NetLogo大地图高效设置指定比例连片咖啡种植区的方法

一、运行效率优化 #

现有代码存在重复统计全局patch和错误选择非农业区patch的问题，针对超大地图场景，优化如下：

• 仅统计一次农业区数量：原代码中ask patches [set total-cultivation ...]会让每个patch都执行一次count操作，完全冗余。直接用全局变量赋值统计农业区（匹配cobertura值18-19.99）：

  globals [total-cultivation agricultural-patches]
  
  to setup-agricultural-areas
    ; 直接筛选符合条件的农业区patch并缓存集合
    set agricultural-patches patches with [cobertura >= 18 and cobertura < 20]
    set total-cultivation length agricultural-patches
  end
  

• 精准选择目标patch：原代码ask n-of ... patches会在所有patch中随机选择，可能选到非农业区。改为仅在缓存的农业区集合中选择：

  to set-coffee-areas
    let target-count round (total-cultivation * %area)
    ; 直接在农业区集合中选指定数量的patch
    ask n-of target-count agricultural-patches [set plant-type coffee]
  end
  

  缓存agricultural-patches集合后，后续操作无需重复筛选patch，大幅减少超大地图下的计算量。

二、实现连片种植区 #

若需要连片的咖啡种植区，可采用种子扩散法，从随机种子patch开始，逐步向相邻农业区扩散，直到达到目标数量：

globals [ agricultural-patches coffee-count-target coffee-patches ]

to setup-connected-coffee-areas
  set agricultural-patches patches with [cobertura >= 18 and cobertura < 20]
  set coffee-count-target round (length agricultural-patches * %area)
  set coffee-patches []
  
  ; 选择初始种子patch（必须是农业区）
  let seed one-of agricultural-patches
  if seed != nobody [
    set coffee-patches lput seed coffee-patches
    ask seed [set plant-type coffee]
  ]
  
  ; 扩散逻辑：从已选咖啡区的邻居中添加新的农业区patch
  while [length coffee-patches < coffee-count-target] [
    ; 获取已选咖啡区的所有未被选中的农业区邻居
    let candidates (neighbors coffee-patches) with [member? self agricultural-patches and not member? self coffee-patches]
    if empty? candidates [ break ] ; 防止农业区不连通导致无法继续扩散
    let new-patch one-of candidates
    set coffee-patches lput new-patch coffee-patches
    ask new-patch [set plant-type coffee]
  ]
end

• 若需要多个独立的连片区块，可在扩散完成后，从剩余未选中的农业区中再次选择种子，重复扩散逻辑，直到达到目标总数。

内容的提问来源于stack exchange，提问作者bksnlow