Tecniche di risparmio idrico a livello locale e mondiale. Approfondimento sulle buone pratiche agricole per un uso sostenibile dell'acqua

Oltre un miliardo di persone non ha oggi accesso all’acqua potabile; più di due miliardi è il numero di coloro che vivono in condizioni igienico-sanitarie realmente proibitive. Sono 80 i paesi nel mondo (con il 40% della popolazione totale) in cui si riscontra difficoltà di approvvigionamento e presenza di risorse idriche che mancano dei requisiti che dovrebbero essere assicurati per la tutela della salute: quotidianamente e sistematicamente il diritto di accesso all’acqua, che nessun individuo dovrebbe vedersi negato, viene violato. Scarsità di acqua e non omogenea distribuzione sulla superficie terrestre sono fattori che concorrono alla crisi della risorsa, cui contribuiscono processsi di natura ambientale (cambiamenti climatici, desertificazione), di natura economica (le sorti dell’industria agroalimentare, la globalizzazione degli scambi, il bisogno crescente di energia), di natura sociale (migrazioni, urbanizzazione, crescita demografica, epidemie), di natura culturale (passaggio dal rurale all’urbano, dall’agricoltura di sussistenza a quella di profitto). Nell’ottica di uno sviluppo sostenibile un aumento indiscriminato dell’offerta non può costituire soluzione al continuo incremento della domanda di acqua. Si rende pertanto necessaria la definizione di politiche e strumenti di cambiamento nei modelli di consumo e nella pianificazione che consentano una riduzione degli squilibri nella distribuzione e nella gestione della risorsa a livello domestico e civile, industriale, agricolo. L’uso efficiente, e quindi sostenibile, dell’acqua è da perseguirsi secondo le modalità: • Risparmio, inteso come minore consumo di acqua all’inizio del ciclo. • Riciclo dell’acqua in circuito chiuso, inteso come riuso dell’acqua di scarico, o uso multiplo dell’acqua. Una idonea utilizzazione dipende da una idonea progettazione, che abbia come finalità: • La destinazione in via prioritaria delle fonti e delle risorse di più elevata qualità agli usi idropotabili, con una graduale sostituzione del consumo per altri usi con risorse di minore pregio. • La regolamentazione dell’uso delle acque sotterranee, mediante la limitazione del ricorso all’impiego di pozzi solo in mancanza di forniture alternative per uso civile, industriale, agricolo. • L’incentivazione ad un uso razionale della risorsa, anche mediante l’attuazione di idonee politiche tariffarie. • L’aumento dell’efficienza delle reti di adduzione e distribuzione, sia civili che irrigue. • La promozione di uso efficiente, riciclo e recupero di acqua nell’industria. • Il miglioramento dell’efficienza ed efficacia delle tecniche di irrigazione. • La promozione del riutilizzo delle acque nei vari settori. • La diffusione nella pratica domestica di apparati e tecnologie progettati per la riduzione degli sprechi e dei consumi di acqua. In ambito agricolo la necessità di un uso parsimonioso della risorsa impone il miglioramento dell’efficienza irrigua, pari appena al 40%. La regione Emilia Romagna a livello locale, Israele a livello internazionale, forniscono ottimi esempi in termini di efficacia dei sistemi di trasporto e di distribuzione, di buona manutenzione delle strutture. Possibili soluzioni verso le quali orientare la ricerca a livello mondiale per arginare la progressiva riduzione delle riserve idriche sono: • Revisione dei costi idrici. • Recupero delle riserve idriche. • Raccolta dell’acqua piovana. • Miglioramento degli impianti di distribuzione idrica. • Scelta di metodi di coltivazione idonei alle caratteristiche locali. • Scelta di colture a basso fabbisogno idrico. • Conservazione della risorsa attraverso un sistema di irrigazione efficiente. • Opere di desalinizzazione. • Trasferimento idrico su vasta scala da un’area all’altra. Si tratta di tecniche la cui attuazione può incrementare la disponibilità media pro capite di acqua, in particolare di coloro i quali non ne posseggono in quantità sufficiente per bere o sono privi di sistemi igienico-sanitari sufficienti.

Production Lot Detection through Synthetic Generated Data

Correctness of information gathered in production environments is an essential part of quality assurance processes in many industries, this task is often performed by human resources who visually take annotations in various steps of the production flow. Depending on the performed task the correlation between where exactly the information is gathered and what it represents is more than often lost in the process. The lack of labeled data places a great boundary on the application of deep neural networks aimed at object detection tasks, moreover supervised training of deep models requires a great amount of data to be available. Reaching an adequate large collection of labeled images through classic techniques of data annotations is an exhausting and costly task to perform, not always suitable for every scenario. A possible solution is to generate synthetic data that replicates the real one and use it to fine-tune a deep neural network trained on one or more source domains to a different target domain. The purpose of this thesis is to show a real case scenario where the provided data were both in great scarcity and missing the required annotations. Sequentially a possible approach is presented where synthetic data has been generated to address those issues while standing as a training base of deep neural networks for object detection, capable of working on images taken in production-like environments. Lastly, it compares performance on different types of synthetic data and convolutional neural networks used as backbones for the model.

