5D Library › 玛雅卓尔金双模齿轮、神谕十字对合变换与六十四闸门网络渗流 · English

第五章 独立模型下的闸门激活与通道渗流

激活过程

一张人类图由 13 个天体(太阳、地球、月亮、南北交点与水星到冥王星)在两个时刻的位置决定:出生时刻,以及太阳位置早 88 度的「设计」时刻,共 26 次「激活」。一个闸门只要被任一次激活落中就算激活;一条通道当且仅当两端的闸门都被激活时导通;一个中心当且仅当至少连着一条导通的通道时被「定义」。

本章先用一个最简单的模型:26 次激活彼此独立,都均匀地落在 64 个闸门上。因为每个闸门宽度相同,这相当于 26 个独立均匀的黄经。第六章再用真实的行星位置检验这个近似。

精确结果

被激活的闸门数服从占位分布,可由第二类斯特林数精确写出。导通通道全无的概率可以精确计数:闸门图是一个 K4K_4 加 30 条孤立边,它的独立集由多项式 (1+4x)(1+2x)30(1+4x)(1+2x)^{30} 计数,再与满射数相乘即可。导通通道数 CC 的期望与方差则由「若干指定闸门全部被激活」的容斥概率得到。

推导

# Chapter 5. p08: The independence model, exactly - one chart (26 draws)
# and a two-person composite (52 draws).
# Suppose the 26 activations (13 bodies, two moments) were independent and
# uniform over the 64 gates (each gate is 5.625 degrees wide). Then:
#  - the number of distinct gates hit has the occupancy law
#    P(D = j) = C(64, j) j! S(26, j) / 64^26 (S = Stirling, 2nd kind);
#  - a channel conducts when both its gates are hit; the gate graph is K4
#    (10, 20, 34, 57) plus 30 disjoint pairs, so its independent sets are
#    counted by (1 + 4x)(1 + 2x)^30, and
#    P(no channel) = sum_j [x^j] * j! S(26, j) / 64^26;
#  - E[C] and Var[C] follow from P(a given set of m gates all hit).
from fractions import Fraction
from math import comb, factorial

G = 64

def stirling2(n):
    S = [[0] * (n + 1) for _ in range(n + 1)]
    S[0][0] = 1
    for i in range(1, n + 1):
        for j in range(1, i + 1):
            S[i][j] = j * S[i - 1][j] + S[i - 1][j - 1]
    return S[n]

def all_hit(m, draws):
    return sum((-1) ** j * comb(m, j) * Fraction(G - j, G) ** draws
               for j in range(m + 1))

def poly_mul(a, b):
    out = [0] * (len(a) + len(b) - 1)
    for i, x in enumerate(a):
        for j, y in enumerate(b):
            out[i + j] += x * y
    return out

CHANNELS = ((1, 8), (2, 14), (3, 60), (4, 63), (5, 15), (6, 59), (7, 31),
            (9, 52), (10, 20), (10, 34), (10, 57), (11, 56), (12, 22),
            (13, 33), (16, 48), (17, 62), (18, 58), (19, 49), (20, 34),
            (20, 57), (21, 45), (23, 43), (24, 61), (25, 51), (26, 44),
            (27, 50), (28, 38), (29, 46), (30, 41), (32, 54), (34, 57),
            (35, 36), (37, 40), (39, 55), (42, 53), (47, 64))

