Are you an EPFL student looking for a semester project?
Work with us on data science and visualisation projects, and deploy your project as an app on top of GraphSearch.
Heart failure is a cardiovascular disease affecting between 1 and 2% of the population of developed countries alone - more than 10 million people - with this number bound to increase due to the ageing of the population. This condition starts to develop itself when the heart is not capable to properly pump the blood in the circulatory system anymore, and it can be due to other cardiovascular diseases, hypertension, heart attacks, etc. When cases become very severe, surgical procedures are then necessary to install cardiac assist devices to help reduce the stress experienced by the heart and allow it to potentially heal itself. However, the devices currently used are bulky and very invasive with risks of infections and rejections.As such, the Center for Artificial Muscles (CAM) was tasked to investigate and develop a new kind of cardiac assist devices which would be much less invasive to install. To do so, the idea envisioned consisted of working with Dielectric Elastomer Actuators (DEAs) which are often nicknamed artificial muscles due to their close resemblance with the natural ones as they, amongst other things, light, thin and can deform themselves significantly. One of the main challenges of working with this technology comes from the fact that DEAs have very uncommon needs concerning the driving electronics. Indeed, to operate to the maximum of their abilities, DEAs require voltages of several thousands of volts - between 5 and 20 kV depending on the actuator - to be applied to them. Currently, these actuators are powered using extremely large and impractical laboratory power supplies which is unacceptable for portable and medical applications. It was therefore critical to investigate and uncover a solution for the electronics so that the DEAs' needs could be met with a system as compact and efficient as possible.Thus, this thesis presents the investigating process followed to obtain in the end an electronics system capable of supplying at least 8 kV to a DEA and recover unused electrical energy stored in the actuator. Process which consisted at first of an investigation of power electronics topologies capable of amplifying a given input voltage to a higher level in order to determine the most suitable one. It was concluded that the DC/DC flyback converter was the best candidate for the job.This was followed by an in depth analysis of the working principle of this converter topology to determine the limiting factors regarding the supply of high voltages. This analysis was confirmed through the development of an analytical model depicting the behavior of the converter when it charges the actuator to the required voltage. Thanks to this a list of solutions to overcome these factors could be given.Then, due to the voltage levels worked with and the necessity to recover the unused electrical energy stored in the actuator, it was revealed that commercially available solutions were scarce to nonexistent to properly control the electronics. As such, control strategies and custom devices were ultimately conceived to resolve this problem.In the end, the final ultra-high voltage bidirectional flyback converter prototype was used to operate a DEA pump placed in a test bench recreating the pressure conditions and blood flow which are expected to be found in a living body. These tests showed that the electronics performed very well and managed to operate without major issues the new cardiac assist device created by the CAM.