This website uses cookies primarily for visitor analytics. Certain pages will ask you to fill in contact details to receive additional information. On these pages you have the option of having the site log your details for future visits. Indicating you want the site to remember your details will place a cookie on your device. To view our full cookie policy, please click here. You can also view it at any time by going to our Contact Us page.

Weidmüller Automated Machine Learning tool

24 August 2020

The Weidmuller Automated Machine Learning (AML) platform uses an intuitive, guided HTML interface approach.

Artificial intelligence and machine learning tools are being used increasingly today in factory automation and process control applications. Together, they make it possible to use a wealth of often readily available data and provide predictive condition monitoring, better quality control and more efficient maintenance scheduling.

The Weidmuller Automated Machine Learning (AML) platform uses an intuitive, guided HTML interface approach, helping engineers with good plant or process knowledge to generate software models by marking up good and bad process or machinery data. This means that personnel without a data science or statistical background can quickly develop and test various models themselves.

The Weidmuller Automated Machine Learning Tool is not application or sector specific; it can be implemented in environments as diverse as highspeed manufacturing, marine, oil and gas, fine chemicals, or any other plant/process. 

The software provides two modules for the user, a Model Builder and a Runtime Environment. With the AML model builder, engineers take their first steps with model creation. Good plant or process knowledge is essential here and since most engineers have the experience to spot regular or abnormal behaviour, they can easily begin design activity.

This combination of input data (in timestamped csv format) and good engineering knowledge allows speedy generation of AML models and the results are fully comparable with solutions created using a conventional, lengthy data science-based approach. The system produces several models using different analysis techniques for consideration by the user. Each model is scored for accuracy and other criteria and the user then selects the most suitable model for the application. The selected model can then be transferred to our Runtime Environment (RTE) which analyses real time data, provides visualisation, model tuning and export of meaningful output to third party SCADA or DCS systems. Solutions can be hosted locally on physical hardware or in the cloud.

In summary, using the Weidmüller Automated Machine Learning Tool shortens the time the from first concept to completion, saving valuable time and resources. At the same time, the user also benefits from the latest developments in the predictive condition monitoring environment. There is no need for extensive training. There is no need to purchase external expertise. In just a few hours, a model can be set up which then automatically detects anomalies. It couldn't be simpler.

For more information, contact Keith Atkinson, Jack Taylor or Nick Rhodes, visit the micro-site or call on 0845 094 2007.

Contact Details and Archive...

Print this page | E-mail this page