# -*- coding: utf-8 -*-
import geatpy as ea # import geatpy
import numpy as np
"""==================================实例化问题对象================================"""
problemName = 'Sphere' # 问题名称
fileName = problemName # 这里因为目标函数写在与之同名的文件里，所以文件名也是问题名称
MyProblem = getattr(__import__(fileName), problemName) # 获得自定义问题类
problem = MyProblem(30) # 生成问题对象
"""==================================种群设置================================"""
Encoding = 'RI' # 编码方式
conordis = 0 # 表示染色体解码后得到的变量是连续的
NIND = 20 # 种群规模
precisions = [50] * problem.Dim # 编码精度（适用于二进制/格雷编码）
Field = ea.crtfld(Encoding, problem.varTypes, problem.ranges, problem.borders, precisions) # 创建区域描述器
population = ea.Population(Encoding, Field, NIND) # 实例化种群对象（此时种群还没被初始化，仅仅是完成种群对象的实例化）
"""==================================算法参数设置================================"""
myAlgorithm = ea.soea_DE_rand_1_bin_templet(problem, population) # 实例化一个算法模板对象
myAlgorithm.MAXGEN = 1000 # 最大进化代数
myAlgorithm.F = 0.5
myAlgorithm.pc = 0.2
myAlgorithm.drawing = 1 # 设置绘图方式（0：不绘图；1：绘制结果图；2：绘制过程动画）
"""=======================调用算法模板进行种群进化=============================="""
[population, obj_trace, var_trace] = myAlgorithm.run() # 执行算法模板，得到最后一代种群以及进化记录器
population.save() # 把最后一代种群的信息保存到文件中
# 输出结果
best_gen = np.argmin(obj_trace[:, 1]) # 记录最优种群是在哪一代
best_ObjV = np.min(obj_trace[:, 1])
print('最优的目标函数值为：%s'%(best_ObjV))
print('最优的控制变量值为：')
for i in range(var_trace.shape[1]):
# f2[exIdx] = f2[exIdx] + np.max(f2) - np.min(f2)
# 利用可行性法则处理约束条件
CV = np.hstack([
2 - x1 - x2, x1 + x2 - 6, -2 - x1 + x2, x1 - 3 * x2 - 2,
(x3 - 3)**2 + x4 - 4, 4 - (x5 - 3)**2 - x4
])
return np.hstack([f1, f2]), CV
def calBest(self):
realBestObjV = None
return realBestObjV
problem = MyProblem() # 生成问题对象
"""==================================种群设置================================"""
Encoding = 'I' # 编码方式
conordis = 1 # 表示染色体解码后得到的变量是离散的
NIND = 50 # 种群规模
Field = ea.crtfld(Encoding, conordis, problem.ranges,
problem.borders) # 创建区域描述器
population = ea.Population(Encoding, conordis, Field,
NIND) # 实例化种群对象（此时种群还没被真正初始化）
"""==================================算法参数设置================================"""
myAlgorithm = ea.moea_NSGA2_templet(problem, population) # 实例化一个算法模板对象
myAlgorithm.MAXGEN = 200 # 最大遗传代数
"""=======================调用算法模板进行种群进化=============================="""
NDSet = myAlgorithm.run() # 执行算法模板，得到帕累托最优解集NDSet
# 输出
print('用时：%s 秒' % (myAlgorithm.passTime))
print('非支配个体数：%s 个' % (NDSet.sizes))
print('单位时间找到帕累托前沿点个数：%s 个' % (int(NDSet.sizes // myAlgorithm.passTime)))
import geatpy as ea # import geatpy
"""
描述:
该案例是moea_demo1的另一个版本，展示了如何定义aimFunc()而不是evalVars()来计算目标函数和违反约束程度值。【见MyProblem.py】
同时展示如何定义outFunc()，用于让算法在每一次进化时调用该outFunc()函数。
"""
if __name__ == '__main__':
# 实例化问题对象
problem = MyProblem()
# 定义outFunc()函数
def outFunc(alg, pop): # alg 和 pop为outFunc的固定输入参数，分别为算法对象和每次迭代的种群对象。
print('第 %d 代' % alg.currentGen)
# 构建算法
algorithm = ea.moea_NSGA2_templet(
problem,
ea.Population(Encoding='RI', NIND=50),
MAXGEN=200, # 最大进化代数
logTras=1, # 表示每隔多少代记录一次日志信息，0表示不记录。
outFunc=outFunc)
# 求解
res = ea.optimize(algorithm,
verbose=False,
