Efficient energy extraction of salinity gradient power through a reverse-electrodialysis system

Abstract

Energy extraction of salinity gradient power (SGP) through a reverse-electrodialysis (RED) stack, which features direct electricity conversion, globally abundant storage, and zero production of pollutants, has great potential in providing our world with a truly sustainable and uninterruptible clean energy production. Nevertheless, the RED technology still witnesses a relatively low power generation efficiency. To advance the development of harnessing SGP, this thesis investigates into the power utilization stage of the RED stack. By dividing the efficient power utilization stage into three segments (i.e., RED modelling approaches, maximum-power-pointtracking (MPPT) techniques, and single-input multiple-output interfaces) and investigating each individually, a complete RED system for harnessing SGP efficiently can be systematically designed and developed. Initially, the existing research works on each of the segment are assessed, of which research gaps are summarized. Subsequently, new findings are presented to fill the gaps. Simulation and experimental results are provided and compared to validate the effectiveness and feasibility of the proposed solutions. The major contributions of this thesis are summarized as follows. Firstly, the framework of an RED system at the power utilization stage, which includes RED modelling approaches, control methods, and power interfaces, is established. The desired characteristics of each segment in the RED system are described in detail, based on the RED stack’s unique features. This provides new perspectives of advancing the RED technology and facilitates future research works associated with it. Secondly, a generalized hybrid model based on the electrical circuit is proposed for RED stacks. It covers two operation modes and is capable of accurately characterizing RED stacks’ electrical dynamics. An improved kinetic battery module is introduced, alleviating complicated calculation processes and simplifying the computation. The proposed model lays a foundation for the design of control methods and power interfaces in the RED system.Thirdly, a self-adaptive-step-size incremental-resistance MPPT technique is for the first time proposed to efficiently extract maximum power from the RED stacks. The principles and a systematic approach for designing and optimizing the scaling factor and the sampling interval in the MPPT controller are provided.Fourthly, catering to RED stacks’ electrical characteristics and application scenarios, a non-isolated single-input triple-output converter with high efficiency is employed for the first time to interface RED stacks with multiple output ports. An easy-to-derive decoupled control method is put forward to avoid the cross-coupling of multiple ports. The design guidelines of inductors and the battery charging current to ensure ZVS operations of all switches are analytically provided.This thesis comprises six chapters. The first chapter delivers a background review of the SGP and RED technologies, while the second chapter reports existing research works regarding the RED stack modelling approaches, MPPT techniques, and single-input multiple-output converters. The third chapter describes the generalized hybrid RED model. The fourth chapter presents the self-adaptive-step-size incremental-resistance MPPT control scheme applicable to RED stacks. The fifth chapter reports a single-input triple-output converter with decoupled control scheme designed for RED stacks. Finally, the last chapter concludes the thesis and suggests possible directions of future works.