File size: 18,275 Bytes
0e051b9 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 |
import sys
import numpy as np
import matplotlib.pyplot as plt
from matplotlib.backends.backend_qt5agg import FigureCanvasQTAgg as FigureCanvas
from matplotlib.figure import Figure
from PyQt5.QtWidgets import (QApplication, QMainWindow, QWidget, QVBoxLayout,
QHBoxLayout, QGroupBox, QLabel, QComboBox,
QDoubleSpinBox, QSpinBox, QPushButton, QTextEdit,
QTabWidget, QGridLayout, QProgressBar)
from PyQt5.QtCore import QThread, pyqtSignal
import random
class PSOThread(QThread):
update_signal = pyqtSignal(dict)
finished_signal = pyqtSignal(dict)
def __init__(self, problem_type, num_particles, max_iter, w, c1, c2):
super().__init__()
self.problem_type = problem_type
self.num_particles = num_particles
self.max_iter = max_iter
self.w = w
self.c1 = c1
self.c2 = c2
self.running = True
def run(self):
# Initialize particles based on problem type
if self.problem_type == "radiative_equilibrium":
bounds = [(-10, 10), (-10, 10)] # Temperature and density parameters
dim = 2
elif self.problem_type == "nuclear_reaction_rate":
bounds = [(0.1, 2.0), (1e-3, 1e-1)] # Temperature (T7) and density parameters
dim = 2
elif self.problem_type == "convective_stability":
bounds = [(0.1, 0.5), (0.1, 0.5), (0.1, 0.5)] # ∇_rad, ∇_ad, ∇_μ
dim = 3
elif self.problem_type == "opacity_optimization":
bounds = [(1e-3, 1e3), (1e4, 1e8)] # Density and temperature
dim = 2
else:
bounds = [(-5, 5), (-5, 5)]
dim = 2
# PSO initialization
particles = np.random.uniform([b[0] for b in bounds], [b[1] for b in bounds],
(self.num_particles, dim))
velocities = np.random.uniform(-1, 1, (self.num_particles, dim))
personal_best_positions = particles.copy()
personal_best_scores = np.array([self.fitness(p, self.problem_type) for p in particles])
global_best_index = np.argmin(personal_best_scores)
global_best_position = personal_best_positions[global_best_index]
global_best_score = personal_best_scores[global_best_index]
# PSO main loop
for iteration in range(self.max_iter):
if not self.running:
break
for i in range(self.num_particles):
# Update velocity
r1, r2 = random.random(), random.random()
velocities[i] = (self.w * velocities[i] +
self.c1 * r1 * (personal_best_positions[i] - particles[i]) +
self.c2 * r2 * (global_best_position - particles[i]))
# Update position
particles[i] += velocities[i]
# Apply bounds
for d in range(dim):
if particles[i, d] < bounds[d][0]:
particles[i, d] = bounds[d][0]
elif particles[i, d] > bounds[d][1]:
particles[i, d] = bounds[d][1]
# Evaluate fitness
current_fitness = self.fitness(particles[i], self.problem_type)
# Update personal best
if current_fitness < personal_best_scores[i]:
personal_best_positions[i] = particles[i].copy()
personal_best_scores[i] = current_fitness
# Update global best
if current_fitness < global_best_score:
global_best_position = particles[i].copy()
global_best_score = current_fitness
# Emit update signal
self.update_signal.emit({
'iteration': iteration,
'global_best': global_best_score,
'position': global_best_position,
'particles': particles.copy()
})
self.finished_signal.emit({
'final_score': global_best_score,
'final_position': global_best_position
})
def fitness(self, x, problem_type):
"""Fitness function based on stellar physics problems from Chapter 5-6"""
if problem_type == "radiative_equilibrium":
# Optimize radiative temperature gradient (Eq. 5.18)
# We want to minimize deviation from ideal radiative equilibrium
T, rho = x[0], x[1]
