node-red-contrib-machine-condition-forecast v1.0.0

node-red-contrib-forecast

This Node-RED custom node allows you to perform time-series forecasting and anomaly detection using exponential smoothing. This can be particularly useful for predictive/preventive maintenance scenarios.

Installation

npm install node-red-contrib-forecast


## Usage

1. Once installed, you will find the forecast node in the function category in the Node-RED interface.
2. Drag and drop the node into your flow.
3. Double-click on the node to configure its properties:

| Property | Description |
|---|---|
| Alpha | The exponential smoothing factor (default value is 0.5). |
| Number of Standard Deviations | This value sets the threshold for anomaly detection. If a data point's deviation from the forecasted value is greater than this many standard deviations, it is considered an anomaly (default value is 2). |

4. Connect the node's input to a source of numeric data and its output to a node that will handle the result.
5. The output of this node will be a JSON object that contains two fields:

| Field | Description |
|---|---|
| value | The current data point value. |
| status | A string that is either "Normal" or "Anomaly". |

## Predictive/Preventive Maintenance Use Case

In predictive/preventive maintenance scenarios, the goal is to predict failures or issues before they occur so they can be prevented. This often involves monitoring some sort of metric or sensor data over time, such as temperature or vibration, and looking for anomalies that may indicate a problem.

The forecast node can be used in such scenarios to analyze the data stream and flag any anomalies. For example, you could have a temperature sensor on a machine that sends data to the forecast node every minute. The forecast node applies exponential smoothing to forecast future values, and if the actual value is significantly different from the forecasted value (i.e., it is an anomaly), it could indicate a problem with the machine.

This could trigger a warning or an automatic shutdown of the machine, preventing damage to the machine and avoiding downtime. The exact reaction to an anomaly would depend on the specific situation and would be handled by other nodes in the Node-RED flow.

## License

This project is licensed under the GPL-3.0 License - see the LICENSE file for details.