When we are in the car, plane or train we are normally too busy to wonder how they work. We cannot imagine that many aspects of our everyday life, like different means of transport or factory plants, health and energy facilities are actually made possible by the integration of electronic equipment called embedded systems.

Published Friday, 17 October, 2014
Updated Thursday, 2 July, 2015

*This Blog Post was written together with Ioannis Bitsios, DG CNECT, and Paul Goulart, Scientific Coordinator of the EMBOCON Project.

These systems, which are invisible to us, are becoming more and more sophisticated and one of the challenges is to improve their performance while minimizing their operation costs (optimisation): a slight improvement in precision and computing speed can make a big difference in some industrial production processes such as the ones in the energy (gas and electricity), chemical, industrial automation robotics, automotive sectors, where time reduction and optimisation of resources can strongly affect the production costs.

Let's take as an example the energy management system in hybrid electric vehicles whose ultimate aim is to consume less fuel and produce less CO2, whilst keeping the performance and safety characteristics of the car up to pre-set standards. In order to decide whether to switch from the electric to the fuel engine and vice versa, this embedded system has to take into considerations many factors (some of them varying while driving) and react in a very short lapse of time, i.e. in real-time, transmitting the best (optimal) decision to the vehicle.

The EMBOCON project is a successful example of how through a complex combination of mathematical algorithms and predictive techniques, these improvements can provide European embedded systems with a competitive advantage in the near-future market.

Find below some esteems and potential impact statements, derived from the projections following the test and demonstration of the EMBOCON technology:

  • several million €/annum is the value of the optimised control methods for the European chemical process industry (source BASF ).
  • 50% reduction on the downtime caused by external factors (such as voltage instabilities coming from the electrical grid/network) to the electrical (A/C) motors that are used in many real-life applications (like in the cement, mining, minerals, water and power industries) . As a result the motor speed can be better adapted to the actual operation needs, thus optimizing energy consumption and reducing CO2 emissions (source ABB, market realization is expected to be within 5 years),
  • more efficient operation (closer to the operation limits) for natural gas compressors leading to a realistic reduction of energy consumption between 2-4% (source ABB, a prototype controller was realized using EMBOCON tools, actual implementations in the market are expected within 5 years),
  • efficient control for gas or liquid transport or refrigeration systems. For a gas turbine or engine driven systems, the potential benefit is to cut fuel consumption by about 5-7% per year corresponding to savings in fuel costs (2-4 million € / year) and significant reductions in CO2 emissions (10-20 thousand tons of CO2 / year). Actual implementation in the market is expected within 2 years (source ABB),
  • 27 full-time technical positions in industry is projected to be created in the next five years due to the exploitation/commercialization of the project's outcomes.

The intelligence of the EMBOCOM system is the result of a harmonic and fruitful academic and industrial collaboration which led to the creation of 2 spin off companies, a patent and first tests in the medical technology market.

  • EMBOTECH, spin-off from ETH Zurich. H/W & S/W solutions for optimization-based decision making on embedded platforms.
  • MatheMedical, (so new that no website is available yet!) spin-off from University of Heidelberg (UHEI), which has patented an innovative approach for determining the cause for irregular heartbeats. Mathemedical uses the EMBOCON optimization tools as a basis for its model-driven decision support systems.

The successful results of the projects open up new markets in automotive and aerospace business to some of the project partners. The EMBOCON project proved to be a major enabler for partner LMS in its strategic transition as design engineering partner from the mechanical to the mechatronic domain. LMS turned EMBOCON ‘Model Predictive Control’ into a tool for accelerating product development in the mechatronic sector. Market leading automotive companies have already invested in over 1 M€ on engineering services projects.

Moreover there is a significant scientific impact and produced knowledge: more than 160 published peer-reviewed scientific papers, 20 PhD theses and an edited special journal issue, across a wide range of theoretical and applied journals and international conferences.

If you want to know more about EMBOCON and its partners, please have a look at the project website. Additionally, you can access its YouTube video channel and watch some of the actual demonstrations that took place during the project's lifespan.

The EMBOCON consortium consisted of 7 academic beneficiaries (Imperial College London as coordinator, the universities of Bucharest, Dortmund, Heidelberg, Leuven, Magdeburg and Zurich) and 3 industrial contractors (BASF SE Germany, Leuven Measurement Systems Belgium (LMS) and IPCOS NV Belgium (SME)).

For further information, please contact:

Paul Goulart, Scientific Coordinator: p.goulart@imperial.ac.uk

Follow us on Twitter: @Complex_Systems