# Simplified radiative equilibrium condition
radiative_flux = (T**3 / rho) if rho > 0 else 1e10
target_flux = 1.0 # Ideal normalized flux
return abs(radiative_flux - target_flux)
elif problem_type == "nuclear_reaction_rate":
# Optimize nuclear reaction rates (Eq. 6.29)
T7, density_param = x[0], x[1] # T7 = T/10^7 K
# Gamow peak-based reaction rate approximation
reaction_rate = (T7**(-2/3)) * np.exp(-1/T7**(1/3)) * density_param
target_rate = 0.5 # Optimal reaction rate
return abs(reaction_rate - target_rate)
elif problem_type == "convective_stability":
# Schwarzschild/Ledoux criterion optimization (Eq. 5.49, 5.50)
grad_rad, grad_ad, grad_mu = x[0], x[1], x[2]
# Stability requires: ∇_rad < ∇_ad - (χ_μ/χ_T)∇_μ
# For ideal gas: χ_μ = -1, χ_T = 1
stability_condition = grad_ad + grad_mu # Ledoux criterion
instability = max(0, grad_rad - stability_condition)
return instability # Minimize instability
elif problem_type == "opacity_optimization":
# Optimize opacity for efficient energy transport
rho, T = x[0], x[1]
# Kramers opacity approximation (Eq. 5.31, 5.32)
opacity = rho * T**(-3.5) if T > 0 else 1e10
# Target opacity range for efficient transport
target_opacity = 1.0
return abs(opacity - target_opacity)
else:
# Default sphere function
return sum(xi**2 for xi in x)
def stop(self):
self.running = False
class MplCanvas(FigureCanvas):
def __init__(self, parent=None, width=5, height=4, dpi=100):
self.fig = Figure(figsize=(width, height), dpi=dpi)
super().__init__(self.fig)
self.setParent(parent)
class PSOWindow(QMainWindow):
def __init__(self):
super().__init__()
self.pso_thread = None
self.init_ui()
def init_ui(self):
self.setWindowTitle("Stellar Physics PSO Optimizer - Chapter 5-6")
self.setGeometry(100, 100, 1200, 800)
central_widget = QWidget()
self.setCentralWidget(central_widget)
layout = QHBoxLayout(central_widget)
# Left panel - Controls
left_panel = QWidget()
left_layout = QVBoxLayout(left_panel)
left_panel.setMaximumWidth(400)
# Problem selection
problem_group = QGroupBox("Stellar Physics Optimization Problem")
problem_layout = QVBoxLayout(problem_group)
self.problem_combo = QComboBox()
self.problem_combo.addItems([
"Radiative Equilibrium",
"Nuclear Reaction Rate",
"Convective Stability",
"Opacity Optimization"
])
problem_layout.addWidget(QLabel("Select Problem:"))
problem_layout.addWidget(self.problem_combo)
# Problem description
self.problem_desc = QTextEdit()
self.problem_desc.setMaximumHeight(150)
self.problem_desc.setReadOnly(True)
problem_layout.addWidget(QLabel("Problem Description:"))
problem_layout.addWidget(self.problem_desc)
left_layout.addWidget(problem_group)
# PSO parameters
pso_group = QGroupBox("PSO Parameters")
pso_layout = QGridLayout(pso_group)
pso_layout.addWidget(QLabel("Number of Particles:"), 0, 0)
self.num_particles = QSpinBox()
self.num_particles.setRange(10, 200)
self.num_particles.setValue(30)
pso_layout.addWidget(self.num_particles, 0, 1)
pso_layout.addWidget(QLabel("Max Iterations:"), 1, 0)
self.max_iter = QSpinBox()
self.max_iter.setRange(50, 1000)
self.max_iter.setValue(100)
pso_layout.addWidget(self.max_iter, 1, 1)
pso_layout.addWidget(QLabel("Inertia Weight (w):"), 2, 0)
self.w_spin = QDoubleSpinBox()
self.w_spin.setRange(0.1, 1.0)
self.w_spin.setValue(0.7)
self.w_spin.setSingleStep(0.1)
pso_layout.addWidget(self.w_spin, 2, 1)
pso_layout.addWidget(QLabel("Cognitive Coefficient (c1):"), 3, 0)
