#!/usr/bin/env python
# Created by "Thieu" at 15:53, 07/07/2021 ----------%
# Email: nguyenthieu2102@gmail.com %
# Github: https://github.com/thieu1995 %
# --------------------------------------------------%
import numpy as np
from mealpy.optimizer import Optimizer
[docs]class OriginalAO(Optimizer):
"""
The original version of: Aquila Optimization (AO)
Links:
1. https://doi.org/10.1016/j.cie.2021.107250
Examples
~~~~~~~~
>>> import numpy as np
>>> from mealpy import FloatVar, AO
>>>
>>> def objective_function(solution):
>>> return np.sum(solution**2)
>>>
>>> problem_dict = {
>>> "bounds": FloatVar(lb=(-10.,) * 30, ub=(10.,) * 30, name="delta"),
>>> "obj_func": objective_function,
>>> "minmax": "min",
>>> }
>>>
>>> model = AO.OriginalAO(epoch=1000, pop_size=50)
>>> g_best = model.solve(problem_dict)
>>> print(f"Solution: {g_best.solution}, Fitness: {g_best.target.fitness}")
>>> print(f"Solution: {model.g_best.solution}, Fitness: {model.g_best.target.fitness}")
References
~~~~~~~~~~
[1] Abualigah, L., Yousri, D., Abd Elaziz, M., Ewees, A.A., Al-Qaness, M.A. and Gandomi, A.H., 2021.
Aquila optimizer: a novel meta-heuristic optimization algorithm. Computers & Industrial Engineering, 157, p.107250.
"""
def __init__(self, epoch=10000, pop_size=100, **kwargs):
"""
Args:
epoch (int): maximum number of iterations, default = 10000
pop_size (int): number of population size, default = 100
"""
super().__init__(**kwargs)
self.epoch = self.validator.check_int("epoch", epoch, [1, 100000])
self.pop_size = self.validator.check_int("pop_size", pop_size, [5, 10000])
self.set_parameters(["epoch", "pop_size"])
self.sort_flag = False
[docs] def evolve(self, epoch):
"""
The main operations (equations) of algorithm. Inherit from Optimizer class
Args:
epoch (int): The current iteration
"""
alpha = delta = 0.1
g1 = 2 * self.generator.random() - 1 # Eq. 16
g2 = 2 * (1 - epoch / self.epoch) # Eq. 17
dim_list = np.array(list(range(1, self.problem.n_dims + 1)))
miu = 0.00565
r0 = 10
r = r0 + miu * dim_list
w = 0.005
phi0 = 3 * np.pi / 2
phi = -w * dim_list + phi0
x = r * np.sin(phi) # Eq.(9)
y = r * np.cos(phi) # Eq.(10)
QF = epoch ** ((2 * self.generator.random() - 1) / (1 - self.epoch) ** 2) # Eq.(15) Quality function
pop_new = []
for idx in range(0, self.pop_size):
x_mean = np.mean(np.array([agent.solution for agent in self.pop]), axis=0)
levy_step = self.get_levy_flight_step(beta=1.5, multiplier=1.0, case=-1)
if epoch <= (2 / 3) * self.epoch: # Eq. 3, 4
if self.generator.random() < 0.5:
pos_new = self.g_best.solution * (1 - epoch / self.epoch) + self.generator.random() * (x_mean - self.g_best.solution)
else:
idx = self.generator.choice(list(set(range(0, self.pop_size)) - {idx}))
pos_new = self.g_best.solution * levy_step + self.pop[idx].solution + self.generator.random() * (y - x) # Eq. 5
else:
if self.generator.random() < 0.5:
pos_new = alpha * (self.g_best.solution - x_mean) - self.generator.random() * \
(self.generator.random() * (self.problem.ub - self.problem.lb) + self.problem.lb) * delta # Eq. 13
else:
pos_new = QF * self.g_best.solution - (g2 * self.pop[idx].solution * self.generator.random()) - \
g2 * levy_step + self.generator.random() * g1 # Eq. 14
pos_new = self.correct_solution(pos_new)
agent = self.generate_empty_agent(pos_new)
pop_new.append(agent)
if self.mode not in self.AVAILABLE_MODES:
agent.target = self.get_target(pos_new)
self.pop[idx] = self.get_better_agent(agent, self.pop[idx], self.problem.minmax)
if self.mode in self.AVAILABLE_MODES:
pop_new = self.update_target_for_population(pop_new)
self.pop = self.greedy_selection_population(self.pop, pop_new, self.problem.minmax)
[docs]class AAO(Optimizer):
"""
The original version of: Adaptive Aquila Optimizer (AAO)
Links:
1. https://doi.org/10.1016/j.rineng.2024.103261
Examples
~~~~~~~~
>>> import numpy as np
>>> from mealpy import FloatVar, AO
>>>
>>> def objective_function(solution):
>>> return np.sum(solution**2)
>>>
>>> problem_dict = {
>>> "bounds": FloatVar(n_vars=30, lb=(-10.,) * 30, ub=(10.,) * 30, name="delta"),
>>> "obj_func": objective_function,
>>> "minmax": "min",
>>> }
>>>
>>> model = AO.AAO(epoch=1000, pop_size=50, sharpness=10.0, sigmoid_midpoint=0.5)
>>> g_best = model.solve(problem_dict)
>>> print(f"Solution: {g_best.solution}, Fitness: {g_best.target.fitness}")
>>> print(f"Solution: {model.g_best.solution}, Fitness: {model.g_best.target.fitness}")
References
~~~~~~~~~~
[1] Al-Selwi, S. M., Hassan, M. F., Abdulkadir, S. J., Ragab, M. G., Alqushaibi, A., & Sumiea, E. H. (2024).
