Name

TIA ASP - Teknopar Machine Learning on StreamPipes as a Data Processor

Defined in Task

T5.5

Short description

TIA ASP is an Apache StreamPipes based tool, it enables none technical users to select and execute machine learning algorithms for predictive maintenance purposes. The performances and the results obtained by the executed Machine Learning algorithms are presented in forms of graphs.

The tool has an easy-to-use, drag and drop user interface. Different data sources can be used as inputs to the ML algorithms of TIA ASP, including data from Kafka, and .csv files. The ML algorithms are MLP, GBT, LSTM, RF, SVM, and KNN.

Example of usage

Figure- A Sample Data Processor: TIA ASP using Kafka Data Source

Data coming from Kafka or .csv files can be used by the user selected machine learning algorithms, and the results of the predictive maintenance applied by the selected algorithms are displayed on the presented GUI above.

Figure- A Sample Data Processor: TIA ASP using .csv Data Source

Interfaces

User selected set of ML algorithms are executed on the stream and the results of the algorithms can be compared and graphically presented. Stream data contains sensor data in vector form.

Subordinates and platform dependencies

Apache StreamPipes, Apache Kafka, Fiware

License

Proprietary/ Subject to License

TRL

The current TRL is 4 running to be TRL 6.

References

n/a

To be considered in particular for the following COGNITWIN pilots

NOKSEL