self.c1_spin = QDoubleSpinBox()
self.c1_spin.setRange(0.1, 3.0)
self.c1_spin.setValue(1.5)
self.c1_spin.setSingleStep(0.1)
pso_layout.addWidget(self.c1_spin, 3, 1)
pso_layout.addWidget(QLabel("Social Coefficient (c2):"), 4, 0)
self.c2_spin = QDoubleSpinBox()
self.c2_spin.setRange(0.1, 3.0)
self.c2_spin.setValue(1.5)
self.c2_spin.setSingleStep(0.1)
pso_layout.addWidget(self.c2_spin, 4, 1)
left_layout.addWidget(pso_group)
# Control buttons
self.run_button = QPushButton("Run PSO")
self.run_button.clicked.connect(self.run_pso)
left_layout.addWidget(self.run_button)
self.stop_button = QPushButton("Stop")
self.stop_button.clicked.connect(self.stop_pso)
self.stop_button.setEnabled(False)
left_layout.addWidget(self.stop_button)
# Progress
self.progress = QProgressBar()
left_layout.addWidget(self.progress)
# Results
results_group = QGroupBox("Results")
results_layout = QVBoxLayout(results_group)
self.results_text = QTextEdit()
self.results_text.setMaximumHeight(150)
results_layout.addWidget(self.results_text)
left_layout.addWidget(results_group)
layout.addWidget(left_panel)
# Right panel - Visualization
right_panel = QTabWidget()
# Convergence plot
self.convergence_canvas = MplCanvas(self, width=6, height=4, dpi=100)
self.convergence_ax = self.convergence_canvas.fig.add_subplot(111)
right_panel.addTab(self.convergence_canvas, "Convergence")
# Particle positions
self.particles_canvas = MplCanvas(self, width=6, height=4, dpi=100)
self.particles_ax = self.particles_canvas.fig.add_subplot(111)
right_panel.addTab(self.particles_canvas, "Particles")
# Fitness landscape
self.landscape_canvas = MplCanvas(self, width=6, height=4, dpi=100)
self.landscape_ax = self.landscape_canvas.fig.add_subplot(111)
right_panel.addTab(self.landscape_canvas, "Fitness Landscape")
layout.addWidget(right_panel)
# Update problem description
self.update_problem_desc()
self.problem_combo.currentTextChanged.connect(self.update_problem_desc)
def update_problem_desc(self):
problem = self.problem_combo.currentText()
descriptions = {
"Radiative Equilibrium":
"Optimize radiative temperature gradient (Eq. 5.18)\n"
"Minimize deviation from ideal radiative equilibrium conditions\n"
"Parameters: Temperature, Density",
"Nuclear Reaction Rate":
"Optimize thermonuclear reaction rates (Eq. 6.29)\n"
"Find optimal conditions for efficient energy generation\n"
"Based on Gamow peak theory\n"
"Parameters: Temperature (T7), Density parameter",
"Convective Stability":
"Apply Schwarzschild/Ledoux criteria (Eq. 5.49, 5.50)\n"
"Minimize convective instability in stellar layers\n"
"Parameters: ∇_rad, ∇_ad, ∇_μ",
"Opacity Optimization":
"Optimize opacity for efficient energy transport\n"
"Based on Kramers opacity law (Eq. 5.31)\n"
"Find optimal density-temperature conditions\n"
"Parameters: Density, Temperature"
}
self.problem_desc.setText(descriptions.get(problem, ""))
def run_pso(self):
if self.pso_thread and self.pso_thread.isRunning():
return
problem_map = {
"Radiative Equilibrium": "radiative_equilibrium",
"Nuclear Reaction Rate": "nuclear_reaction_rate",
"Convective Stability": "convective_stability",
"Opacity Optimization": "opacity_optimization"
}
problem_type = problem_map[self.problem_combo.currentText()]
self.pso_thread = PSOThread(
problem_type=problem_type,
num_particles=self.num_particles.value(),
max_iter=self.max_iter.value(),
w=self.w_spin.value(),
c1=self.c1_spin.value(),
c2=self.c2_spin.value()
)
self.pso_thread.update_signal.connect(self.update_plots)
