关于「舒适区间与适应:最优温度轨迹,而不是越热越好」,《按摩油扩散、温感与留存》给出的定量结论是什么?

结论:第五章用适应模型搜索温度安排:平衡最优是中性底色、靠后温热段再回落,平均 6.84、峰终 8.55;只追峰终的方案平均仅 4.88,本书拒绝采用。

数据口径:书中浓度、用量、蒸气压、温度与评分参数均为示意,全部计算由书中程序生成;本书不是医疗建议,不宣称任何精油、载体、温度或手法的功效,不给任何剂量或浓度建议,稀释、斜贴测试、孕期与儿童是否适用交由合格专业人员判断;本书也不是个人化经营建议,不点名任何真实品牌或门店。 第一至三章修正立项蓝图:萜烯与中链甘油三酯(C8 / C10 辛癸酸甘油三酯)任意比例互溶,不存在饱和溶解度的天花板,纯油活度约为 2 % 配方的 18.12 倍;示意角质层的滞后时间约 62.5 分钟,浓度加倍不会让剖面更深。

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。 程序只用 Python 标准库与结论吻合的打印值: 4.88, 6.84, 8.55

best for mean    : 36 36 37 36 37 36 36 37 36 36 36 36
   mean 7.07, peak-end 7.77, hottest block 37 C
best for peak-end: 41 41 41 41 41 41 41 41 41 41 39 36
   mean 4.88, peak-end 8.68, hottest block 41 C
best for balanced: 36 36 37 36 37 36 36 36 37 40 39 36
   mean 6.84, peak-end 8.55, hottest block 40 C
steady 36 C: mean 7.06, peak-end 7.77
steady 41 C: mean 4.31, peak-end 6.46
查看完整验算程序
# Chapter 5. p53: searching for the best warmth plan, 12 blocks of 5 min
# Same rating model as p51 (DT 10 s). Contact levels 34 to 41 C in 1 C
# steps; neighbouring blocks may differ by at most 3 C. Coordinate ascent
# from 20 seeded random starts, for the mean rating, for the peak-end
# score, and for their average ("balanced"). Illustrative model.
import math
import random

TAU_S, TAU_A, DT = 45.0, 240.0, 10.0
LEVELS = list(range(34, 42))


def rating(t, a):
    static = 2.0 * (1.0 - ((t - 36.5) / 4.0) ** 2)
    dyn = 2.0 * math.tanh((t - a) / 1.5) * math.tanh((35.5 - a) / 1.0)
    pen = 1.5 * (t - 41.0) ** 2 if t > 41.0 else 0.0
    return min(10.0, max(0.0, 5.0 + static + dyn - pen))


def evaluate(blocks):
    t, a, minute_r = 33.0, 33.0, []
    for c in blocks:
        for _ in range(5):
            rs = []
            for _ in range(int(60 / DT)):
                t += DT * (c - t) / TAU_S
                a += DT * (t - a) / TAU_A
                rs.append(rating(t, a))
            minute_r.append(math.fsum(rs) / len(rs))
    mean = math.fsum(minute_r) / len(minute_r)
    end = math.fsum(minute_r[-3:]) / 3.0
    pe = (max(minute_r) + end) / 2.0
    return mean, pe, (mean + pe) / 2.0


def feasible(b):
    return all(abs(b[i] - b[i + 1]) <= 3 for i in range(len(b) - 1))


def ascend(start, which):
    best = list(start)
    val = evaluate(best)[which]
    improved = True
    while improved:
        improved = False
        for i in range(12):
            for lv in LEVELS:
                if lv == best[i]:
                    continue
                cand = best[:i] + [lv] + best[i + 1:]
                if not feasible(cand):
                    continue
                v = evaluate(cand)[which]
                if v > val + 1e-12:
                    best, val, improved = cand, v, True
    return best, val


rng = random.Random(20261120)
starts = []
while len(starts) < 20:
    s = [rng.choice(LEVELS) for _ in range(12)]
    if feasible(s):
        starts.append(s)
for which, label in ((0, "mean"), (1, "peak-end"), (2, "balanced")):
    found = [ascend(s, which) for s in starts]
    plan, val = max(found, key=lambda x: (x[1], x[0]))
    m, pe, _ = evaluate(plan)
    print("best for {:8s}: {}".format(label, " ".join(map(str, plan))))
    print("   mean {:.2f}, peak-end {:.2f}, hottest block {} C".format(
        m, pe, max(plan)))
for c in (36, 41):
    m, pe, _ = evaluate([c] * 12)
    print("steady {} C: mean {:.2f}, peak-end {:.2f}".format(c, m, pe))

唯一数据源与引用

范雅福(Fan, Y.). (2026). 《草本芳香按摩油菲克透皮扩散、微循环热舒适度与理疗空间留存母典》第 5 章「舒适区间与适应:最优温度轨迹,而不是越热越好」(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-339 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026339,
  author    = {Fan, Yafu},
  title     = {Fickian Transdermal Diffusion of Botanical Massage Oils, Thermal Comfort Kinetics and Wellness Studio Retention},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

本书其他章节结论