Democritus University of Thrace (DUTH)
Ανάπτυξη κατανεμημένου δικτύου ευφυών πρακτόρων υβριδικής νόησης, για την ανάλυση – παρακολούθηση περιβαλλοντικών προβλημάτων με σκοπό τη διαχείρισή τους σε πραγματικό χρόνο
Abstract
dc:descriptionThe process of analyzing, monitoring and managing complex environmental problems, has often led to the modeling and development of complex, monolithic systems strictly oriented towards a specific narrow approach. The main reason for this is the complexity and nonlinearity of environmental problems, which specifically requires the adoption of Soft Computing methods.The aim of this thesis is to develop new algorithms and hybrid approach standards that will lead to the design and implementation of an integrated innovative intelligent computational Intelligence system. The system will use state of the art technologies and will have the capacity of modeling, analyzing real time monitoring and control, of a wide range of complex and crucial environmental cases through a perpetual two-way process.The system modifies the environment through actions that it imposes, but then again the environment offers new stimuli that determine the need for new actions by the system. So through this bidirectional feed, a real time continuous monitoring and addressing of contemporary environmental problems is achieved. The object-oriented multiparametric system comprises of a large layered network of independent intelligent hybrid cognitive agents. These agents are invited to receive various real time stimuli from the environment, to calculate and process complex data and parameters on a second level, to implement Fuzzy Logic models that will enable understanding of real world concepts and to pay Mathematical meaning in Linguistics. Finally, they are invited to use a complex set of rules in order to choose actions that modify the environment, aiming to solve the current problem. In addition, another set of agents, facilitates the modeling of the problem, as well as the development and use of competitive neural networks ensembles, that achieve the forecasting of the evolution of the phenomenon and ensure the real time function even in cases of large loss of primary data. Finally, on another level, there are intelligent agents that control the proper functioning of the system and ensure early detection of potential malfunctions.The distributed and modular structure of the system, in conjunction with the encorporated Database, enables its interface with existing systems and/or its easy potential expansion for a wide variety of potential problems. The functionality and efficiency of the system was tested and validated in three independent environmental problems of different nature and orientation in the framework of two Major, distinct competitive research projects, under the National Strategic Framework of Development (ESPA) (Cooperation) and (Enforcement of Small Medium Enterprises).
Degree
thesis:*- Grantor dc:publisher
- Democritus University of Thrace (DUTH)
- Year dc:date
- 2016
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Papaleonidas, Antonios
- Παπαλεωνίδας, Αντώνιος
Subjects
dc:subject × 28- Ασαφή σύνολα
- Ασαφής λογική
- Βάσεις δεδομένων
- Δίκτυα Ευφυών Πρακτόρων
- Ευφυή υπολογιστικά συστήματα
- Κατανεμημένα Δίκτυα Ευφυών Πρακτόρων
- Περιβαλλοντικά προβλήματα
- Πολυπρακτορικά συστήματα
- Συστάδες Νευρωνικών Δικτύων
- Συστήματα πραγματικού χρόνου
- Συστήματα λήψης αποφάσεων
- Υπολογιστική νοημοσύνη
- Fuzzy logic
- Data bases
- Intelligent agents
- Computational intelligence
- Multiagent system
- Distributed systems
- Artificial neural networks
- Real time systems
- Decision support systems
- Environmental problems
- Φυσικές Επιστήμες
- Επιστήμη Ηλεκτρονικών Υπολογιστών και Πληροφορική
- Γεωεπιστήμες και Επιστήμες Περιβάλλοντος
- Natural Sciences
- Computer and Information Sciences
- Earth and Related Environmental Sciences
Rights
dc:rights- Statement dc:rights
-
- BY
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/38104