import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
from datetime import datetime

# Read Phase 2 CSV file
df = pd.read_csv('ONYX LN2 2026 PH 2.csv')

# Set column names based on the actual data structure
df.columns = ['Time', 'Temp', 'Alm1', 'mA', 'Alm2', 'Empty']

# Convert Time to datetime
df['Time'] = pd.to_datetime(df['Time'])

# Convert mA to SCFH: 4mA = 0, 20mA = 10000
# Formula: SCFH = (mA - 4) / 16 * 10000
df['SCFH'] = (df['mA'] - 4) * (10000 / 16)

# Filter out negative flow rates for analysis (they might be measurement errors)
df_clean = df[df['SCFH'] >= 0]

# Calculate key metrics for Phase 2
peak_idx = df_clean['SCFH'].idxmax()
peak_time = df_clean.loc[peak_idx, 'Time']
peak_flow = df_clean.loc[peak_idx, 'SCFH']

# Trends by day of week
df_clean['DayOfWeek'] = df_clean['Time'].dt.day_name()
df_clean['IsWeekend'] = df_clean['Time'].dt.dayofweek >= 5

weekend_avg = df_clean[df_clean['IsWeekend']]['SCFH'].mean()
weekday_avg = df_clean[~df_clean['IsWeekend']]['SCFH'].mean()

# Time of day
df_clean['Hour'] = df_clean['Time'].dt.hour
hourly_avg = df_clean.groupby('Hour')['SCFH'].mean()

# Calculate overall statistics
total_records = len(df)
total_records_clean = len(df_clean)
start_time = df['Time'].min()
end_time = df['Time'].max()
duration_days = (end_time - start_time).days

# Calculate the average flow rate
avg_flow = df_clean['SCFH'].mean()

# Calculate variance and standard deviation
std_dev = df_clean['SCFH'].std()
min_flow = df_clean['SCFH'].min()
max_flow = df_clean['SCFH'].max()

print("=== PHASE 2 ANALYSIS RESULTS (Clean Data) ===")
print(f"Duration: {start_time.strftime('%m/%d/%y')} to {end_time.strftime('%m/%d/%y')} ({duration_days} days)")
print(f"Total Records: {total_records}")
print(f"Clean Records: {total_records_clean}")
print(f"Records with negative flow rates: {total_records - total_records_clean}")
print(f"Average Flow Rate: {avg_flow:.2f} SCFH")
print(f"Peak Usage: {peak_flow:.2f} SCFH at {peak_time}")
print(f"Minimum Flow: {min_flow:.2f} SCFH")
print(f"Maximum Flow: {max_flow:.2f} SCFH")
print(f"Standard Deviation: {std_dev:.2f} SCFH")
print(f"Weekend Average: {weekend_avg:.2f} SCFH")
print(f"Weekday Average: {weekday_avg:.2f} SCFH")

# Create a detailed hourly analysis
print("\nHourly averages:")
hourly_analysis = []
for hour in range(24):
    avg_hourly = hourly_avg.get(hour, 0)
    hourly_analysis.append((hour, avg_hourly))
    print(f"  {hour:02d}:00 - {avg_hourly:.2f} SCFH")

print("\n=== COMPARISON WITH PHASE 1 ===")
print("Phase 1 Results (from HTML report):")
print("- Variance: -0.87% (Logger High)")
print("- Peak Usage: 1062.50 SCFH on Feb 27, 2026 at 11:16 AM")
print("- Weekday Average: 430.76 SCFH")
print("- Weekend Average: 386.53 SCFH")
print("- Data Resolution: 4.5 minutes (corrected to 4 minutes in Phase 2)")

# Calculate variance between phases
phase1_peak = 1062.50
phase2_peak = peak_flow
peak_variance = ((phase2_peak - phase1_peak) / phase1_peak) * 100

print(f"\nPeak Usage Variance: {peak_variance:.2f}%")
if abs(peak_variance) <= 1:
    print("✓ Peak usage variance within acceptable limits (±1%)")
else:
    print("⚠ Peak usage variance exceeds acceptable limits (>1%)")

# Create some visualizations
plt.figure(figsize=(15, 10))

# Plot 1: Time series of flow rates (clean data only)
plt.subplot(2, 2, 1)
plt.plot(df_clean['Time'], df_clean['SCFH'], linewidth=0.8, alpha=0.7, color='blue')
plt.title('LN2 Gas Consumption Over Time (Phase 2) - Clean Data')
plt.xlabel('Date/Time')
plt.ylabel('Flow Rate (SCFH)')
plt.xticks(rotation=45)
plt.grid(True, alpha=0.3)

# Plot 2: Hourly average flow rates
plt.subplot(2, 2, 2)
hours = list(range(24))
avg_hourly = [hourly_avg.get(h, 0) for h in hours]
plt.plot(hours, avg_hourly, marker='o', linewidth=2, markersize=4, color='green')
plt.title('Average Flow Rate by Hour of Day')
plt.xlabel('Hour of Day')
plt.ylabel('Average Flow Rate (SCFH)')
plt.xticks(range(0, 24, 2))
plt.grid(True, alpha=0.3)

# Plot 3: Weekday vs Weekend comparison
plt.subplot(2, 2, 3)
days = ['Weekday', 'Weekend']
averages = [weekday_avg, weekend_avg]
bars = plt.bar(days, averages, color=['#1f77b4', '#ff7f0e'])
plt.title('Average Flow Rate by Day Type')
plt.ylabel('Flow Rate (SCFH)')
plt.grid(True, alpha=0.3, axis='y')

# Add value labels on bars
for bar, avg in zip(bars, averages):
    plt.text(bar.get_x() + bar.get_width()/2, bar.get_height() + 10,
             f'{avg:.1f}', ha='center', va='bottom')

# Plot 4: Flow rate distribution (clean data)
plt.subplot(2, 2, 4)
plt.hist(df_clean['SCFH'], bins=30, alpha=0.7, color='skyblue', edgecolor='black')
plt.title('Distribution of Flow Rates (Clean Data)')
plt.xlabel('Flow Rate (SCFH)')
plt.ylabel('Frequency')
plt.grid(True, alpha=0.3)

plt.tight_layout()
plt.savefig('phase2_analysis_clean.png', dpi=300, bbox_inches='tight')
plt.close()

print("\nAnalysis complete. Visualizations saved as 'phase2_analysis_clean.png'")

# Also create a summary for the HTML report
print("\n=== SUMMARY FOR HTML REPORT ===")
print("Phase 2 Summary:")
print(f"- Duration: {start_time.strftime('%m/%d/%y')} to {end_time.strftime('%m/%d/%y')}")
print(f"- Average Flow Rate: {avg_flow:.2f} SCFH")
print(f"- Peak Usage: {peak_flow:.2f} SCFH")
print(f"- Weekend Average: {weekend_avg:.2f} SCFH")
print(f"- Weekday Average: {weekday_avg:.2f} SCFH")

# Check if data meets requirements
if peak_variance <= 1:
    print("- Data variance within acceptable limits")
else:
    print("- Note: Peak usage variance exceeds acceptable limits (>1%)")