【智能优化算法】磷虾群算法KHA附Python代码.zip

import numpy as np import matplotlib.pyplot as plt #待求解问题，求解问题为求最小值 def function(x): y1 = 0 for i in range(len(x) - 1): y2 = 100 * ((x[i + 1] - x[i] ** 2) ** 2) + (x[i] - 1) ** 2 y1 = y1 + y2 y = abs(0 - y1) return y m = 50 #种群数量 imax = 100 #迭代次数 dimen = 20 #解的维度 rangelow = -10 #解的最小取值 rangehigh = 10 #解的最大取值 Nmax = 0.01 #最大诱导速度 Vf = 0.02 #觅食速度 #pop用于存储种群个体的位置信息，pop_fitness用于存储个体对应的适应度值 pop = np.zeros((m,dimen)) pop_fitness = np.zeros(m) #对种群个体进行初始化并计算对应适应度值 for j in range(m): pop[j] = np.random.uniform(low=rangelow, high=rangehigh,size=(1, dimen)) pop_fitness[j] = function(pop[j]) #allbestpop,allbestfit分别存储种群在历史迭代过程中最优个体解及对应适应度 allbestpop,allbestfit = pop[pop_fitness.argmin()].copy(),pop_fitness.min() #bestpop,bestfit分别存储每个个体历史最优解及对应适应度 bestpop,bestfit = pop.copy(),pop_fitness.copy() #分别为当前种群适应度最优个体适应度值、位置信息与适应度最差个体适应度值 bestfitness = pop_fitness.min() bestplace = pop[pop_fitness.argmin()].copy() worstfitness = pop_fitness.max() #his_bestfit存储每次迭代时种群历史适应度值最优的个体适应度 his_bestfit=np.zeros(imax) #Nold为历史诱导值，Fold为历史捕食移动量 Nold = np.zeros((m,dimen)) Fold = np.zeros((m,dimen)) #计算变量的上下界差距之和 sum1 = 0 for d in range(dimen): sum1 = sum1 + (rangehigh - rangelow) #开始迭代训练 for i in range(imax): print("The iteration is:",i+1) #当前迭代次数与总迭代次数的比例 iratio = (i+1)/imax #将Ct、wn以及wf这三个参数设置为在取值范围内随迭代次数减小的数，也按照取值范围可设置为常数 Ct = 0.01 + 2 * (1 - iratio) wn = 0.01 + 1 * (1 - iratio) wf = 0.01 + 1 * (1 - iratio) #对每个个体依次进行位移 for j in range(m): # 1、其他个体引起的移动 # 计算个体受到影响的范围距离 sum2 = 0 for l in range(m): e = np.sum(np.square(pop[l] - pop[j])) sum2 = sum2 + e d = sum2 / (5 * m) #计算受到一定范围内其他个体的影响alphalocal alphalocal = 0 for l in range(m): sum3 = np.sum(np.square(pop[l] - pop[j])) if sum3 <= d: k = (pop_fitness[l] - pop_fitness[j])/(worstfitness - bestfitness) x = (pop[l] - pop[j]) / (sum3 + 0.0001) alphalocal = alphalocal + k * x #计算种群最优目标的影响alphatarget Cbest = 2 * (np.random.rand() + iratio) Kbest1 = (pop_fitness[j] - bestfitness) / (worstfitness - bestfitness) xbest1 = (bestplace - pop[j]) / (np.sum(np.square(bestplace - pop[j])) + 0.0001) alphatarget = Cbest * Kbest1 * xbest1 #计算第i个个体受到的综合影响 alphai = alphalocal + alphatarget #计算其他个体引起的移动Nnew Nnew = Nmax * alphai + wn * Nold[j] # 2、觅食移动 # 食物影响系数 Cfood = 2 * (np.random.rand() - iratio) # 计算食物的估计位置Xfood及对应的适应度值 multi1 = 0 multi2 = 0 for j in range(m): multi1 = multi1 + (1 / pop_fitness[j]) * pop[j] multi2 = multi2 + (1 / pop_fitness[j]) Xfood = multi1 / multi2 fitnessfood = function(Xfood) #计算受到食物的影响移动量belta Kbest2 = (pop_fitness[j] - bestfit[j]) / (worstfitness - bestfitness) xbest2 = (bestpop[j] - pop[j]) / (np.sum(np.square(bestpop[j] - pop[j])) + 