Measurement Stream
GreenBus Measurements represent a particular value of data for a particular time in the stream of values associated with a Point. The Measurement data structure provides a common representation of time-series data across varied sources and communications protocols.
Each Measurement contains the following attributes:
| Attribute | Description |
|---|---|
| Value | The data value. Can be a boolean, integer, floating point, or string. |
| Time | The time associated with the value. May also include device time |
| Quality | The quality of the Measurement, based on the IEC 61850 standard for telemetry quality. |
Measurements enter the system through front ends, which represent an external connection to a data source. Those data sources may include:
- An intelligent electronic device (IED) communicating with an industry protocol such as Modbus or DNP3.
- A device publishing to a messaging solution such as MQTT.
- A gateway to an external web service, such as weather forecasting.
- Values published by a GreenBus application acting as a synthetic data source. ` Data sources are partitioned in the GreenBus model into Endpoints, which correspond to an individual, logical process providing data (and receiving controls). For every Endpoint receiving data, there is a logical process in a protocol front end that has registered itself with a logical process in the measurement processor, a GreenBus core component. A single instance of the measurement processor software running on a computing node may support many Endpoints, and similarly front end connections for a particular protocol may be served by the same running application, or distributed across many.
Furthermore, "protocols" that are actually external web services or message-oriented technology like MQTT may serve many logical devices with a single Endpoint. In these cases Endpoints may signify a particular web URI or connection to a broker.
Processing
Front end protocols post batches of Measurements to the measurement processor. The values have at that point been normalized regardless of source device or communication protocol, and the following processing steps are then applied according to the system modeling of the individual Point:
| Attribute | Description |
|---|---|
| White list | Values may only enter the system that have Points configured for them. |
| Overrides | Individual Point may have new values suppressed by setting their state to not in service (NIS) and optionally providing a synthetic Measurement to replace the current state of the stream. |
| Filter | To reduce the flow of data in the system, duplicate values and updates that occur within a deadband can be filtered before entering the database and event bus. |
| Transformation | Examples may include linear scaling and offset, discrete mappings to enumerations, or re-intepreting the Measurement quality. |
| Event | Conditions based on Measurement value or quality can be configured to create Events that notify the system of an occurrence or exception state. |
Data Access
The measurement processing subsystem stores Measurement values in two data stores: the current value database, and the historical database. The actual storage technology used may depend on the performance requirements of the system deployment. The GreenBus services API provides methods to query the current Measurement values for a set of Points, or to query historical values with parameters for time boundaries.
Notifications are also posted to both the individual Measurement subscription channel as well as the Measurement batch subscription channel. These provide immediate, event-driven delivery of data updates to processes that had previously subscribed.