1,720,956 research outputs found
Efficient control of domestic space heating systems and intermittent energy resources
No full textMeeting the ever-growing global energy demand while reducing carbon emissions is one of the most prominent challenges of our era. In this context, efficient control of an operation, service or production process is a key tool to achieve this goal. While there are many opportunities for efficient control within the energy sustainability agenda, this work focuses on domestic space heating systems and intermittent energy resources. This is because in many countries, such as the UK and the US, the domestic sector accounts for more than 20% of the total energy consumption and over 40% of this share is related to space heating. In addition, in recent years, an increasing number of intermittent energy resources, such as photovoltaic systems and wind turbine generators are being integrated into the grid. As such, efficient control of domestic space heating systems and intermittent energy resources can lead to a major reduction in energy consumption and the corresponding CO2 emission.In more detail, domestic space heating automation systems (DHASs) aim to optimize the control process of domestic space heating systems with minimum user-input. Moreover, in the case of electricity-based heating, such systems can also incorporate economic control to exploit the energy buffer that heating loads provide in order to shift the heating consumption according to financial incentives, such as variable electricity import tariffs and/or the availability of cheap electricity coming from house-integrated intermittent energy resources. In the latter case, the financial benefits of economic control can be further amplified in domestic coalitions where a number of houses share their energy generation to minimize the collective energy imported from the grid.Against this background, the first main strand of work in this thesis is to develop a new DHAS, AdaHeat, that overcomes limitations of previous approaches regarding: (i) their efficiency in dealing with the thermal dynamics of houses, (ii) their efficiency in dealing with the inherent uncertainty of the occupancy schedule in domestic settings, (iii) their usability and effectiveness in meeting the user preferences, (iv) their ability to work in conjunction with a diverse range of heating systems, and (v) their ability to efficiently consider economic control in the case of electricity-based heating, exploiting also, for the first time, the aforementioned coalition potential. The backbone of AdaHeat is an adaptive model predictive control approach along with a new general heating schedule planning algorithm based on dynamic programming. In the case of economic control in the presence of house-integrated intermittent energy resources, our planning approach relies on stochastic predictions of the shared intermittent energy resource power output. To this end, we also develop a new adaptive site-specific calibration technique to improve such predictions based on Gaussian process modeling. We present thorough evaluation of the proposed system, and show its effectiveness in terms of Pareto efficiency and usability criteria against state-of-the-art DHASs. We also show that collective economic control, in the presence of house-integrated IERs, can improve heating cost-efficiency by up to 60%, compared to independent economic control, and even more when compared to no economic control.The second strand of work is concerned with increasing the efficiency of intermittent energy resources themselves, through efficient control. In particular, specifically for photovoltaic systems, solar tracking can be used to orient the system towards the greatest possible levels of incoming solar irradiance. This can increase the power output of a photovoltaic system by up to 100%. However, current solar tracking techniques suffer from several drawbacks: (i) they usually do not consider the forecasted or prevailing weather conditions; even when they do, they (ii) rely on complex closed-loop controllers and sophisticated instruments; and (iii) typically, they do not take the energy consumption of the trackers into account. As such, in this work, we propose PreST; a novel, low-cost and generic solar tracking approach that overcomes the above limitations, utilizing optimal control (proposed for the first time for solar tracking). In particular, our approach is able to calculate appropriate trajectories for efficient and effective day-ahead (predictive) solar tracking, based on available weather forecasts (that can come from on-line providers for free). To this end, we propose a new approximating policy iteration algorithm, suitable for large Markov decision processes, and a novel and generic solar tracking consumption model. Our simulations show that our approach can increase the power output of a photovoltaic system considerably, when compared to standard solar tracking techniques, that can lead to significant monetary gains.As outlined above, apart from their great share in contemporary economies, both domestic space heating systems and intermittent energy resources provide considerable opportunities for energy efficient improvements through efficient control. In this work we exploit this potential and propose respective systems that improve their independent, as well as their interaction, efficiency. This can considerably reduce the respective energy consumption and the corresponding CO2 emission towards fulfilling our goal for an energy sustainable future
Μηχανική μάθηση για χαμηλού κόστους πρόβλεψη ηλιακής ακτινοβολίας σε ευρεία περιοχή
No full textΔιπλωματική εργασία που κατατέθηκε στη σχολή ΗΜΜΥ του Πολυτεχνείου Κρήτης για την πλήρωση των προϋποθέσεων λήψης προπτυχιακού διπλώματος.Summarization: In the past few years, solar radiation prediction has been paramount in a multitude of sectors, from energy production via renewable energy sources, to tracking climate change, among others.
