-
Notifications
You must be signed in to change notification settings - Fork 8
Prognosers
Prognosers, the core of the GSAP system, contain the fundamental logic for performing prognostics. Prognosers take in live sensor data, configuration and future loading parameters, and output state estimation and prognostic results.
A new prognoser is created to support a new method for performing prognostics. However, many prognostics systems that do follow a common model-based structure, do not require the creation of a new prognoser, only the creation of a new model that will be used by the ModelBasedPrognoser class.
GSAP is shipped with one standard prognoser that is listed below. Click on the prognoser name for more information.
- Model Based Prognoser This prognoser is used to conduct model-based prognostics.
Individual prognosers can have their own configuration parameters, which are documented with each specific prognoser.
General configuration parameters that would be accepted by any prognoser:
Parameter | Description |
---|---|
type [Required] | The type of prognoser (ex: modelBasedPrognoser). |
name [Required] | The name of the component being prognosed (ex: battery1). |
id [Required] | A unique identifier for the piece of hardware being prognosed (ex: Serial Number). |
inTags [Required] | A list of tags expected from communicators in format local_name : global_name (e.g., inTags: voltage:volts, power:pwr, temperature:temp). |
histPath [Optional] | A path for the history files. |
resetHist [Optional] | A flag to reset the recorded history for the component. Will archive the current history file and start a new one. |
Visit the specific prognoser, model, or algorithm description pages for information on additional configuration parameters.
New Prognosers will be added when the Prognosers do not follow the ModelBasedPrognoser class pattern.
Note: If you are using the ModelBasedPrognoser class pattern, you do not need to create a new Prognoser, but you will need to create a model that will be used by the ModelBasedPrognoser class.
If you are not using ModelBasedPrognoser class, follow the following instructions to create a new Prognoser:
- Copy EmptyPrognoser.cpp and EmptyPrognoser.h from the templates directory and rename to the name of your prognoser
- Follow the instructions inside EmptyPrognoser.cpp to create your prognoser