PREDICTIVE MAINTENANCE

QUALIGON develops predictive maintenance solutions by combining its rich expertise in software, hardware, signal processing, communications and machine learning. 

Based on the requirements we implement

  • cloud-based systems where the machine learning part is hosted in the cloud to gain scalable performance
  • edge computing solutions completely on-premise with data in direct and secured access and low latency by architecture, especially for safety-critical applications

 

Predictive Maintenance Use-Cases

Engine Maintenance Prediction

Engines like the one used in ships or in the power generation represent huge investments. Failures of these system can endanger human lives and create high risk from the economic point of view. 

We are supplying solutions to analysis and predict the health-status of engines during their lifecycle to organise the engine maintenance. This enables the user increase the safety and to create the best-possible return-on-investment. 

Microcontroller and FPGAs Programming

Microcontroller and FPGAs (Field Programmable Gate Array) are perfect entities to implement customer and product specific solution implementations. With our expertise in programing, but also in the selection and connection of sensors we develop complete systems. 

Time Sensitive Network Implementations

 The usage of Time Sensitive Networks or TSN offers a variety of advantages in the implementation of solutions that require a reliable and synchronised communication. Often Time Sensitive Networks represent the missing piece of the puzzle to achieve the solution. We are dealing with wired and wireless TSN implementations following IEEE standards. 

 

 R&D Activities in Edge Computing

QUALIGON is part of the HORIZON2020 R&D project FRACTAL founded by European Commission and the Federal Ministry of Education and Research - BMBF of Germany. The objective is to create a reliable computing node that will create a Cognitive Edge under industry standards. Implementations are based on RISC-V processors and the PULP- and VERSAL platforms.