基于Python求解最优化无约束优化问题（源码）.rar

import pandas as pd import numpy as np import matplotlib.pyplot as plt import matplotlib as mpl import random from collections import Counter from matplotlib.animation import FuncAnimation #fname指定字体文件 选简体显示中文 myfont = mpl.font_manager.FontProperties(fname=r"C:\Windows\Fonts\STFANGSO.TTF") mpl.rcParams['axes.unicode_minus'] = False mpl.rcParams['font.sans-serif'] = ['SimHei'] class GA: def __init__(self,NP,NG,t,f): ''' NP:种群规模 NG:最大代数 t:初始时刻 f:计算适值函数，算子 ''' self.NP = NP self.NG = NG self.t = t self.f = f def gun(self,flask = False): ''' 返回x，y flask为Ture显示图像 ''' def mod1(L,a=0,b=10): ''' 产生随机数长度为L，范围为a到b a，b默认0，10 ''' return [random.randint(a,b) for i in range(L)] def bianma(x): ''' 整数转二进制 基因编码 ''' y = "{0:b}".format(x) return y def min_max(x): ''' 0-1标准化 ''' return [(i - min(x))/(max(x) - min(x)) for i in x] def choose(ss,pp,L): ''' ss:随机序列范围0-1 pp:转轮 L:序列长度 return index:选择母体下标 选择培育后代母体 ''' index = [] for i in range(L): for j in range(L): if j == 0: if ss[i] >= pp[j] and ss[i] < pp[j+1]: index.append(0) if ss[i] >= pp[j] and ss[i] < pp[j+1]: index.append(j) elif j == L - 1: if ss[i] >= pp[j] and ss[i] <= pp[j+1]: index.append(j) return index def jiaca(x,y,n): ''' 交叉运算 x:父亲 y:母亲 n:交叉位点 return s:子代 ''' x = str(x) y = str(y) s = x[:n]+y[n:] s = int(s, 2) return s L = self.NG # 生成初始种群 x = mod1(L) for kt in range(self.NG): ii = 0 # print('第',kt+1,'轮:','\n种群:',x) # 计算适值函数 xt = [bianma(i) for i in x] # 基因编码xt f = [self.f(i) for i in x] # 计算f(x) plt.ion() plt.cla() plt.scatter(x,f) plt.title('第'+str(kt+1)+'轮计算') plt.xlabel('种群取值') plt.ylabel('适值函数值') if flask: plt.pause(1) # print('适值函数:',f) # 选择 p = [i/sum(f) for i in f] # 计算个体选择概率p pp = [sum(p[0:i]) for i in range(len(p))] pp.append(1) ss1 = min_max(mod1(L)) # 生成随机数0-1 ss2 = min_max(mod1(L)) # 生成随机数0-1 index_1 = choose(ss1,pp,L) # 父亲下标 index_2 = choose(ss2,pp,L) # 母亲下标 x1 = [xt[i] for i in index_1] # 父亲编码 x2 = [xt[i] for i in index_2] # 母亲编码 # 遗传运算 n = mod1(L) y = [jiaca(x1[i],x2[i],n[i]) for i in range(L)] # 子代y yt = [bianma(i) for i in y] # 子代基因编码yt # 判断 if x == y: break # 更新 x = y result = Counter(x) xx = max(result,key=result.get) fx = self.f(xx) return xx,fx def fun(x): f = -1 * x**2 + 100 return f ga = GA(50,50,0,fun) x,y = ga.gun(flask=True) print(x,' ',y)