Smart grid stability prediction using adaptive aquila optimizer and ensemble stacked bilstm. Results in Engineering, 24, 103261.
"""
def __init__(self, epoch=10000, pop_size=100, sharpness=10.0, sigmoid_midpoint=0.5, **kwargs):
"""
Args:
epoch (int): maximum number of iterations, default = 10000
pop_size (int): number of population size, default = 100
sharpness (float): is a positive variable that controls the sharpness of the transition between exploration and exploitation, default is 10.0, Valid range: [0.1, 10000.0].
sigmoid_midpoint (float): a variable that controls the midpoint of the sigmoid function as it determines when the transition should be applied, default is 0.5, Valid range: [0.0, 1.0].
"""
super().__init__(**kwargs)
self.epoch = self.validator.check_int("epoch", epoch, [1, 100000])
self.pop_size = self.validator.check_int("pop_size", pop_size, [5, 10000])
self.sharpness = self.validator.check_float("sharpness", sharpness, [0.1, 10000.0])
self.sigmoid_midpoint = self.validator.check_float("sigmoid_midpoint", sigmoid_midpoint, [0.0, 1.0])
self.set_parameters(["epoch", "pop_size", "sharpness", "sigmoid_midpoint"])
self.sort_flag = False
[docs] def evolve(self, epoch):
"""
The main operations (equations) of algorithm. Inherit from Optimizer class
Args:
epoch (int): The current iteration
"""
alpha = delta = 0.1
g1 = 2 * self.generator.random() - 1 # Eq. 16
g2 = 2 * (1 - epoch / self.epoch) # Eq. 17
dim_list = np.array(list(range(1, self.problem.n_dims + 1)))
miu = 0.00565
r0 = 10
r = r0 + miu * dim_list
w = 0.005
phi0 = 3 * np.pi / 2
phi = -w * dim_list + phi0
x = r * np.sin(phi) # Eq.(9)
y = r * np.cos(phi) # Eq.(10)
QF = epoch ** ((2 * self.generator.random() - 1) / (1 - self.epoch) ** 2) # Eq.(15) Quality function
pop_new = []
for idx in range(0, self.pop_size):
x_mean = np.mean(np.array([agent.solution for agent in self.pop]), axis=0)
levy_step = self.get_levy_flight_step(beta=1.5, multiplier=1.0, case=-1)
# Dynamically balance the exploration and exploitation phases
sigmoid_factor = 1 / (1 + np.exp(-self.sharpness * (epoch / self.epoch - self.sigmoid_midpoint)))
if np.random.rand() <= (1 - sigmoid_factor):
if self.generator.random() < 0.5:
pos_new = (self.g_best.solution * (1 - epoch / self.epoch) +
self.generator.random() * (x_mean - self.g_best.solution)) # Eq. (3) and Eq. (4)
else:
idx = self.generator.choice(list(set(range(0, self.pop_size)) - {idx}))
pos_new = self.g_best.solution * levy_step + self.pop[idx].solution + self.generator.random() * (y - x) # Eq. 5
else:
if self.generator.random() < 0.5:
pos_new = (alpha * (self.g_best.solution - x_mean) - self.generator.random() *
(self.generator.random() * (self.problem.ub - self.problem.lb) + self.problem.lb) * delta) # Eq. 13
else:
pos_new = (QF * self.g_best.solution - (g2 * self.pop[idx].solution * self.generator.random()) -
g2 * levy_step + self.generator.random() * g1) # Eq. 14
pos_new = self.correct_solution(pos_new)
agent = self.generate_empty_agent(pos_new)
pop_new.append(agent)
if self.mode not in self.AVAILABLE_MODES:
agent.target = self.get_target(pos_new)
self.pop[idx] = self.get_better_agent(agent, self.pop[idx], self.problem.minmax)
if self.mode in self.AVAILABLE_MODES:
pop_new = self.update_target_for_population(pop_new)
self.pop = self.greedy_selection_population(self.pop, pop_new, self.problem.minmax)