So far, work in the area lacks in large area coverage, ease of access, or uses past solar radiation readings, relying on related equipment being already on-site. In this work, we provide insight into the efficacy of neural networks in the area, accompanied with data sourced from varying providers.
In order to achieve this, we create and vet a weather reading dataset from a large variety of stations, which are more indicative of what smaller organizations or individuals may have access to, instead of more tailored datasets. We utilize this dataset to train a number of neural networks, each with different architectures, and evaluate their results so as to set a standard to be improved upon in later work utilizing a similar type of dataset.
The results indicate that, even utilizing a much broader dataset than what has been used in the past, neural networks show promise in this area, especially with more targeted implementations.Περίληψη: Τα τελευταία χρόνια, η πρόβλεψη ηλιακής ακτινοβολίας έχει γίνει αναπόσπαστο κομμάτι πλήθους τομέων, όπως, μεταξύ άλλων, η παραγωγή ενέργειας μέσω ανανεώσιμων πηγών και η παρακολούθηση της κλιματικής αλλαγής.
Ως τώρα, οι σχετικές εργασίες γύρω από το θέμα είτε καλύπτουν πολύ μικρό χώρο, είτε είναι δυσπρόσιτες/μη προσβάσιμες από τον περισσότερο κόσμο, είτε χρησιμοποιούν χρονοσειρές μετρήσεων ηλιακής ακτινοβολίας ως εισόδους, βασιζόμενες στην ύπαρξη σχετικού εξοπλισμού στο σημείο-στόχο. Στην παρούσα διπλωματική εργασία, εξετάζουμε την αποδοτικότητα των νευρωνικών δικτύων εκπαιδευμένων με δεδομένα από διάφορες πηγές.
Συγκεκριμένα, δημιουργούμε και ελέγχουμε ένα σύνολο μετρήσεων καιρικών συνθηκών από μεγάλο πλήθος σταθμών, ως πιο ενδεικτικό του τύπου δεδομένων στα οποία μπορούν να έχουν πρόσβαση μικρότερες οργανώσεις ή επιμέρους άτομα. Προτεραιοποιούμε τη χρήση τέτοιων μετρήσεων έναντι της χρήσης πιο προσαρμοσμένων δεδομένων. Αξιοποιούμε το σύνολο δεδομένων μας για να εκπαιδεύσουμε ένα πλήθος νευρωνικών δικτύων με διαφορετικές αρχιτεκτονικές, και αξιολογούμε τα αποτελέσματά τους ώστε να θέσουμε ένα πρότυπο προς βελτίωση σε μελλοντικές εργασίες που θα χρησιμοποιούν ένα παρόμοιο σύνολο δεδομένων.
