node-red-contrib-condition-monitoring 0.1.2

Node-RED Nodes for anomaly detection, predictive maintenance, and time series analysis

npm install node-red-contrib-condition-monitoring

node-red-contrib-condition-monitoring

A comprehensive Node-RED module for anomaly detection, predictive maintenance, and time series analysis.

---

Project Status: BETA (v0.1.2)

This is the first public release - currently in beta testing.

• First Release: All core features are implemented and functional
• Beta Phase: Undergoing real-world validation and testing
• Feedback Welcome: Please report issues and share your experience
• API May Change: Breaking changes possible before v1.0 stable release
• Production Use: Use with caution and proper testing in your environment
• Goal: Reach v1.0.0 stable after community feedback and validation

---

Important Disclaimer

This software is provided for condition monitoring and predictive maintenance purposes.

• NOT a replacement for safety-critical systems
• NOT suitable as the sole means of safety decision-making
• Should be used as an additional monitoring layer
• Always validate results with domain experts
• Follow proper safety protocols and regulations for your industry

Use at your own risk. See LICENSE file for full legal terms.

Features

• 10 Anomaly Detection Methods - Z-Score, IQR, Moving Average, Isolation Forest, Threshold, Percentile, EMA, CUSUM, Multi-Value
• 7 Predictive Maintenance Nodes - Trend Prediction (RUL), FFT Analysis, Vibration Features, Health Index, Rate of Change, Peak Detection, Correlation Analysis
• Real-time Processing - Continuous data stream analysis
• Ready-to-Use Examples - 5 complete example flows in /examples directory
• Fully Documented - Built-in help for every node

Installation

npm install node-red-contrib-condition-monitoring

Or install directly from Node-RED:

1. Menu → Manage palette
2. Install tab
3. Search for node-red-contrib-condition-monitoring
4. Click install

Quick Start

With Docker Compose (Recommended)

# Start Node-RED with the module
docker-compose up -d

# Access at http://localhost:1880

Import Example Flows

1. Open Node-RED: http://localhost:1880
2. Menu → Import → Examples
3. Select one of the 4 example flows:
   • Example 1: Motor Monitoring (Z-Score, Trend Prediction, Health Index, Correlation)
   • Example 2: Bearing Vibration Analysis (FFT, Peak Detection, IQR)
   • Example 3: Process Monitoring (Threshold, CUSUM, EMA, Percentile)
   • Example 4: ML Anomaly Detection (Isolation Forest)

See /examples/README.md for detailed documentation of all examples.

Available Nodes

Anomaly Detection (10 Nodes)

Node	Method	Best For	Output
Z-Score	Statistical	General purpose anomalies	2 outputs (normal/anomaly)
IQR	Quartile-based	Robust to outliers	2 outputs
Moving Average	Trend-based	Gradual changes	2 outputs
Isolation Forest	Machine Learning	Complex patterns	2 outputs
Threshold	Min/Max limits	Hard boundaries	2 outputs
Percentile	Rank-based	Dynamic thresholds	2 outputs
EMA	Exponential smoothing	Recent changes	2 outputs
CUSUM	Cumulative sum	Drift detection	2 outputs
Multi-Value Anomaly	Any method	Multiple sensors	2 outputs
Multi-Value Splitter	Utility	Split sensor arrays	1 output

Predictive Maintenance (7 Nodes)

Node	Function	Output	Use Case
Trend Prediction	RUL calculation	Future values, time-to-threshold	"Motor fails in 48h"
FFT Analysis	Frequency analysis	Peaks, spectral features	Bearing fault detection
Vibration Features	Feature extraction	RMS, Crest Factor, Kurtosis, Skewness	Comprehensive vibration analysis
Health Index	Multi-sensor aggregation	0-100 health score	Overall equipment status
Rate of Change	Derivative analysis	Speed of change, acceleration	Rapid temperature rise
Peak Detection	Impact detection	Peak events	Bearing impacts, shocks
Correlation Anomaly	Sensor relationship	Correlation coefficient	Temp vs Power relationship

Which Node Should I Use?

