关于「十二种流月窗口:断联与复合的时机」,《钦天双人合盘飞化》给出的定量结论是什么?

结论:本章把时间放进合盘:一年十二个月,月干按五虎遁排出,化出的禄、权、科、忌若落在你的夫妻宫、他的夫妻宫或两边同时,就打开十二种窗口中的一种。精确枚举表明,每一种化平均每年恰好有一个月落进你的夫妻宫;一对人有没有某种窗口取决于这对人自己,不取决于年份。双方窗口很少:禄-双方只属于 0.0670 的配对,拥有者平均等 5.4342 个月;任何一方的禄窗口属于 0.8802 的配对,平均等 4.2257 个月;从忌-双方的月份到下一个禄月平均 4.2286 个月。每一种窗口都写明了这个月对联系、距离、断联与复合做了什么,该做的动作,以及绝不要做的事。这些是诠释;本章给出了一项可以预先登记的检验及其样本量。

数据口径:全部计算由书中程序生成;本书不是婚姻、心理、法律或投资建议。 书中以标注框区分已发表的证据、可运行的 Python 计算、原典与作者的假说。

English version

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。 与结论吻合的打印值: 0.0670, 0.8802, 4.2257, 4.2286, 5.4342

lu-bothpairs that ever get it 0.0670mean wait 5.4342 monthslongest gap 10
lu, any sidepairs that ever get it 0.8802mean wait 4.2257 monthslongest gap 10
ji-bothpairs that ever get it 0.0670mean wait 5.4364 monthslongest gap 10
ji, any sidepairs that ever get it 0.8568mean wait 4.2049 monthslongest gap 10
pairs with a ji-both month 0.0670; of these, no lu month ever 0.0244
from a ji-both month to the next lu month: mean 4.2286 months
查看完整验算程序
# Chapter 7. p36: Waiting times between windows, in months.
# Same charts and Spouse palace stem sets as p35. Month stems run on without
# a break across year ends (month twelve of one year and month one of the
# next are consecutive stems; a leap month repeats the stem of the month it
# follows and is left out here), so the stem sequence has
# period 10 and a waiting time depends only on the set of stems that open
# the window. From a month taken at random we count the months until the
# next qualifying month (the current month excluded): k in 1..10.
# Reported over all chart pairs: share of pairs that ever get the window,
# mean wait among those, longest possible gap, and the bridge from a
# ji-both month to the next lu month of any side.
NAYIN = "310231423042104" * 2
BUREAU = {4: 2, 0: 3, 3: 4, 2: 5, 1: 6}
SIHUA = [(5, 13, 3, 2), (1, 11, 0, 7), (4, 1, 16, 5), (7, 4, 1, 9), (8, 7, 15, 1),
         (3, 8, 11, 17), (2, 3, 7, 4), (9, 2, 17, 16), (11, 0, 14, 3), (13, 9, 7, 8)]
KIND = ["lu", "quan", "ke", "ji"]

def stem_sets():
    """Counter over charts of (lu set, quan set, ke set, ji set) as 10-bit stem masks."""
    seen = {}
    for ys in range(10):
        y0 = (ys % 5 * 2 + 2) % 10
        for month in range(1, 13):
            for hour in range(12):
                ming = (1 + month - hour) % 12
                stem = (y0 + (ming - 2) % 12) % 10
                bu = BUREAU[int(NAYIN[((6 * stem - 5 * ming) % 60) // 2])]
                for day in range(1, 31):
                    k = 0
                    while (day + k) % bu:
                        k += 1
                    z = (2 + (day + k) // bu - 1 + (k if k % 2 == 0 else -k)) % 12
                    t = (4 - z) % 12
                    pos = (z, z - 1, z - 3, z - 4, z - 5, z - 8, t, t + 1, t + 2, t + 3, t + 4,
                           t + 5, t + 6, t + 10, 3 + month, 11 - month, 10 - hour, 4 + hour)
                    sp = {i for i, b in enumerate(pos) if (ming - b) % 12 == 2}
                    key = tuple(sum(1 << s for s in range(10) if SIHUA[s][h] in sp) for h in range(4))
                    seen[key] = seen.get(key, 0) + 1
    return seen

def wait(mask, c):
    return next((k for k in range(1, 11) if mask >> ((c + k) % 10) & 1), None)

def report(name, pick, dist):
    tot = ever = wsum = 0
    gap = 0
    for a, wa in dist.items():
        for b, wb in dist.items():
            w, x = wa * wb, pick(a, b)
            tot += w
            if x:
                ever += w
                wsum += w * sum(wait(x, c) for c in range(10)) / 10
                gap = max(gap, max(wait(x, c) for c in range(10) if x >> c & 1))
    print("   {:22s} pairs that ever get it {:.4f}  mean wait {:.4f} months  longest gap {}".format(
        name, ever / tot, wsum / ever, gap))

if __name__ == "__main__":
    seen = stem_sets()
    lu, ji, both = {}, {}, {}
    for key, w in seen.items():
        lu[key[0]] = lu.get(key[0], 0) + w
        ji[key[3]] = ji.get(key[3], 0) + w
        both[(key[0], key[3])] = both.get((key[0], key[3]), 0) + w
    report("lu-both", lambda a, b: a & b, lu)
    report("lu, any side", lambda a, b: a | b, lu)
    report("ji-both", lambda a, b: a & b, ji)
    report("ji, any side", lambda a, b: a | b, ji)
    tot = has = wsum = never = 0
    for (la, ja), wa in both.items():
        for (lb, jb), wb in both.items():
            w = wa * wb
            tot += w
            jb2 = ja & jb
            if jb2:
                has += w
                cs = [c for c in range(10) if jb2 >> c & 1]
                ws = [wait(la | lb, c) for c in cs]
                never += w if None in ws else 0
                wsum += w * sum(v for v in ws if v) / len(cs) if None not in ws else 0
    print("   pairs with a ji-both month {:.4f}; of these, no lu month ever {:.4f}".format(
        has / tot, never / has))
    print("   from a ji-both month to the next lu month: mean {:.4f} months".format(wsum / (has - never)))

唯一数据源与引用

范雅福(Fan, Y.). (2026). 《钦天双人合盘与宿命情缘飞化母典》第 7 章「十二种流月窗口:断联与复合的时机」(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-120 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026120,
  author    = {Fan, Yafu},
  title     = {The Qintian Synastry Codex of Cross-Flying Transformations},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

本书其他章节结论