• Energy Research
• 2025-2025close
• Open Access close
• Closed Access close
• Restricted close
• Open Source close
• 11. Sustainability close
• 6. Clean water close

Machine learning's application in solar-thermal desalination is limited by data shortage and inconsistent analysis. This study develops an optimized dataset collection and analysis process for the representative solar still. By ultra-hydrophilic treatment on the condensation cover, the dataset collection process reduces the collection time by 83.3%. Over 1,000 datasets are collected, which is nearly one order of magnitude larger than up-to-date works. Then, a new interdisciplinary process flow is proposed. Some meaningful results are obtained that were not addressed by previous studies. It is found that Radom Forest might be a better choice for datasets larger than 1,000 due to both high accuracy and fast speed. Besides, the dataset range affects the quantified importance (weighted value) of factors significantly, with up to a 115% increment. Moreover, the results show that machine learning has a high accuracy on the extrapolation prediction of productivity, where the minimum mean relative prediction error is just around 4%. The results of this work not only show the necessity of the dataset characteristics' effect but also provide a standard process for studying solar-thermal desalination by machine learning, which would pave the way for interdisciplinary study.

Sustainable Development (SD) is a significant, high-visibility endeavor. However, there is no comprehensive synthesis of the concept. Instead, there is diversity and vagueness to the point where the term may have lost utility. The following dissertation investigates both theories and practices of SD, aiming to identify, validate, and apply a simple, informative, and operationalizable core idea behind SD. The research objectives are three-fold: (1) to propose a definition of SD that balances complexity and simplicity while accounting for variation in SD approaches; (2) to create an evaluation framework that effectively and minimally differentiates between adequate and inadequate approaches; and (3) to offer guidelines for SD implementation based on the framework’s application to selected case studies. The research consists of three studies, each focused on one of the objectives and presented in chapters two, three, and four.The first study seeks to identify a core concept for SD that is both simple and informative. The study argues that four concepts are fundamental to adequately understand SD: development, sustainability, justice, and governance. First, as a directional change in the net quality of life, development is needed to ensure desirable living standards for the overall population and subpopulations. Second, as the ability to maintain a system feature or state over a time horizon, sustainability is required to guarantee that the quality of life remains uncompromised in the face of social and environmental constraints and trade-offs. Third, justice is critical because while development can be unjust and injustices can be sustained, these outcomes are inconsistent with SD objectives. Fourth, governance is fundamental to regulate and enforce the desired traits of the system characterized by the previous three concepts, particularly to oppose existing tendencies toward inequity and injustice. The second study focuses on devising a diagnostic tool, i.e., an evaluation framework, to address the need to assess and compare the abundance of theoretical and practical approaches. Based on an intense interdisciplinary literature review, the developed framework consists of ten questions addressing development, sustainability, justice, and governance. The study argued that an adequate SD approach should cover and address all these questions promptly. The first two questions capture development by determining which metrics to use and how to measure developmental success. The next three questions cover sustainability by addressing what to sustain, how to determine success, and the intended time horizon. The following two questions focus on justice, addressing the target recipients and the distribution of benefits and burdens. Lastly, the final three questions encompass governance, including whether it is a shared practice, whether it results in collective actions toward SD, and whether SD objectives are integrated. Based on the analysis, SD is defined as Sustainable Development as the evolution of a particular Coupled Human and Natural System (CHANS) resulting from an intervention to improve or maintain the net quality of life for the entire system within the environmental and social constraints of the system, while ensuring that the increase in the relative quality of life for the least advantaged members of the system is at least greater than for the remainder, over multi-generation time horizons for system participants, and opposing existing tendencies toward inequity and injustice via appropriate governance. The third study applies the evaluation framework to two case studies, the Burning Man Project (BMP) and the UN’s Sustainable Development Goals (SDGs), to demonstrate the diagnostic tools’ utility, validate its robustness, and extract cross-scales insights. The investigation reveals that both BMP and SDGs meet the minimum criteria for adequate SD approaches despite implicit or partial responses to the ten questions. Recommendations for improving these approaches include explicating the time horizon of sustainability and a more thorough integration of objectives. Furthermore, based on the comparison of the two cases, the study discusses the role of the population, the importance of integrating objectives, and scale considerations and provides practical insights. For instance, both BMP and SDGs need to uphold a minimum commitment among their target population to their respective practices and ensure meeting sustainability constraints. However, the reasons, potential solutions, and management mechanisms differ between the two cases due to context-dependent factors and variations in scale. The aspiration is that this research offers a foundation for future investigation and application of SD, providing a comprehensive and operationalizable framework to evaluate and compare diverse SD approaches and facilitating more effective development and implementation of SD strategies. The insights gained from this study can help inform the development of new SD approaches, monitoring and assessment tools, and foster a more nuanced understanding of the intricacies inherent in the pursuit of SD.

In the last decade, the manufacturing capacity of silicon, the dominant PV technology, has increasingly been concentrated in China. This has led to PV cost reduction of approximately 80%, while, at the same time, posing risks to PV supply chain security. Recent advancements of novel perovskite tandem PV technologies as an alternative to traditional silicon-based PV provide opportunities for diversification of the PV manufacturing capacity and for increasing the GHG emission benefit of solar PV. Against this background, we estimate the current and future cost-competitiveness and GHG emissions of a set of already commercialized as well as emerging PV technologies for different production locations (China, USA, EU), both at residential and utility-scale. We find EU and USA-manufactured thin-film tandems to have 2 to 4% and 0.5 to 2% higher costs per kWh and 37 to 40%and 32 to 35% less GHG emissions per kWh at residential and utility-scale, respectively. Our projections indicate that they will also retain competitive costs (up to 2% higher)and a 20% GHG emissions advantage per kWh in 2050.