For Anomaly Detection:

Simple Use Cases:

• Hard boundaries (min/max)? → Threshold Anomaly

  • Example: Temperature must stay between 20-80°C

• Statistical outliers? → Z-Score or IQR Anomaly

  • Z-Score: Best for normally distributed data
  • IQR: More robust, works with any distribution

Trend & Drift Detection:

• Slow gradual changes? → CUSUM Anomaly

  • Example: Pump flow slowly decreasing over days

• Moving baseline? → Moving Average or EMA Anomaly

  • Moving Average: Equal weight to all values in window
  • EMA: Recent values weighted more (faster response)

Advanced Cases:

• Complex patterns, no clear rules? → Isolation Forest

  • Machine learning approach, learns automatically

• Extreme values only? → Percentile Anomaly

  • Example: Detect only top 5% and bottom 5%

Multiple Sensors:

• Analyze multiple sensors together? → Multi-Value Anomaly
• Split sensor array for separate processing? → Multi-Value Splitter

---

For Predictive Maintenance:

Vibration Analysis:

• Time-domain features (RMS, Crest Factor, Kurtosis)? → Vibration Features

  • Best for: Bearing condition, overall vibration health

• Frequency analysis (FFT, harmonics)? → FFT Analysis

  • Best for: Finding specific fault frequencies (bearing, gear defects)

• Count impacts/shocks? → Peak Detection

  • Best for: Impact counting, shock detection

Trend & Prediction:

• Predict when threshold will be reached? → Trend Prediction

  • Calculates Remaining Useful Life (RUL)
  • Example: "Temperature will exceed 100°C in 48 hours"

• Measure rate of degradation? → Rate of Change

  • Detects rapid changes (acceleration)
  • Example: "Temperature rising 5°C per hour"

Health Assessment:

• Single health score from multiple sensors? → Health Index

  • Combines temperature, vibration, pressure into 0-100% score

• Validate sensor relationships? → Correlation Anomaly

  • Example: Check if temperature and power consumption correlate correctly

---

Quick Decision Tree:

Do you have historical data?
├─ NO  → Start with Threshold or Z-Score
└─ YES → Continue below

Is it vibration data?
├─ YES → Vibration Features + FFT Analysis + Peak Detection
└─ NO  → Continue below

Single sensor or multiple?
├─ SINGLE → Z-Score / Moving Average / CUSUM
└─ MULTIPLE → Multi-Value Splitter + Individual Analysis → Health Index

Need to predict failures?
└─ YES → Trend Prediction + Rate of Change + Health Index

Usage Examples

Simple Temperature Monitoring

[MQTT Sensor] → [Z-Score Anomaly] → [Normal] → [Dashboard]
                                   → [Anomaly] → [Alarm]

Motor Predictive Maintenance

[Sensors] → [Multi-Value Splitter] → [Z-Score]
                                   → [Trend Prediction] → RUL Display
                                   → [FFT Analysis] → Frequency Chart
         → [Health Index] → Health Dashboard

Bearing Vibration Analysis

[Vibration Sensor] → [Vibration Features] → RMS, Crest Factor, Kurtosis
                   → [FFT Analysis] → Frequency Peaks
                   → [Peak Detection] → Impact Counter
                   → [IQR Anomaly] → Outlier Detection

Documentation

Node-Specific Help

Each node has comprehensive built-in documentation:

1. Drag node to canvas
2. Select it
3. Click Info in sidebar
4. Read detailed docs with examples

Additional Documentation

• examples/README.md - Detailed guide for all 5 example flows
• PAYLOAD_FORMAT.md - Input format specifications
• MULTI_VALUE.md - Working with multiple sensors
• DOCKER.md - Docker deployment guide

Node Configuration

Example: Z-Score Anomaly

// Input
msg.payload = 42.5;

// Output (Anomaly) - v0.1.2+
{
  "payload": 42.5,
  "zScore": 3.2,
  "mean": 35.0,
  "stdDev": 2.3,
  "isAnomaly": true,
  "severity": "critical",     // NEW: "normal", "warning", or "critical"
  "threshold": 3.0,
  "warningThreshold": 2.1,    // NEW
  "bufferSize": 100,          // NEW
  "windowSize": 100           // NEW
}

