Towards partial power processing with built­ in impedance spectroscopy

Abstract

The world is moving away from using the traditional fossil fuels as energy sources and towards a system with an increasingly share of renewable energy. Many types of renewable energies come from forces of nature such as wind and sun, and are therefore as unreliable as the weather. The volatile nature of renewable energy creates a demand for energy storage to ensure supply of energy when the wind isn’t blowing and the sun isn’t shining. The energyproducedbytheserenewableenergysourcesisintheformofelectricity. Electricity can’t be stored directly, but has to be converted to other energy forms to be stored. The energy might be stored as chemical, potential or kinetic energy, but in order to store the energy a conversion from one potential to another is almost certainly needed. In this thesis the focus will be on electrochemical devices such as an electrolyzer cell or batteries as storage elements. Every bit of energy that is produced should be used, but the fact is that there are losses in transporting the energy, converting it, storing it and then converting it back to electricity before the energy can finally be used. The fact that there are two conversions, makes the efficiency of the converter extremely important to not loose any more energy then absolutely necessary. In this thesis the partial power configuration is presented, which has the ability to boost the efficiency of the energy conversion using a regular converter. Traditionally the source, dc-dc converter and load has been connected one after the other, forcing the converter to process the full power delivered to the load. The partial power configuration has the source, one side of the dc-dc converter and load in series, leading to the converter processing the full current, but only the difference in voltage between source and load. Since the difference in voltage potentials of source and load is processed, it is required that the potentials should be relatively close to get the maximum effect of the partial configuration. The remaining side of the dc-dc converter can either feed the converted power back to the source or forward it to the load. Therefore only part of the power enters the dc-dc converter and is affected by the converters efficiency. Since only part of the power is being processed, even a low efficiency converter can be used to achieve a highly efficient system energy transfer. A synthesis of all different partial power configurations shows that there are only two different configurations for partial power processing, when using a bipolar, bidirectional dc-dc converter. The thesis considers the design of dc-dc converters in order to test partial power processing. A graphica linterface is introduced,which visualizes the consequence of each individual design choice on component loss, voltage and current. The calculated losses of each component is then used to estimate the total loss and efficiency of the designed converter. Three different prototype dc-dc converters are presented, which are all part of the ”Smart Chargers” project. The first prototype is a ”proof-of-concept”, and is the only one to be run at full power. For the second prototype the layout of the power loop is considered in conjunction with the converter topology, i.e. the natural occurring inductance and capacitance of the converter topology is considered for optimizing the layout for a capacitance and induction reduction. The third prototype is designed with anon-board electrochemical impedance spectroscopy measurement setup, such that the impedance spectrum can be measured while the converter is charging or discharging a battery or electrolyzer stack. Lastly electrochemical impedance spectroscopy is introduced to partial power processing. Electrochemical impedance spectroscopy is ananalysis tool that can be utilized to measure the impedance spectrum of a device. To enable online measurement of a device the electrochemical impedance spectroscopy is built in to the dc-dc converter. This means that it will be possible to monitor the state of the chemical device, while it is being charged or discharged. The measurements are done both with a 2-point and a 4-point measurement and the benefits of using either is discussed. The thesis shows a potential way towards a more efficient, monitored electrical setup that includes more fluctuating energy sources and storage elements.