Τα αποτελέσματα δείχνουν ότι ακόμη και με ένα πολύ μεγαλύτερο/ευρύτερο σύνολο δεδομένων από όσα έχουν χρησιμοποιηθεί ως τώρα, συγκεκριμένες στοχευμένες υλοποιήσεις νευρωνικών δικτύων μπορεί να είναι αποτελεσματικές στο συγκεκριμένο πρόβλημα
Μία καινοτόμος μετα-ευρετική μέθοδος αναζήτησης για καθολική βελτιστοποίηση συνεχών συναρτήσεων
No full textSummarization: Artificial intelligence research in optimization and search is concerned with reaching the maxima or minima of an objective function, while potentially searching among a vast range of value choices for the function’s variables. Global continuous optimization methods, in particular, seek to reach the optima of complex continuous mathematical functions. Meta-heuristics are commonly used in order to solve such problems. Typically, however, meta-heuristics originally designed for solving discrete optimization problems are later adapted to continuous tasks, which consumes considerable time. Also, there is a chance that they will get stuck to local optima as the complexity of configuration spaces increases. Furthermore, generally meta-heuristics accept worse solutions, in order to achieve a broader exploration of the configuration space. This results in algorithms that do not improve in an anytime manner, and an arbitrary interruption of the algorithm’s flow, can lead to a waste of computation time as the current solution might be worse than a solution discovered earlier on.
In this work, a novel single-point meta-heuristic is proposed, which is specifically designed to tackle continuous optimization problems. Our algorithm, Buggy Pinball, is an anytime algorithm inspired by the well-known pinball game: It employs a trajectory-based search, and each proposed solution ensures the improvement of the configuration space. In order to evaluate our algorithm, we used a number of standard testbed functions, which can also be applied on multiple dimensions. We compared our results to the performance of some of the most widely used meta-heuristics, namely simulated annealing, threshold accepting, and particle swarm optimization. Our systematic evaluation shows that our algorithm is very efficient in global continuous optimization tasks, with performance that is particularly successful in complex configuration spaces.Περίληψη: Ένας εκ των παραδοσιακών πυλώνων της Τεχνητής Νοημοσύνης είναι αυτός που ασχολείται με τεχνικές αναζήτησης και βελτιστοποίησης, που επιχειρούν την εύρεση της βέλτιστης λύσης εντός ενός μεγάλου φάσματος επιλογών. Η καθολική συνεχής βελτιστοποίηση ασχολείται με την επίλυση προβλημάτων βελτιστοποίησης πολύπλοκων συνεχών συναρτήσεων. Οι μετα-ευρετικοί αλγόριθμοι χρησιμοποιούνται ευρέως για την επίλυση τέτοιων προβλημάτων. Οι περισσότεροι όμως μετα-ευρετικοί αλγόριθμοι, έχουν σχεδιαστεί για την επίλυση διακριτών προβλημάτων και αργότερα αναπροσαρμόστηκαν για τη χρήση τους σε συνεχή προβλήματα. Αυτό αυξάνει το χρονικό κόστος εξεύρεσης λύσης. Επίσης, η πιθανότητα να καταλήξουν σε τοπικά βέλτιστα αυξάνεται με την πολυπλοκότητα του χώρου. Ακόμη, τείνουν να αποδέχονται χειρότερες λύσεις, για να επιτύχουν πιο ευρεία εξερεύνηση του χώρου. Αυτό έχει σαν αποτέλεσμα να αποτελούν συνήθως μεθόδους αναζήτησης, οι οποίες δεν βελτιώνονται συνεχώς με την πάροδο του χρόνου. Ως εκ τούτου, μια πιθανή πρόωρη διακοπή της ροής του αλγορίθμου, μπορεί να καταστήσει μεγάλο μέρος της πρότερης υπολογιστικής διαδικασίας κενό νοήματος, αφού οι τελευταίες λύσεις μπορεί να είναι χειρότερες από την καλύτερη που έχει βρεθεί ως εκείνη τη στιγμή.
