Landolfo, Matteo (2024) Adaptive nutrient management in indoor cultivation: a modular intelligent decision support system approach integrating artificial intelligence and fuzzy logic. [Tesi di dottorato]
![[thumbnail of PhDThesis_LandolfoMatteo_37ciclo.pdf]](https://www.fedoa.unina.it/style/images/fileicons/other.png "PhDThesis_LandolfoMatteo_37ciclo.pdf") |
Documento PDF
PhDThesis_LandolfoMatteo_37ciclo.pdf Visibile a [TBR] Amministratori dell'archivio Download (4MB) | Richiedi una copia |
| Tipologia del documento: | Tesi di dottorato |
|---|---|
| Lingua: | English |
| Titolo: | Adaptive nutrient management in indoor cultivation: a modular intelligent decision support system approach integrating artificial intelligence and fuzzy logic |
| Autori: | Autore Email Landolfo, Matteo matteo.landolfo@unina.it |
| Data: | 12 Dicembre 2024 |
| Numero di pagine: | 88 |
| Istituzione: | Università degli Studi di Napoli Federico II |
| Dipartimento: | Biologia |
| Dottorato: | Intelligenza artificiale Area Agrifood e ambiente |
| Ciclo di dottorato: | 37 |
| Coordinatore del Corso di dottorato: | nome email Loreto, Francesco francesco.loreto@unina.it |
| Tutor: | nome email Pennisi, Giuseppina [non definito] Gabrielli, Maurizio [non definito] |
| Data: | 12 Dicembre 2024 |
| Numero di pagine: | 88 |
| Parole chiave: | Intelligent decision support system; Plants nutrient management; Fuzzy inference system; Vertical Farms |
| Settori scientifico-disciplinari del MIUR: | Area 07 - Scienze agrarie e veterinarie > AGR/04 - Orticoltura e floricoltura |
| Informazioni aggiuntive: | Appartengo al 37° ciclo. Per problema tecnico ho inserito il ciclo 36 |
| Depositato il: | 27 Ott 2025 15:10 |
| Ultima modifica: | 12 Ago 2026 05:38 |
| URI: | https://www.fedoa.unina.it/id/eprint/16529 |
Abstract
Population growth and climate change impose new challenges on agriculture, necessitating innovative solutions to enhance productivity and resource-use efficiency. Vertical farms, based on controlled environment cultivation, represent a promising model, although nutrient management remains a significant limitation. This PhD project developed a modular intelligent decision support system to optimize autonomous fertilization in indoor cultivation, integrating artificial intelligence algorithms and fuzzy logic. The first experiment involved the use of convolutional neural networks (CNNs) to classify plant species, ensuring accurate identification (88% accuracy) essential for species-specific nutrient management. The second experiment employed a regression model based on Multi-Layer Perceptron to estimate the SPAD value, a reliable indicator of plant nutritional status. The model achieved a correlation coefficient of 0.92, demonstrating high accuracy and reliability in real-time, non-destructive monitoring, while reducing dependency on traditional invasive methods. Lastly, the system leveraged a fuzzy inference model to autonomously adjust nutrient dosing based on phenotypic requirements, validated through simulations on basil plants. Results showed the fuzzy system significantly improved nutrient use efficiency, achieving a 25.9% reduction in fertilizer application compared to static models and a 20.4% reduction compared to linear models. These results underscore the effectiveness of the intelligent decision support system in optimizing nutritional efficiency, reducing waste, and achieving stable nutrient management. This study provides a foundation for the development of a comprehensive modular system for the autonomous management of a vertical farm, with potential applications for integrated irrigation, lighting, and microclimate management, thus promoting more sustainable and resilient indoor agriculture that meets diverse crop requirements.
Downloads
Downloads per month over past year
Actions (login required)
 |
Modifica documento |
