Interpreting Charts¶
Learn to read and analyze the temperature and humidity visualizations in GreenCop.
Chart Types¶
Bar Charts¶
Used For: Comparing current readings across multiple sensors
Location: Dashboard main area
How to Read: - Each sensor has two bars side-by-side - Left bar (orange): Current temperature - Right bar (blue): Current humidity - Y-axis: Value (°C or %) - X-axis: Sensor names
Insights: - Quickly identify outlier sensors - Compare environmental conditions across rooms - Spot sensors reporting abnormal values
Line Charts¶
Used For: Viewing trends over time
Locations: Dashboard (historical trends), Sensor Detail Page
How to Read: - X-axis: Time (chronological) - Y-axis: Temperature (°C) or Humidity (%) - Orange line: Temperature progression - Blue line: Humidity progression - Hover for exact values at any point
Insights: - Identify daily cycles - Spot gradual changes - Detect anomalies or sudden spikes - Correlate temperature and humidity
Reading Patterns¶
Normal Temperature Patterns¶
Daily Cycle:
- Rises during day (people, activity) - Peaks afternoon (2-4 PM) - Drops overnight - Smooth curvesStable Environment:
- Flat or minor variation - Indicates good HVAC control - Consistent loadAbnormal Temperature Patterns¶
Cooling Failure:
- Sharp upward spike - No recovery - Immediate action neededGradual Overheating:
- Slow increase over days - Cooling capacity declining - Plan maintenanceSensor Malfunction:
- Erratic readings - Impossible fluctuations - Check sensor hardwareHumidity Patterns¶
Normal Humidity: - Relatively stable (±5%) - Minor daily variations - Gradual changes only
Concerning Humidity: - Sudden spikes (>10% in 1 hour) - Sustained above 60% - Wild fluctuations - Inverse correlation with temp (unusual)
Chart Interactions¶
Hover Tooltips¶
Desktop: 1. Move mouse over chart 2. Tooltip appears with: - Exact timestamp - Temperature value - Humidity value 3. Follows cursor
Mobile: - Tap and hold on data point - Tooltip appears - Release to dismiss
Time Range Selection¶
Sensor Detail Page: - Three buttons: "1 Hour", "24 Hours", "7 Days" - Click to change displayed range - Active button highlighted - Chart re-renders with new data
Effect on Chart: - 1h: Highly detailed, many points - 24h: Moderate detail, daily patterns visible - 7d: Sampled data, weekly trends
Legend Interaction¶
Desktop Charts: - Click legend item (Temperature/Humidity) - Hides/shows that data series - Useful for focusing on one metric - Click again to restore
Comparative Analysis¶
Multiple Sensors¶
On Dashboard: 1. View bar chart 2. Compare heights of bars 3. Identify which sensors are hottest/coolest 4. Investigate outliers
Questions to Ask: - Why is one sensor 5°C hotter? - Are sensors in same room similar? - Is one sensor closer to heat source?
Temperature vs Humidity¶
Look For: - Inverse relationship (temp up, humidity down) - Correlated changes (both rising together) - Independent variation
Typical Patterns: - Temp ↑ Humidity ↓: Normal (hot air holds more moisture) - Both ↑: External humidity source (leak, weather) - Both ↓: Dry, cool conditions
Alert Correlation¶
Chart Context for Alerts¶
When alert triggers: 1. Go to sensor detail page 2. View chart leading up to alert 3. Identify: - Gradual buildup vs sudden spike - Time of day pattern - Related humidity change
Root Cause Analysis¶
Example: Temperature alert at 2 PM
Chart Shows: - Gradual rise since 10 AM - Humidity stable - Previous days had same pattern
Conclusion: Insufficient cooling during peak hours
Action: Increase HVAC capacity or cooling schedule
Performance Indicators¶
Healthy System¶
Charts Show: - Smooth lines (no jagged edges) - Predictable daily cycles - All sensors within 3°C of each other - Humidity 40-50% stable - No alerts
Warning Signs¶
Charts Show: - Increasing baseline over days - One sensor diverging from others - Erratic, noisy data - Humidity climbing - Frequent threshold breaches
Best Practices¶
Regular Review¶
- Check dashboard daily
- Review sensor detail pages weekly
- Compare week-over-week trends
- Investigate any changes in patterns
Pattern Recognition¶
- Learn your normal baseline
- Document expected daily cycles
- Note seasonal variations
- Set expectations for "normal"
Documentation¶
- Screenshot abnormal patterns
- Note dates of significant events
- Correlate with facility changes
- Build historical knowledge base
Troubleshooting Charts¶
Chart Not Loading¶
Symptoms: Blank chart area or loading spinner
Solutions: - Refresh page - Check internet connection - Verify backend API is running - Look for JavaScript errors in console
No Data Points¶
Symptoms: Empty chart with axes
Causes: - Sensor not publishing data - Time range has no readings - BigQuery ingestion issue
Solutions: - Verify sensor is online - Select different time range - Check BigQuery table for data
Choppy/Jagged Lines¶
Normal: Actual sensor variation Abnormal: Chart rendering issue, refresh page