Στην παρούσα διπλωματική εργασία προτείνεται ένας νέος μετα-ευρετικός αλγόριθμος μονής-λύσης που είναι σχεδιασμένος για να λύνει συνεχή προβλήματα και που μπορεί να διακοπεί ανά πάσα στιγμή με την εγγύηση ότι η τελευταία λύση θα είναι πάντα καλύτερη από τις προηγούμενες. Ο αλγόριθμός μας, επονομαζόμενος "ελαττωματικό φλιπεράκι", είναι εμπνευσμένος από το διαδεδομένο παιχνίδι φλίπερ, όπου εφαρμόζεται ένας τρόπος εύρεσης με τη δημιουργία τροχιάς και κάθε καινούρια λύση βελτιώνει την ήδη υπάρχουσα. Αξιολογήσαμε τον αλγόριθμό μας σε πλείστες κλασσικές συναρτήσεις συνεχούς βελτιστοποίησης, και συγκρίναμε τις επιδόσεις του με αυτές κάποιων από τους πιο διαδεδομένους μετα-ευρετικούς αλγορίθμους αναζήτησης: την προσομοιωμένη ανόπτηση, την αποδοχή ορίου, και την βελτιστοποίηση σμήνους σωματιδίων. Η συστηματική διαδικασία αξιολόγησης που εφαρμόσαμε, αποδεικνύει την αποτελεσματικότητα του αλγορίθμου μας σε καθολικά συνεχή προβλήματα, και ιδιαίτερα την σημαντική υπεροχή του έναντι των ανταγωνιστών του ειδικά σε πολύπλοκους χώρους αναζήτησης
Going Beyond Counting First Authors in Author Co-citation Analysis
Get PDFThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
Variations on the Author
Get PDF“Variations on the Author” discusses two of Eduardo Coutinho’s recent films (Um Dia na Vida, from 2010, and Últimas Conversas, posthumously released in 2015) and their contribution to the general question of documentary authorship. The director’s filmography is characterized by a consistent yet self-effacing form of authorial self-inscription: Coutinho often features as an interviewer that rather than express opinions propels discourses; an interviewer that is good at listening. This mode of self-inscription characterizes him as an author who is not expressive but who is nonetheless markedly present on the screen. In Um Dia na Vida, however, Coutinho is completely absent form the image, while Últimas Conversas, on the contrary, includes a confessional prologue that moves the director from the margins to the center of his films. This article examines the ways in which these works stand out in the filmography of a director who offers new insights into the notion of cinematic authorship
Appropriate Similarity Measures for Author Cocitation Analysis
Get PDFWe provide a number of new insights into the methodological discussion about author cocitation analysis. We first argue that the use of the Pearson correlation for measuring the similarity between authors’ cocitation profiles is not very satisfactory. We then discuss what kind of similarity measures may be used as an alternative to the Pearson correlation. We consider three similarity measures in particular. One is the well-known cosine. The other two similarity measures have not been used before in the bibliometric literature. Finally, we show by means of an example that our findings have a high practical relevance.information science;Pearson correlation;cosine;similarity measure;author cocitation analysis
Dispelling the Myths Behind First-author Citation Counts
Get PDFWe conducted a full-scale evaluative citation analysis study of scholars in the XML research field to explore just how different from each other author rankings resulting from different citation counting methods actually are, and to demonstrate the capability of emerging data and tools on the Web in supporting more realistic citation counting methods. Our results contest some common arguments for the continued
use of first-author citation counts in the evaluation of scholars, such as high correlations between author rankings by first-author citation counts and other citation
counting methods, and high costs of using more realistic citation counting methods that are not well-supported by the ISI databases. It is argued that increasingly available digital full text research papers make it possible for citation analysis studies to go beyond what the ISI databases have directly supported and to employ more
sophisticated methods
koamabayili/VECTRON-author-checklist: VECTRON author checklist
No full textWe have done our best to complete the author checklist relating to the use of animals in the hut study. Note that the objective for the hut study was to evaluate the IRS treatment applications for residual efficacy against Anopheles mosquitoes, including the local An. coluzzii mosquito population. Cows were only used to attract mosquitoes into the huts and no tests were carried out directly on the cows. The author checklist is intended for use with studies where experiments are carried out on animals, which is why we have had such difficulty in completing this for the hut study, as many of the questions do not relate to how the cows were used
- …