drawing=1,
outputMsg=True,
drawLog=True,
saveFlag=False)
print(res)
kpPole, kiPole, kdPole = Vars[i, [3]], Vars[i, [4]], Vars[i, [5]]
pidController = cartPoleController(kpCart, kiCart, kdCart, kpPole,
kiPole, kdPole)
cartPole.run(pidController)
objValues[i] = cartPole.reward()
# 将目标函数值赋值给pop的ObjV属性
pop.ObjV = objValues
cartPole = plant(200)
problem = optimizePID()
# 种群设置
Encoding = 'RI' # 格雷码
NIND = 30 # 种群数量
Field = ea.crtfld(Encoding, problem.varTypes, problem.ranges, problem.borders)
population = ea.Population(Encoding, Field, NIND)
#算法参数设置
myAlgorithm = ea.soea_SEGA_templet(problem, population)
myAlgorithm.MAXGEN = 30 # 最大遗传代数
myAlgorithm.mutOper.F = 0.5 # 设置差分进化的变异缩放因子
myAlgorithm.recOper.XOVR = 0.5 # 设置交叉概率
myAlgorithm.drawing = 1 # 绘图方式
[population, obj_trace, var_trace] = myAlgorithm.run() # 执行优化算法
cartPole.close()
best_gen = np.argmin(obj_trace[:, 1]) # 找出最优种群
best_ObjV = obj_trace[best_gen, 1]
print('最优的目标函数值为：%s' % (best_ObjV))
bestSolution = []
for i in range(var_trace.shape[1]):
bestSolution.append(var_trace[best_gen, i])
def __init__(self,
X,
y,
target,
model,
l,
u,
drawing=1,
maxgen=100,
moea=1):
'''===============================实例化问题对象=================================='''
self.problem = MyProblem(target=target,
X=X,
y=y,
model=model,
l=l,
u=u)
'''===========================种群设置=============================='''
Encoding = 'RI' # 'RI':实整数编码，即实数和整数的混合编码；
NIND = 100 # 种群规模
Field = ea.crtfld(Encoding, self.problem.varTypes, self.problem.ranges,
self.problem.borders) # 创建区域描述器
self.population = ea.Population(Encoding, Field, NIND)
'''实例化种群对象（此时种群还没被真正初始化，仅仅是生成一个种群对象）'''
"""=========================算法参数设置============================"""
if moea == 1:
self.myAlgorithm = ea.moea_NSGA2_templet(
self.problem, self.population) # 实例化一个算法模板对象
elif moea == 2:
self.myAlgorithm = ea.moea_NSGA2_DE_templet(
self.problem, self.population)
elif moea == 3:
self.myAlgorithm = moea_NSGA2_toZero(self.problem, self.population)
elif moea == 4:
self.myAlgorithm = moea_NSGA2_DE_toZero(self.problem,
self.population)
elif moea == 5:
self.myAlgorithm = ea.moea_NSGA3_templet(self.problem,
self.population)
elif moea == 6:
self.myAlgorithm = ea.moea_NSGA3_DE_templet(
self.problem, self.population)
elif moea == 7:
self.myAlgorithm = ea.moea_awGA_templet(self.problem,
self.population)
elif moea == 8:
self.myAlgorithm = ea.moea_RVEA_templet(self.problem,
self.population)
elif moea == 9:
self.myAlgorithm = moea_NSGA2_10p_toZero(self.problem,
self.population)
elif moea == 10:
self.myAlgorithm = moea_NSGA2_20p_toZero(self.problem,
self.population)
elif moea == 11:
self.myAlgorithm = moea_NSGA2_30p_toZero(self.problem,
self.population)
elif moea == 12:
self.myAlgorithm = moea_NSGA2_random10p_toZero(
self.problem, self.population)
elif moea == 13:
self.myAlgorithm = moea_NSGA2_random20p_toZero(
self.problem, self.population)
elif moea == 14:
self.myAlgorithm = moea_NSGA2_random30p_toZero(
self.problem, self.population)
else:
print('error moea method!!')
self.myAlgorithm.MAXGEN = maxgen # 设置最大遗传代数
self.myAlgorithm.drawing = drawing