DeviceBoard Digital Twin Implementation User Guide
DeviceBoard – Documentation
DeviceBoard – Digital Twin Implementation User Guide
DeviceBoard provides a powerful Digital Twin framework that enables organizations to create virtual replicas of physical devices, machines, assets, buildings, or entire operational environments. These Digital Twins unify telemetry, attributes, alarms, rules, analytics, and AI insights into a single, real-time data model.
This guide explains how to configure, use, and extend Digital Twins in DeviceBoard, including:
- Digital Twin concepts
- Twin creation workflow
- Data mapping (telemetry, attributes, control commands)
- Twin modeling using Device Models & Asset Models
- Twin relationships (hierarchy & topology)
- AI-driven enhancements
- RulesFlow processing
- Dashboards & analytics visualization
- Integration with provisioning and automation
1. Introduction to Digital Twins in DeviceBoard
A Digital Twin in DeviceBoard is a virtual representation of:
- A device (sensor, machine, vehicle, etc.)
- An asset (building, floor, equipment, location)
- A logical system (production line, pumping station, solar field)
- A fleet or facility (multi-layer composite twin)
The Digital Twin mirrors:
- Real-time telemetry
- Configurations
- Operational parameters
- Behaviors & states
- Alarms & events
- AI predictions & anomalies
DeviceBoard makes Digital Twin creation seamless through Device Models, Asset Models, RulesFlow, AI Models, and visual dashboards.
2. Key Concepts of Digital Twin Architecture
2.1 Physical Layer
Real devices or assets that generate telemetry or status information.
Examples:
- Pump
- Energy meter
- Temperature/humidity sensor
- Industrial PLC
- HVAC equipment
2.2 Connectivity Layer
Defines how data reaches DeviceBoard:
- Built-in protocols (MQTT, CoAP, HTTP, LwM2M, SNMP)
- IoT Gateway (Modbus, OPC-UA, CAN, BACnet, BLE, custom drivers)
- LoRaWAN
- Sigfox
- NB-IoT / Cellular IoT
- Proprietary integrations
2.3 Digital Twin Layer
A virtual entity (device or asset) that stores:
- Telemetry
- Server/client/shared attributes
- Computed values
- Alarms
- AI predictions
- Health scores
- States
- Metadata
- Hierarchical relationships
2.4 Processing & Analytics Layer
Digital Twins integrate tightly with:
- RulesFlow (logic, enrichment, alarm generation, data routing)
- AI Models (supervised & unsupervised)
- Time-series analytics
- Anomaly detection
- Aggregations & computations
2.5 Visualization Layer
Digital Twins are visualized using:
- Dashboards
- Widgets
- 2D/3D layout maps
- Trend charts
- Health indicators
- Alarm panels
- Digital Twin UI components
3. Digital Twin Components in DeviceBoard
3.1 Telemetry
Real-time sensor data mapped to:
- Numeric values (temp, voltage, current)
- Boolean states (door open, motor running)
- Strings (status messages)
- Complex structures (JSON payloads)
3.2 Attributes
Attributes define configuration or static data:
Types:
- Client Attributes (from device → DeviceBoard)
- Server Attributes (from DeviceBoard → device)
- Shared Attributes (two-way attributes for configuration)
3.3 Commands (RPC / Control Operations)
Digital Twins support:
- Parameter updates
- Start/stop operations
- Actuator commands
- Diagnostic commands
These are defined in Device Models.
3.4 Alarm Rules
Digital Twins automatically detect:
- Threshold violations
- State changes
- AI anomalies
- Predictive risk indicators
- Connectivity issues
Alarm states are maintained in the twin’s context.
3.5 AI & Behavioral Models
Digital Twins incorporate:
- Supervised predictions
- Anomaly detection (unsupervised models)
- Health scoring
- Predictive maintenance indicators
These enrich the twin with behavioral intelligence.
3.6 Relationships & Hierarchies
DeviceBoard supports:
- Device → Asset mapping
- Asset → Sub-Asset mapping
- Multi-level facility modeling
- Parent-child digital twins
Example:
Factory → Production Line → Machine → Sub-Component → Sensor
4. Creating a Digital Twin in DeviceBoard
A Digital Twin is created automatically when a device or asset is registered.
4.1 Creating a Device Twin
Step-by-Step
- Navigate to Devices → Add Device
- Select a Device Model
- Set device name, description
- Configure connectivity
- Assign device groups
- Save → Digital Twin is automatically created
4.2 Creating an Asset Twin
Steps
- Navigate to Assets → Add Asset
- Select an Asset Model
- Provide metadata
- Assign asset groups
- Save → Asset Twin is created
Assets can represent:
- Buildings
- Rooms
- Transformers
- Vehicles
- Workstations
5. Device Models & Asset Models for Digital Twins
Device Models define:
- Expected telemetry keys
- Attribute structure
- Connectivity method
- RulesFlow pipeline
- Alarm configuration
- Security credentials
- Command definitions
- AI model assignments
Asset Models define:
- Child assets
- Device relationships
- Virtual telemetry (computed values)
- Maintenance attributes
- Status & behavioral rules
This allows:
Model Once → Use Repeatedly
Any new device using the same Device Model will automatically inherit its Digital Twin structure.