A study on tackling visual odometry by a transformer architecture

This dissertation describes a deepening study about Visual Odometry problem tackled with transformer architectures. The existing VO algorithms are based on heavily hand-crafted features and are not able to generalize well to new environments. To train them, we need carefully fine-tune the hyper-parameters and the network architecture. We propose to tackle the VO problem with transformer because it is a general-purpose architecture and because it was designed to transformer sequences of data from a domain to another one, which is the case of the VO problem. Our first goal is to create synthetic dataset using BlenderProc2 framework to mitigate the problem of the dataset scarcity. The second goal is to tackle the VO problem by using different versions of the transformer architecture, which will be pre-trained on the synthetic dataset and fine-tuned on the real dataset, KITTI dataset. Our approach is defined as follows: we use a feature-extractor to extract features embeddings from a sequence of images, then we feed this sequence of embeddings to the transformer architecture, finally, an MLP is used to predict the sequence of camera poses.

Towards the sustainable production of methyl methacrylate: A comparative analysis of industrial chemical production processes using life cycle assessment methodology

Increasing environmental awareness has been a significant driving force for innovations and process improvements in different sectors and the field of chemistry is not an outlier. Innovating around industrial chemical processes in line with current environmental responsibilities is however no mean feat. One of such hard to overhaul process is the production of methyl methacrylate (MMA) commonly produced via the acetone cyanohydrin (ACH) process developed back in the 1930s. Different alternatives to the ACH process have emerged over the years and the Alpha Lucite process has been particularly promising with a combined plant capacity of 370,000 metric tonnes in Singapore and Saudi Arabia. This study applied Life Cycle Assessment methodology to conduct a comparative analysis between the ACH and Lucite processes with the aim of ascertaining the effect of applying principles of green chemistry as a process improvement tool on overall environmental impacts. A further comparison was made between the Lucite process and a lab-scale process that is further improvement on the former, also based on green chemistry principles. Results showed that the Lucite process has higher impacts on resource scarcity and ecosystem health whereas the ACH process has higher impacts on human health. On the other hand, compared to the Lucite process the lab-scale process has higher impacts in both the ecosystem and human health categories with lower impacts only in the resource scarcity category. It was observed that the benefits of process improvements with green chemistry principles might not be apparent in some categories due to some limitations of the methodology. Process contribution analysis was also performed and it revealed that the contribution of energy is significant, therefore a sensitivity analysis with different energy scenarios was performed. An uncertainty analysis using Monte Carlo analysis was also performed to validate the consistency of the results in each of the comparisons.

Biopolyester fibers for sustainable and biocompatible applications

The rising of concerns around the scarcity of non-renewable resources has raised curiosity around new frontiers in the polymer science field. Biopolymers is a general term describing different kind of polymers that are linked with the biological world because of either monomer derivation, end of life degradation or both. The current work is aimed at studying one example of both biopolymers types. Polyhydroxibutyrate (P3HB) is a biodegradable microbial-produced polymer which holds massive potentiality as a substitute of polyolefins such as polypropylene. Though, its highly crystalline nature and stereoregularity of structure make it difficult to work with. The project P3HB-Mono take advantage of polarized Raman spectroscopy to see how annealing of chains with different weights influence the crystallinity and molecular structure of the polymer, eventually reflecting on its mechanical properties. The technique employed is also optimal in order to see how mesophase, a particular conformation of chains different from crystalline and amorphous phase, develops in the polymer structure and changes depending on temperature and mechanical stress applied to the fiber. Polycaprolactone (PCL) on the other hand is a biodegradable fossil-fuel polymer which has biocompatibility and bio-resorbability features. As a consequence this material is very appealing for medical industry and can be used for different applications in this field. One interesting option is to produce narrow and long liquid filled fibers for drug delivery inside human body, using a traditional technique in an innovative way. The project BioLiCoF investigates the feasability of producing liquid filled fibers using melt-spinning techniques and will examine the role that melt-spinning parameters and liquids employed as a core solution have on the final fiber. The physical analysis of the fibers is also interpreted and idea on future developments of the trials are suggested.

Synthetic generation of amyloid PET images by non-linear dimensionality reduction inversion

Privacy issues and data scarcity in PET field call for efficient methods to expand datasets via synthetic generation of new data that cannot be traced back to real patients and that are also realistic. In this thesis, machine learning techniques were applied to 1001 amyloid-beta PET images, which had undergone a diagnosis of Alzheimer’s disease: the evaluations were 540 positive, 457 negative and 4 unknown. Isomap algorithm was used as a manifold learning method to reduce the dimensions of the PET dataset; a numerical scale-free interpolation method was applied to invert the dimensionality reduction map. The interpolant was tested on the PET images via LOOCV, where the removed images were compared with the reconstructed ones with the mean SSIM index (MSSIM = 0.76 ± 0.06). The effectiveness of this measure is questioned, since it indicated slightly higher performance for a method of comparison using PCA (MSSIM = 0.79 ± 0.06), which gave clearly poor quality reconstructed images with respect to those recovered by the numerical inverse mapping. Ten synthetic PET images were generated and, after having been mixed with ten originals, were sent to a team of clinicians for the visual assessment of their realism; no significant agreements were found either between clinicians and the true image labels or among the clinicians, meaning that original and synthetic images were indistinguishable. The future perspective of this thesis points to the improvement of the amyloid-beta PET research field by increasing available data, overcoming the constraints of data acquisition and privacy issues. Potential improvements can be achieved via refinements of the manifold learning and the inverse mapping stages during the PET image analysis, by exploring different combinations in the choice of algorithm parameters and by applying other non-linear dimensionality reduction algorithms. A final prospect of this work is the search for new methods to assess image reconstruction quality.