|
PSOIndividual.py
1 import numpy as np
2 import ObjFunction
3 import copy
4
5
6 class PSOIndividual:
7
8 '''
9 individual of PSO
10 '''
11
12 def __init__(self, vardim, bound):
13 '''
14 vardim: dimension of variables
15 bound: boundaries of variables
16 '''
17 self.vardim = vardim
18 self.bound = bound
19 self.fitness = 0.
20
21 def generate(self):
22 '''
23 generate a rondom chromsome
24 '''
25 len = self.vardim
26 rnd = np.random.random(size=len)
27 self.chrom = np.zeros(len)
28 self.velocity = np.random.random(size=len)
29 for i in xrange(0, len):
30 self.chrom = self.bound[0, i] + \
31 (self.bound[1, i] - self.bound[0, i]) * rnd
32 self.bestPosition = np.zeros(len)
33 self.bestFitness = 0.
34
35 def calculateFitness(self):
36 '''
37 calculate the fitness of the chromsome
38 '''
39 self.fitness = ObjFunction.GrieFunc(
40 self.vardim, self.chrom, self.bound)
PSO.py
1 import numpy as np
2 from PSOIndividual import PSOIndividual
3 import random
4 import copy
5 import matplotlib.pyplot as plt
6
7
8 class ParticleSwarmOptimization:
9
10 '''
11 the class for Particle Swarm Optimization
12 '''
13
14 def __init__(self, sizepop, vardim, bound, MAXGEN, params):
15 '''
16 sizepop: population sizepop
17 vardim: dimension of variables
18 bound: boundaries of variables
19 MAXGEN: termination condition
20 params: algorithm required parameters, it is a list which is consisting of[w, c1, c2]
21 '''
22 self.sizepop = sizepop
23 self.vardim = vardim
24 self.bound = bound
25 self.MAXGEN = MAXGEN
26 self.params = params
27 self.population = []
28 self.fitness = np.zeros((self.sizepop, 1))
29 self.trace = np.zeros((self.MAXGEN, 2))
30
31 def initialize(self):
32 '''
33 initialize the population of pso
34 '''
35 for i in xrange(0, self.sizepop):
36 ind = PSOIndividual(self.vardim, self.bound)
37 ind.generate()
38 self.population.append(ind)
39
40 def evaluation(self):
41 '''
42 evaluation the fitness of the population
43 '''
44 for i in xrange(0, self.sizepop):
45 self.population.calculateFitness()
46 self.fitness = self.population.fitness
47 if self.population.fitness > self.population.bestFitness:
48 self.population.bestFitness = self.population.fitness
49 self.population.bestIndex = copy.deepcopy(
50 self.population.chrom)
51
52 def update(self):
53 '''
54 update the population of pso
55 '''
56 for i in xrange(0, self.sizepop):
57 self.population.velocity = self.params[0] * self.population.velocity + self.params[1] * np.random.random(self.vardim) * (
58 self.population.bestPosition - self.population.chrom) + self.params[2] * np.random.random(self.vardim) * (self.best.chrom - self.population.chrom)
59 self.population.chrom = self.population[
60 i].chrom + self.population.velocity
61
62 def solve(self):
63 '''
64 the evolution process of the pso algorithm
65 '''
66 self.t = 0
67 self.initialize()
68 self.evaluation()
69 best = np.max(self.fitness)
70 bestIndex = np.argmax(self.fitness)
71 self.best = copy.deepcopy(self.population[bestIndex])
72 self.avefitness = np.mean(self.fitness)
73 self.trace[self.t, 0] = (1 - self.best.fitness) / self.best.fitness
74 self.trace[self.t, 1] = (1 - self.avefitness) / self.avefitness
75 print("Generation %d: optimal function value is: %f; average function value is %f" % (
76 self.t, self.trace[self.t, 0], self.trace[self.t, 1]))
77 while self.t < self.MAXGEN - 1:
78 self.t += 1
79 self.update()
80 self.evaluation()
81 best = np.max(self.fitness)
82 bestIndex = np.argmax(self.fitness)
83 if best > self.best.fitness:
84 self.best = copy.deepcopy(self.population[bestIndex])
85 self.avefitness = np.mean(self.fitness)
86 self.trace[self.t, 0] = (1 - self.best.fitness) / self.best.fitness
87 self.trace[self.t, 1] = (1 - self.avefitness) / self.avefitness
88 print("Generation %d: optimal function value is: %f; average function value is %f" % (
89 self.t, self.trace[self.t, 0], self.trace[self.t, 1]))
90
91 print("Optimal function value is: %f; " % self.trace[self.t, 0])
92 print "Optimal solution is:"
93 print self.best.chrom
94 self.printResult()
95
96 def printResult(self):
97 '''
98 plot the result of pso algorithm
99 '''
100 x = np.arange(0, self.MAXGEN)
101 y1 = self.trace[:, 0]
102 y2 = self.trace[:, 1]
103 plt.plot(x, y1, 'r', label='optimal value')
104 plt.plot(x, y2, 'g', label='average value')
105 plt.xlabel("Iteration")
106 plt.ylabel("function value")
107 plt.title("Particle Swarm Optimization algorithm for function optimization")
108 plt.legend()
109 plt.show()
运行程序:
1 if __name__ == "__main__":
2
3 bound = np.tile([[-600], [600]], 25)
4 pso = PSO(60, 25, bound, 1000, [0.7298, 1.4962, 1.4962])
5 pso.solve()
ObjFunction见简单遗传算法-python实现。 |
|