6. Mapping Data to Digital Twin
Once the device sends telemetry via supported protocol, DeviceBoard:
- Receives raw data
- Maps data to telemetry fields
- Processes via RulesFlow
- Stores data in the time-series database
- Updates Digital Twin’s live state
- Triggers alarms or AI
- Updates dashboards
7. RulesFlow for Digital Twin Behavior
RulesFlow is the logic engine that powers Digital Twins.
Capabilities:
- Telemetry transformation
- Unit conversion
- Derived values computation
- AI inference
- Anomaly scoring
- Alarm generation
- Event routing
- Device-to-asset data propagation
- Twin state updates
Example Flows:
- Compute running hours
- Calculate energy cost
- Determine system efficiency
- Apply predictive models
- Generate alarms when twin state changes
8. Digital Twin Dashboards & Visualization
DeviceBoard offers rich visualization tools for Digital Twins.
8.1 Device Twin Dashboard
Includes:
- Latest telemetry
- Time-series charts
- AI predictions
- Anomaly graph
- Health score widget
- Attribute tables
- Location maps
- Alarm history
- Commands panel
- Event history
8.2 Asset Twin Dashboard
Displays aggregated analytics:
- Energy consumption
- Occupancy/usage
- Environmental conditions
- Device performance within asset
- Floor/building heatmaps
- Maintenance KPIs
8.3 Hierarchical Navigation
Users can navigate:
Building → Floor → Room → Sensor
or
Cluster → Device Group → Device Twin
9. Digital Twin AI Integration
Digital Twins support built-in AI:
9.1 Supervised Models
Use cases:
- Predict failure probability
- Estimate sensor readings
- Predict energy usage
- Compute performance scores
Output is stored in the twin as:
- Predicted_value
- Probability_failure
- AI_health_score
- AI_status_label
9.2 Unsupervised Models
Use cases:
- Anomaly detection
- Behavioral drift detection
- State deviation scoring
Outputs:
- anomaly_score
- health_score
- anomaly_flag
- anomaly_type
Twin automatically updates these values.
10. Digital Twin Alarms
Alarms may be triggered by:
- Telemetry thresholds
- Attribute changes
- AI anomalies
- Predicted failures
- State transitions
- Connectivity loss
Alarm data is stored in:
- Twin’s alarm table
- Global Alarm Center
- Reports engine
Twin supports:
- Acknowledge
- Clear
- Severity tracking
- Alarm lifecycle analytics
11. Twin-to-Twin Relationships
Digital Twins support relationships such as:
11.1 Parent → Child
- Room → Sensor
- Machine → Sub-Devices
- Plant → Pumping Station
11.2 Asset → Device Mapping
Assign devices to assets for high-level visualization.
11.3 Multi-Level Hierarchies
DeviceBoard supports unlimited hierarchical depth.
12. Using Digital Twins in Reports & Analytics
Analytics support:
- Device performance across twins
- Asset-level aggregation
- AI prediction trends
- Anomaly history
- Alarm distribution
- Efficiency analytics
- Environmental assessment
Reports include:
- Digital Twin performance report
- Asset utilization report
- Predictive maintenance report
- Health score timeline
- Twin hierarchy summary
13. Extending Digital Twins (Customization)
Advanced users can implement:
13.1 Virtual Telemetry
Derived values computed without device sending them.
Examples:
- Efficiency = energy_output / energy_input
- Power_factor = real_power / apparent_power
- Running_state = (rpm > 100) ? “ON” : “OFF”
13.2 Virtual Assets
Used to represent complex systems:
- Entire factories
- Pipelines
- Vehicle fleets
- Cooling tower clusters
13.3 Custom Widgets
To visualize digital twin behaviors such as:
- 3D building layouts
- Rotating machinery animations
- Dynamic schematics
- Custom health indicators
13.4 Custom Connectors
If your device uses a legacy or proprietary protocol, DeviceBoard’s Solution Team can create:
- Protocol adapters
- Payload decoders
- Modeling templates
- Virtual attribute maps
14. Security & Access Control for Digital Twins
Digital Twin access is governed by:
14.1 Device Groups / Asset Groups
User sees twin only if device/asset is assigned.
14.2 RBAC
Controls which modules the user can access:
- Dashboards
- Device view
- Asset view
- AI analytics
- Reports
14.3 ABAC
Controls what user can do:
- Acknowledge alarms
- Send commands
- Edit attributes
- Modify twin relationships
15. Troubleshooting Digital Twin Issues
- ✔ Device may not be connected
- ✔ Payload decoding misconfigured
- ✔ Wrong Device Model mapping
- ✔ Wrong unit conversion
- ✔ Incorrect RulesFlow transformation
- ✔ AI model not deployed
- ✔ Wrong telemetry mapping to AI node
- ✔ Parent-child relation not configured
16. Summary
DeviceBoard’s Digital Twin architecture enables:
- Real-time virtual representation of devices & assets
- Hierarchical modeling of infrastructure
- AI-enhanced insight generation
- Strong automation via RulesFlow
- Multi-protocol connectivity
- Rich visualization dashboards
- Predictive maintenance capabilities
- Actionable alarms & events
- Data-driven decision support
This guide provides everything needed to build, deploy, and operate Digital Twins effectively within DeviceBoard.