def report(DRAWS):
    S = stirling2(DRAWS)
    total = G ** DRAWS
    occ = [Fraction(comb(G, j) * factorial(j) * S[j], total)
           for j in range(DRAWS + 1)]
    mean_d = sum(j * p for j, p in enumerate(occ))
    sd = (sum(j * j * p for j, p in enumerate(occ)) - mean_d ** 2) ** 0.5
    print("   {} draws: distinct gates E = {:.4f}, s.d. {:.4f}".format(
        DRAWS, float(mean_d), float(sd)))
    mode = max(range(DRAWS + 1), key=lambda j: occ[j])
    print("     most likely {} gates (P = {:.4f})".format(
        mode, float(occ[mode])))
    ind = [1]
    for poly in [[1, 4]] + [[1, 2]] * 30:
        ind = poly_mul(ind, poly)
    p0 = sum(Fraction(ind[j] * factorial(j) * S[j], total)
             for j in range(min(len(ind), DRAWS + 1)))
    if p0 < Fraction(1, 10000):
        man, ex = "{:.3e}".format(float(p0)).split("e")
        print("     P(no channel conducts) = {} x 10^{}".format(man, int(ex)))
    else:
        print("     P(no channel conducts) = {:.5f}".format(float(p0)))
    mean_c = len(CHANNELS) * all_hit(2, DRAWS)
    second = sum(all_hit(len(set(e) | set(f)), DRAWS)
                 for e in CHANNELS for f in CHANNELS)
    print("     channels conducting: E[C] = {:.4f}, Var[C] = {:.4f}".format(
        float(mean_c), float(second - mean_c ** 2)))
    print("     P(a given channel conducts) = {:.5f}".format(
        float(all_hit(2, DRAWS))))

if __name__ == "__main__":
    print("   check: 64(1 - (63/64)^26) = {:.4f}".format(
        64 * (1 - (63 / 64) ** 26)))
    report(26)
    report(52)
#    check: 64(1 - (63/64)^26) = 21.5033
#    26 draws: distinct gates E = 21.5033, s.d. 1.6338
#      most likely 22 gates (P = 0.2362)
#      P(no channel conducts) = 0.00462
#      channels conducting: E[C] = 3.9603, Var[C] = 2.5290
#      P(a given channel conducts) = 0.11001
#    52 draws: distinct gates E = 35.7818, s.d. 2.3607
#      most likely 36 gates (P = 0.1680)
#      P(no channel conducts) = 6.662 x 10^-9
#      channels conducting: E[C] = 11.1618, Var[C] = 5.5687
#      P(a given channel conducts) = 0.31005

单人盘中被激活的闸门数期望为 21.5033,标准差 1.6338,最可能的值是 22(0.2362)。每条通道导通的概率为 0.11001,导通通道数的期望为 3.9603,方差 2.5290(小于期望,比泊松分布更集中)。一条通道都不导通的精确概率为 0.00462。

两人合盘把 52 次激活放在一起:被激活的闸门数期望 35.7818,每条通道导通的概率升到 0.31005,导通通道数的期望为 11.1618,方差 5.5687;一条通道都不导通的概率只有 6.662×10−96.662 \times 10^{-9}。

中心、分裂与类型

程序按人类图公开的规则计算被定义的中心、定义的连通块数(一块为单一定义,两块为「二分」,三块、四块为「三分」「四分」)与「类型」:没有中心被定义为反映者;荐骨被定义为生产者,若还有动力中心经由被定义的中心连到喉咙则为显示生产者;荐骨未定义而有动力中心连到喉咙为显示者;其余为投射者 (Hu and Bunnell 2011)。

推导

# Chapter 5. p09: Defined centres, splits and "types" in the independence
# model. A centre is defined when it touches a conducting channel. The
# defined centres fall into connected groups (1 = single definition,
# 2 = split, 3 = triple split, 4 = quadruple split). Type rules (as the
# system states them): none defined -> Reflector; Sacral defined ->
# Generator, "Manifesting" if a motor (Sacral, Heart, Solar Plexus, Root)
# connects to the Throat through defined centres; Sacral open and a motor
# connected to the Throat -> Manifestor; otherwise Projector.
import random

CENTRES = {
    "Head": (64, 61, 63), "Ajna": (47, 24, 4, 17, 43, 11),
    "Throat": (62, 23, 56, 35, 12, 45, 33, 8, 31, 20, 16),
    "G": (1, 13, 25, 46, 2, 15, 10, 7), "Heart": (21, 40, 26, 51),
    "Spleen": (48, 57, 44, 50, 32, 28, 18),
    "SolarPl": (6, 37, 22, 36, 30, 55, 49),
    "Sacral": (5, 14, 29, 59, 9, 3, 42, 27, 34),
    "Root": (53, 60, 52, 19, 39, 41, 58, 38, 54)}
CHANNELS = ((1, 8), (2, 14), (3, 60), (4, 63), (5, 15), (6, 59), (7, 31),
            (9, 52), (10, 20), (10, 34), (10, 57), (11, 56), (12, 22),
            (13, 33), (16, 48), (17, 62), (18, 58), (19, 49), (20, 34),
            (20, 57), (21, 45), (23, 43), (24, 61), (25, 51), (26, 44),
            (27, 50), (28, 38), (29, 46), (30, 41), (32, 54), (34, 57),
            (35, 36), (37, 40), (39, 55), (42, 53), (47, 64))
HOME = {g: c for c, gs in CENTRES.items() for g in gs}
MOTORS = ("Sacral", "Heart", "SolarPl", "Root")

