斑图、小众口碑品牌陌生城市据点涌现与区域垄断扩散比条件母典》第 1 章「均匀为什么会失稳:图灵条件与扩散比门槛」(版本 1.0.0). 5D Library. DOI 待发布
DOI 待发布 · 专著页正式发布后上线(本站暂无专著页) · 5DL-B-247 · v1.0.0 · CC-BY-NC-ND-4.0
BibTeX
@book{fan2026247,
author = {Fan, Yafu},
title = {Turing Reaction-Diffusion Patterns: How the Outposts of a Niche Word-of-Mouth Brand Emerge in a New City, and the Diffusion-Ratio Condition for Regional Dominance},
year = {2026},
publisher = {5D Library},
edition = {1.0.0},
note = {DOI pending}
}