Skip to content

Data Analytics

Analyze sensor data with interactive charts, statistics, and historical trends.

Overview

GreenCop provides comprehensive data analytics to help you understand environmental patterns, identify trends, and make data-driven decisions about your infrastructure.

Data Sources

Real-Time Data

  • Latest sensor readings from BigQuery
  • Updated every few seconds
  • Displayed on dashboard and sensor detail pages

Historical Data

  • All sensor readings stored in BigQuery
  • Partitioned by date for fast queries
  • Queryable via API for custom time ranges

Analytics Features

Statistics

Sensor Statistics API: GET /api/v1/data/stats/{sensor_id}

Provides aggregated stats for multiple time windows: - Last 1 hour - Last 24 hours - Last 7 days - Last 30 days

Metrics Provided: - Average temperature - Minimum temperature - Maximum temperature - Average humidity - Minimum humidity - Maximum humidity

Time-Series Charts

Temperature Trends: - Line chart showing temperature over time - Customizable time ranges (1h, 24h, 7d) - Smooth interpolation - Interactive tooltips

Humidity Trends: - Line chart showing humidity over time - Same time range options - Parallel to temperature for correlation analysis

Comparative Charts: - Bar charts comparing all sensors - Side-by-side temperature and humidity - Identify outliers quickly

Data Visualization

Chart Types

Line Charts (Trends): - Best for: Time-series analysis - Shows: Patterns, cycles, anomalies - Used in: Sensor detail page, dashboard

Bar Charts (Comparison): - Best for: Multi-sensor comparison - Shows: Current state across fleet - Used in: Dashboard

Interactive Features

  • Hover Tooltips: Exact values on mouse over
  • Zoom: Click and drag to zoom (future)
  • Legend: Toggle series visibility
  • Responsive: Adapts to screen size

Querying Data

Via Dashboard

  1. Navigate to sensor detail page
  2. Select time range (1h/24h/7d)
  3. View charts automatically
  4. No manual query needed

Via API

Latest Reading: 

GET /api/v1/data/latest/{sensor_id}

Historical Range: 

GET /api/v1/data/historical/{sensor_id}?start_time=2025-01-01T00:00:00Z&end_time=2025-01-15T23:59:59Z

Multi-Sensor: 

POST /api/v1/data/multi-sensor
Body: {"node_ids": ["sensor1", "sensor2"]}

Statistics: 

GET /api/v1/data/stats/{sensor_id}

Data Insights

Identifying Patterns

Daily Cycles: - Temperature typically rises during day - Drops at night - Chart shows regular wave pattern

Anomalies: - Sudden spikes or drops - Sustained high/low periods - Irregular patterns

Trends: - Gradual increase over days (cooling failure) - Gradual decrease (over-cooling) - Seasonal variations

Use Cases

Capacity Planning: - Analyze peak usage times - Size cooling infrastructure - Plan maintenance windows

Efficiency: - Identify over-cooling periods - Optimize HVAC schedules - Reduce energy costs

Compliance: - Generate reports for audits - Prove SLA adherence - Document environmental controls

Data Export

Future Feature

CSV/Excel export is planned. Currently, use API to retrieve data programmatically.

Current Options

API Export: - Query historical data endpoint - Parse JSON response - Process in your own tools

BigQuery Direct: - Connect to BigQuery dataset - Use SQL for complex analytics - Export to Google Sheets, Data Studio, etc.

Performance

Optimization

  • BigQuery partitioned by day
  • Clustered by node_id for fast lookups
  • Dashboard limits to last 50 points
  • Caching planned for future release

Query Limits

  • Historical data: Up to 30 days via API
  • Statistics: Pre-computed for common windows
  • BigQuery: Unlimited with direct access

Best Practices

For Accurate Analysis

  1. Sufficient Data: Wait for 24h of data before analyzing trends
  2. Time Zones: All timestamps in UTC, convert for local analysis
  3. Outliers: Investigate extreme values, may be sensor errors
  4. Context: Consider external factors (weather, usage patterns)

Interpretation

Normal Patterns: - Smooth curves without sudden changes - Daily temperature cycles of 3-5°C - Humidity relatively stable

Warning Signs: - Sharp temperature spikes - Sustained above-threshold conditions - Erratic, noisy data (sensor issue)

Next Steps