// Reset the node
msg.reset = true;  // NEW: Clears buffer and restarts learning

Example: Trend Prediction

// Input
msg.payload = 75.2;  // Temperature
msg.timestamp = Date.now();

// Output
{
  "payload": 75.2,
  "trend": "increasing",
  "slope": 0.5,
  "predictedValues": [76.2, 76.7, 77.2, ...],
  "timeToThreshold": 172800000,  // 48 hours in ms
  "stepsToThreshold": 96
}

Example: FFT Analysis

// Input (continuous stream at 1000 Hz)
msg.payload = 0.45;  // Vibration amplitude

// Output
{
  "payload": 0.45,
  "peaks": [
    { "frequency": 30, "magnitude": 0.5 },
    { "frequency": 157, "magnitude": 0.3 }  // Bearing fault!
  ],
  "dominantFrequency": 30,
  "features": {
    "spectralCentroid": 85.2,
    "crestFactor": 3.5,  // High = impulsive behavior
    "rms": 0.42
  }
}

Learning Path

1. Start Simple - Import Example 3 (Process Monitoring)
2. Learn Basics - Understand threshold and Z-Score detection
3. Advanced Methods - Try FFT and Trend Prediction
4. Combine Nodes - Build complete predictive maintenance system

Real-World Applications

• Manufacturing - Machine health monitoring, quality control
• Energy - Battery degradation, power quality monitoring
• Automotive - Vehicle diagnostics, fleet management
• HVAC - Climate system optimization, energy efficiency
• Water Treatment - Pump monitoring, leak detection
• Aerospace - Engine monitoring, structural health
• Medical - Equipment monitoring, vital sign analysis

Technical Details

Statistical Methods

Method	Type	Complexity	Speed	Accuracy
Threshold	Rule-based	Low	Fast	Medium
Z-Score	Statistical	Low	Fast	High
IQR	Statistical	Medium	Fast	High
Percentile	Statistical	Medium	Fast	High
Moving Average	Trend	Low	Fast	Medium
EMA	Trend	Low	Fast	Medium
CUSUM	Cumulative	Medium	Fast	High
Isolation Forest	ML	High	Medium	Very High

Predictive Maintenance Capabilities

Feature	Node	Output
RUL Estimation	Trend Prediction	Time until failure
Frequency Analysis	FFT Analysis	Fault frequencies
Overall Health	Health Index	0-100 score
Change Speed	Rate of Change	Derivative
Impact Events	Peak Detection	Peak count
Sensor Validation	Correlation	Relationship strength

Development

Run with Docker

# Development mode (with hot-reload)
docker-compose -f docker-compose.dev.yml up

# Production mode
docker-compose up

Local Development

# Install dependencies
npm install

# Link to Node-RED
npm link
cd ~/.node-red
npm link node-red-contrib-condition-monitoring

# Restart Node-RED
node-red-restart

Dependencies

Required

• Node-RED >= 1.0.0
• Node.js >= 14.0.0

Optional

• ml-isolation-forest - For Isolation Forest node (falls back to Z-Score if not available)
• simple-statistics - For advanced statistical functions

Contributing

Contributions are welcome! Please:

1. Fork the repository
2. Create a feature branch
3. Add tests if applicable
4. Submit a pull request

License

MIT License - see LICENSE file for details.

Author

blanpa

Issues & Support

• Bug Reports: Open an issue on GitHub
• Questions: Check /examples/README.md first
• Feature Requests: Submit via GitHub issues

Roadmap

• Dashboard UI components
• Export/import of trained models
• MQTT examples
• Real-time charting integration
• More ML algorithms (LSTM, Prophet)
• Automated reporting

Show Your Support

If you find this useful, please consider:

• Starring the repository
• Sharing with others
• Reporting bugs
• Suggesting features

---

Made with love for the Node-RED community

Get Started: Import an example flow and start monitoring in minutes!