Hydroponic system customization and sensor integration for operation optimization and physiology study

Abstract

Food security and climate change pose great threats to the future of the agriculture industry. By 2050, with 20% arable land reduction and an additional 2.2 billion people, agriculture will need to adapt. Before this research, precision agriculture had shown significant advantages in improving resource usage efficiency and yields with less land. However, most of the popular precision agriculture approaches are associated with the tag of high technical barriers to implementation and scalability. This dissertation aims to answer the central question: What approaches should be applied for precision agriculture, through both hardware and software applications, to understand the scientific connection between plant physiological response and changing environmental conditions, ultimately enhancing agricultural production? Several published, submitted, and drafted studies prepared the foundation for this interdisciplinary research discussed in Chapters 2 - 5. Chapter 2 delves into a manufacturing pathway dedicated to customizing cost-effective and performance-effective sensors for precision agriculture. Through extensive experimentation involving temperature, speed, and applied voltage, we employ electric field-assisted direct writing technology to fabricate 3D metallic structures. This innovative approach empowers us to adapt sensor functional and dimensional variations to suit the specific requirements for collecting environmental and physiological data, as elaborated in Chapter 5. The ability to rapidly prototype these sensors enhances the efficiency of plant science research, contributing to the broader goals of precision agriculture. In Chapter 3, we focus on developing a germination substrate with precise control over mechanical gripping forces, water regulation, plant respiration, and real-time root growth visualization. We achieve distinct mechanical characteristics by leveraging the anisotropic properties of 3D-printed Carboxylated methylcellulose (C-MC), ensuring liquid-free water access, oxygen supply, and carbon dioxide removal. This transparent substrate enables 100%germination, non-destructive evaluation of germination and facilitates root phenotyping, offering flexibility in experimental conditions for plant science research. Furthermore, it validates the viability of seed germination in hydrogels, which explores environmental conditions for lettuce growth and leverages the anisotropic mechanical performance afforded by additive manufacturing. Building upon the foundation laid in Chapter 3, which showcased the potential of MC hydrogel, we conducted further experiments involving formulation chemistry modifications. Chapter 4 explores the most influential environmental parameters affecting plants during germination. Through three rounds of trials and two iterations of Hoagland formulation optimizations tested with more than 900 seeds, we identify pH, hydrogel content, and photo energy as the most influential factors impacting germination. This chapter validates the earlier findings regarding the impact of mechanical differentiation on germination, as established in Chapter 3. The insights gained in this study provide valuable recommendations for selecting environmental parameters in Chapter 5, enhancing the effectiveness of plant physiological research. In Chapter 5, we turn our attention to the operational aspects of our research, emphasizing the collection of efficient data, signal processing, and the translation of data into valuable evaluation indices. We integrate the sensing matrix into a customized hydroponic growing system capable of monitoring controlled environmental conditions while non-destructively evaluating lettuce’s physiological responses. By processing and analyzing high-frequency weight measurement data, we create 14 plant phenotyping and system operation efficiency evaluation indices. These indices serve two vital purposes: firstly, they scientifically elucidate the connection between environmental condition shifts and the corresponding physiological responses of the plants; secondly, they enable us to evaluate system performance and provide actionable operational insights. This study employs a system engineering framework to design a hydroponic system that serves as a platform for plant physiology research. Its functions include the validation of results from Chapter 4, the establishment of scientific connections between quantifiable data and plant physiology, the proposal of a non-destructive and in-situ data acquisition system that significantly reduces sample requirements and resource usage, and, perhaps most importantly, the advancement of crop modeling and precision agriculture while minimizing technological barriers and costs.