self.pso_thread.finished_signal.connect(self.optimization_finished)
self.run_button.setEnabled(False)
self.stop_button.setEnabled(True)
self.progress.setValue(0)
self.progress.setMaximum(self.max_iter.value())
self.convergence_ax.clear()
self.particles_ax.clear()
self.landscape_ax.clear()
self.best_scores = []
self.iterations = []
self.pso_thread.start()
def stop_pso(self):
if self.pso_thread:
self.pso_thread.stop()
self.pso_thread.wait()
self.run_button.setEnabled(True)
self.stop_button.setEnabled(False)
def update_plots(self, data):
iteration = data['iteration']
best_score = data['global_best']
position = data['position']
particles = data['particles']
# Update convergence plot
self.best_scores.append(best_score)
self.iterations.append(iteration)
self.convergence_ax.clear()
self.convergence_ax.plot(self.iterations, self.best_scores, 'b-', linewidth=2)
self.convergence_ax.set_xlabel('Iteration')
self.convergence_ax.set_ylabel('Best Fitness')
self.convergence_ax.set_title('PSO Convergence')
self.convergence_ax.grid(True, alpha=0.3)
self.convergence_canvas.draw()
# Update particles plot (2D projection)
self.particles_ax.clear()
if particles.shape[1] >= 2:
self.particles_ax.scatter(particles[:, 0], particles[:, 1],
c='blue', alpha=0.6, s=20)
self.particles_ax.scatter([position[0]], [position[1]],
c='red', s=100, marker='*', label='Global Best')
self.particles_ax.set_xlabel('Parameter 1')
self.particles_ax.set_ylabel('Parameter 2')
self.particles_ax.set_title('Particle Positions')
self.particles_ax.legend()
self.particles_ax.grid(True, alpha=0.3)
self.particles_canvas.draw()
# Update fitness landscape for 2D problems
if particles.shape[1] == 2:
self.update_fitness_landscape(position, particles)
# Update progress
self.progress.setValue(iteration + 1)
# Update results
self.results_text.setText(
f"Iteration: {iteration + 1}\n"
f"Best Fitness: {best_score:.6f}\n"
f"Best Position: {position}\n"
)
def update_fitness_landscape(self, best_position, particles):
self.landscape_ax.clear()
# Create meshgrid for fitness landscape
x = np.linspace(-5, 5, 50)
y = np.linspace(-5, 5, 50)
X, Y = np.meshgrid(x, y)
# Calculate fitness for each point
Z = np.zeros_like(X)
for i in range(X.shape[0]):
for j in range(X.shape[1]):
Z[i, j] = self.pso_thread.fitness([X[i, j], Y[i, j]],
self.pso_thread.problem_type)
# Plot contour
contour = self.landscape_ax.contourf(X, Y, Z, levels=50, alpha=0.8)
self.landscape_ax.contour(X, Y, Z, levels=10, colors='black', alpha=0.3)
# Plot particles
self.landscape_ax.scatter(particles[:, 0], particles[:, 1],
c='white', s=20, alpha=0.7)
self.landscape_ax.scatter([best_position[0]], [best_position[1]],
c='red', s=100, marker='*', label='Global Best')
self.landscape_ax.set_xlabel('Parameter 1')
self.landscape_ax.set_ylabel('Parameter 2')
self.landscape_ax.set_title('Fitness Landscape')
self.landscape_canvas.draw()
def optimization_finished(self, data):
self.run_button.setEnabled(True)
self.stop_button.setEnabled(False)
self.progress.setValue(self.max_iter.value())
final_text = (
f"Optimization Completed!\n"
f"Final Fitness: {data['final_score']:.8f}\n"
f"Optimal Parameters: {data['final_position']}\n"
f"Total Iterations: {self.max_iter.value()}\n"
f"Number of Particles: {self.num_particles.value()}"
)
self.results_text.setText(final_text)
def main():
app = QApplication(sys.argv)
window = PSOWindow()
window.show()
sys.exit(app.exec_())
if __name__ == '__main__':
main() |