HEXDATA

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
# Parameters
weeks = 52 * 15  # total number of weeks for data collection (3 years)
cleaning_interval_weeks = 52  # number of weeks between each cleaning (approximately 6 months)
initial_efficiency = 95  # starting efficiency (%)
base_scaling_rate = 0.6  # base efficiency drop per week due to scaling (%)
min_efficiency = 65  # Minimum efficiency throughout operation (70%)
recovery_decline_rate = 1  # Rate of decline in maximum efficiency after each cleaning

# Generate week numbers and dates
week_number = np.arange(1, weeks + 1)
dates = pd.date_range(start="2008-01-01", periods=weeks, freq="W-SUN")  # Create date range for 3 years (weekly)

# Simulate seasonal effects using a sine wave to modulate the scaling rate
seasonal_effects = 1.1*np.sin(2 * np.pi* (week_number % 52) / 52)  # One full cycle per year
scaling_rate = base_scaling_rate + 0.2 * seasonal_effects  # Modulate scaling rate with seasonality

# Initialize efficiency array
efficiency = np.zeros(weeks)
max_efficiency = initial_efficiency  # Maximum efficiency that decreases slightly after each cleaning

# Simulate efficiency drop and recovery with seasonal effects and reduced noise
for week in range(weeks):
    if week % cleaning_interval_weeks == 0 and week != 0:
        # After cleaning, efficiency recovers to near initial efficiency with reduced noise
        efficiency[week] = np.random.uniform(0.9, 0.99)*max_efficiency + np.random.uniform(-0.5, 0.5)
    else:
        if week == 0:
            efficiency[week] = np.random.uniform(0.82, 0.9)*initial_efficiency + np.random.uniform(-0.5, 0)  # Initial value with reduced noise
        else:
            efficiency[week] = efficiency[week - 1] - np.random.uniform(0.9, 0.99)*scaling_rate[week] + np.random.uniform(0, 0.5)  # Drop with reduced noise
            # Ensure efficiency does not drop below minimum
            '''if efficiency[week] < min_efficiency:
                efficiency[week] = np.random.uniform(1, 1.1)*min_efficiency + np.random.uniform(0, 0.5)'''

# Create a DataFrame
data = pd.DataFrame({
    'Date': dates,
    'Week Number': week_number,
    'Efficiency (%)': efficiency
})

# Display the first few rows
print(data)

# Plot the efficiency over time
plt.figure(figsize=(12, 6))
plt.plot(data['Date'][::], data['Efficiency (%)'][::],'o')
plt.title('Heat Exchanger Efficiency Over Time (10 Years of Weekly Data)')
plt.xlabel('Date')
plt.ylabel('Efficiency (%)')
plt.legend(['Efficiency'])
plt.ylim(min_efficiency - 10, initial_efficiency + 10)
plt.grid(True)
plt.show()

# Save the data to a CSV file
%cd /content/drive/MyDrive/
data.to_csv('HEXeff_data.csv')
print(data.shape)