0.0001) beltabest = xbest2 * Kbest2 Kfood = (pop_fitness[j] - fitnessfood) / (worstfitness - bestfitness) xbestfood = (Xfood - pop[j]) / (np.sum(np.square(Xfood - pop[j])) + 0.0001) beltafood = Kfood * xbestfood belta = beltafood + beltabest #计算觅食移动量Fnew Fnew = Vf * belta + wf * Fold[j] #3、计算物理扩散位移量 Dmax = np.random.uniform(low=0.002 , high=0.01) D = Dmax * iratio * np.random.uniform(low=-1, high=1,size=(1, dimen)) #4、计算前面三种行为的综合影响 change = Nnew + Fnew + D #计算速度的比例因子向量deltat deltat = Ct * sum1 #5、计算当前个体在移动后的新位置 pop[j] = pop[j] + deltat * change # 对历史诱导值Nold，历史捕食移动量Fold，个体适应度值进行更新 Nold[j] = Nnew Fold[j] = Fnew pop_fitness[j] = function(pop[j]) #对当前种群适应度最优个体适应度值、位置信息与适应度最差个体适应度值进行更新 bestfitness = pop_fitness.min() worstfitness = pop_fitness.max() bestplace = pop[pop_fitness.argmin()].copy() #6、对每个个体进行交叉操作： for j in range(m): #计算交叉概率 Cr = 0.2 * (1 - (worstfitness - pop_fitness[j])/(worstfitness - bestfitness)) for a in range(dimen): #进行交叉操作 r1 = np.random.rand() if r1 < Cr: rangem = list(range(0, m)) rangem.pop(j) b = np.random.choice(rangem) pop[j][a] = pop[b][a] #7、对每个个体进行变异操作： for j in range(m): #计算变异概率Mu Mu = 0.05 * (1 - (worstfitness - pop_fitness[j])/(worstfitness - bestfitness)) for a in range(dimen): #进行变异操作 r2 = np.random.rand() if r2 < Mu: r3 = np.random.rand() rangem = list(range(0, m)) rangem.remove(j) c = np.random.choice(rangem) rangem.remove(c) d = np.random.choice(rangem) pop[j][a] = bestplace[a] + r3 * (pop[c][a] - pop[d][a]) #计算交叉、变异后的个体适应度值，对个体历史最优适应度值与位置信息进行更新 for j in range(m): pop_fitness[j] = function(pop[j]) if pop_fitness[j] < bestfit[j]: bestfit[j] = pop_fitness[j] bestpop[j] = pop[j] #对种群历史最优位置信息与适应度值进行更新 if bestfit.min() < allbestfit: allbestfit = bestfit.min() allbestpop = bestpop[bestfit.argmin()].copy() #对当前种群适应度最优个体适应度值、位置信息与适应度最差个体适应度值进行更新 bestfitness = pop_fitness.min() worstfitness = pop_fitness.max() bestplace = pop[pop_fitness.argmin()].copy() #存储当前迭代下的种群历史最优适应度值并输出 his_bestfit[i] = allbestfit print("The best fitness is:", allbestfit) print("After iteration, the best pop is:",allbestpop) print("After iteration, the best fitness is:",allbestfit) #将结果进行绘图 fig=plt.figure(figsize=(12, 10), dpi=300) plt.rcParams['font.sans-serif']=['Arial Unicode MS'] plt.title('最优适应度值的变化情况',fontdict={'weight':'normal','size': 30}) x=range(1,101,1) plt.plot(x,his_bestfit,color="red",label="KHA",linewidth=3.0, linestyle="-") plt.tick_params(labelsize=25) plt.xlim(0,101) plt.yscale("log") plt.xlabel("迭代次数",fontdict={'weight':'normal','size': 30}) plt.ylabel("适应度值",fontdict={'weight':'normal','size': 30}) plt.xticks(range(0,101,10)) plt.savefig("KHA.png") plt.show()