Wearable ultrasound systems hold significant potential for continuous health monitoring but face challenges in minimizing energy consumption and device size. This research investigates the design and simulation of an energy-recycling inductive power converter for CMUT-based wearable ultrasound applications. A comparison between conventional class-D and resonant pulsers highlights the trade-off between operational flexibility and energy efficiency, emphasizing the critical impact of high-voltage bias generation, which can consume hundreds of times more energy than a single transmit pulse.

A novel four-switch converter topology is proposed, enabling both high-voltage biasing and pulsing with energy recycling. Unlike traditional buck–boost converters, the proposed converter operates in a unique regime: it only functions during the initial charging of capacitive loads, after which the output can be fully decoupled, allowing the inductor to perform other tasks. This places the design in a fundamentally different and largely unexplored operating point within power-converter research. It also facilitates energy recycling, allowing the recovered energy to be reused after pulse-acquisitions in other systems where power is required.

System-level simulations explore the influence of inductor sizing, switch dimensioning, and current optimization on efficiency, while break-even analysis indicates that conventional pulsers may remain more practical for typical pulse counts due to lower complexity. Circuit-level implementation in Cadence confirms functional operation, high-voltage generation, pulsing and energy recycling.

This work identifies the primary sources of energy inefficiency and proposes control strategies, including optimized switch timing and parasitic energy recovery, to enhance performance. The study provides a validated foundation for future experimental research and practical implementation, demonstrating the feasibility of energy-efficient wearable ultrasound systems using 180nm technology.

Sustainable energy technologies are ever-increasing in popularity, not only among companies but also among consumers. Photovoltaic, or PV, cells are amongst the most well known in this field. This project was set up for designing a PV-powered UAV that is able to generate an albedo map of an area equal or bigger than the campus of the TU Delft in a single run. PV cells, however, are highly susceptible to changes in weather conditions and have a range of voltages at which these cells can supply power. Besides that, unnecessary losses occur if the so-called ’Maximum Power Point’ (MPP) is not tracked. In this thesis, a boost converter is proposed for the aforementioned system to implement the maximum power point tracking and supply the, by the PV system produced, power as efficiently as possible to the UAV power system.