def analyse(gates):
    on = [(HOME[a], HOME[b]) for a, b in CHANNELS
          if a in gates and b in gates]
    adj = {}
    for a, b in on:
        adj.setdefault(a, set()).add(b)
        adj.setdefault(b, set()).add(a)
    comps, seen = [], set()
    for c in adj:
        if c in seen:
            continue
        stack, comp = [c], set()
        while stack:
            v = stack.pop()
            if v not in comp:
                comp.add(v)
                stack.extend(adj[v] - comp)
        seen |= comp
        comps.append(comp)
    throat = next((k for k in comps if "Throat" in k), set())
    motor_throat = any(m in throat for m in MOTORS)
    if not adj:
        kind = "Reflector"
    elif "Sacral" in adj:
        kind = "Manifesting Gen." if motor_throat else "Generator"
    else:
        kind = "Manifestor" if motor_throat else "Projector"
    return len(on), len(adj), len(comps), kind

def summary(samples, label):
    n = len(samples)
    kinds = {}
    for _, _, _, k in samples:
        kinds[k] = kinds.get(k, 0) + 1
    split = {}
    for _, d, c, _ in samples:
        if d:
            split[c] = split.get(c, 0) + 1
    print("   {}: mean defined centres {:.3f}".format(
        label, sum(s[1] for s in samples) / n))
    for k in ("Generator", "Manifesting Gen.", "Projector", "Manifestor",
              "Reflector"):
        print("     {:17s} {:.4f}".format(k, kinds.get(k, 0) / n))
    print("     definition groups 1/2/3/4: " + " ".join(
        "{:.4f}".format(split.get(c, 0) / n) for c in (1, 2, 3, 4)))

if __name__ == "__main__":
    rng = random.Random(72009)
    T = 200000
    samples = [analyse({rng.randrange(1, 65) for _ in range(26)})
               for _ in range(T)]
    summary(samples, "independence model, 26 draws")
#    independence model, 26 draws: mean defined centres 5.260
#      Generator         0.3159
#      Manifesting Gen.  0.4056
#      Projector         0.1527
#      Manifestor        0.1212
#      Reflector         0.0047
#      definition groups 1/2/3/4: 0.3567 0.4821 0.1470 0.0095

在独立模型下,平均有 5.260 个中心被定义;生产者占 0.3159,显示生产者 0.4056,投射者 0.1527,显示者 0.1212,反映者 0.0047(与上面的精确值 0.00462 一致)。有定义的盘中,单一定义占 0.3567,二分 0.4821,三分 0.1470,四分 0.0095。

注

这一模型的局限。 26 次激活并不独立:地球永远在太阳对面,南北交点互为对冲,水星与金星离太阳不远,外行星在 88 天内几乎不动。第六章用真实的行星位置重算,结果在反映者的比例上有明显差别。

小结

在独立均匀模型下,单人盘平均激活 21.5033 个闸门、导通 3.9603 条通道,零导通的精确概率为 0.00462;两人合盘零导通的概率为 6.662×10−96.662 \times 10^{-9}。类型与分裂的比例完全由激活过程与图的结构决定。

参考文献

Hu, Ra Uru, and Lynda Bunnell. 2011. The Definitive Book of Human Design: The Science of Differentiation. HDC Publishing.

Fan, Y. (2026). Mayan Tzolkin Modular Gears, Oracle-Cross Involutions and Sixty-Four-Gate Network Percolation: Coprime Cycle Synchronization, Antipode-Analog Permutations and Nine-Center Connected Components [玛雅卓尔金双模齿轮、神谕十字对合变换与六十四闸门网络渗流] (Version 1.0.0). 5D Library. https://doi.org/10.5281/zenodo.24090193 · Licensed under Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0).