Synthesis of regional networks for biomass and biofuel production Hon Loong Lam
, 2010, dissertation
Abstract: This thesis presents two different approaches to the synthesis of regional networks for biomass and biofuel production and supply: Mathematical Programming and Graph Theoretic approach. The optimisation criterion for both approaches is the maximisation of profit.
The first approach is based on a generic optimisation model of biomass production and supply networks. This superstructure approach is based on a flexible number of network layers: plantation, collection using a pre-treatment, process, and consumption. A Mixed Integer Linear Programming (MILP) model has been successfully developed during this work.
However, the solution of this biomass production network model is very challenging due to the large sizes of the networks and the number of interconnections. The huge number of redundant variables reduces model efficiency (time taken to solve the model and the interpretation of the results). This model when representing very large size networks cannot be solved over a reasonable time even by professional mathematical programming software tools. Several model-size reduction techniques are therefore proposed for the solution of large-scale networks. In particular, methods are proposed for (i) reducing the connectivity within a biomass supply chain network by setting the maximum allowable distance between the supply zones to the collection centres, (ii) eliminating unnecessary variables and constrains to reduce the zero-flows in the full model, and (iii) aggregating the network and hence the synthesis process by merging the collection centres.
The network synthesis is also carried out by P-graph (Process Graph) tools. P-graph is a directed bipartite graph, having two types of vertices — one for operating units and another for those objects representing material or energy flows/quantities. In this procedure, firstly a maximum feasible superstructure for biomass production network is generated from which the optimal structure is then selected by the Branch and Bound method. This graph-based method clearly shows where, how, and what kind of material and energy carriers will be transferred from one supply chain layer to another.
In order to test the efficiency of the model, a small regional renewable network problem was solved using both methods. Their performances were tested and the results confirmed the applicability on a regional scale. The proposed model-size reduction techniques were also tested. A large-scale regional case study was created to demonstrate these techniques. The results are very positive and some suggestions for future work are given in the conclusion.
Keywords: Biomass and bioenergy network synthesis, Model-size reduction
techniques, Mathematical Programming, MILP, P-Graph
Published: 06.01.2011; Views: 2607; Downloads: 87
Full text (4,25 MB)
The complexity of porous structure of building materialsMarko Samec
, 2011, dissertation
Abstract: This thesis seeks to establish the link between the structure (in a topological sense) of porous space and charged particle dynamics in porous matter, specifically in constituent elements of sustainable building materials such as clay, cement and soil. The work done is a combination of experimental research and modelling of analysed data using advanced and expanded network models to model pore structure and generalized conductivity model. The main outcome of this doctoral thesis is the demonstration that there is a correlation between the large scale structure of the pore space and the properties of the motion of charged particles through the pore space. This was achieved by conducting two experiments: the structure of pore space of selected porous materials (soil samples, clays, cements, clay-cement mixtures) was investigated using state-of-the-art X-ray computed microtomography, while the dynamics of charged particles in the samples was probed using low-frequency dielectric spectroscopy. The research done and described in the thesis is directed towards the advancement of understanding the transport phenomena and the structure of porous media which is of paramount importance for solving problems in building physics dealing with moist transport in building's envelope, the building-ground interaction, and in transport of contaminants in the vicinity of the repositories where the transfer of moist through soil can be the source of contamination.
Keywords: porous matter, clay-water system, hydrating cement, fractional dynamics, dielectric response, X-ray computed tomography, image analysis, complex network
Published: 11.05.2011; Views: 3556; Downloads: 152
Full text (34,69 MB)
Approach to optimization of cutting conditions by using artificial neural networksFranc Čuš
, Uroš Župerl
, 2006, original scientific article
Abstract: Optimum selection of cutting conditions importantly contribute to the increase of productivity and the reduction of costs, therefore utmost attention is paid to this problem in this contribution. In this paper, a neural network-based approach to complex optimization of cutting parameters is proposed. It describes the multi-objective technique of optimization of cutting conditions by means of the neural networks taking into consideration the technological, economic and organizational limitations. To reach higher precision of the predicted results, a neural optimization algorithm is developed and presented to ensure simple, fast and efficient optimization of all important turning parameters. The approach is suitable for fast determination of optimum cutting parameters during machining, where there is not enough time for deep analysis. To demonstrate the procedure and performance of the neural network approach, an illustrative example is discussed in detail.
Keywords: optimization, cutting parameter optimization, genetic algorithm, cutting parameters, neural network algorithm, machining, metal cutting
Published: 30.05.2012; Views: 1457; Downloads: 75
Link to full text
Advanced modelling and simulation methods for communication networksJože Mohorko
, Matjaž Fras
, Saša Klampfer
, 2008, original scientific article
Abstract: This paper contains a brief overview of suitable simulation tools available for education and research on network technologies and protocols. Among the mentioned tools we present advanced methods for network simulations using the OPNET modeler simulation tool. This is one of the most widespread simulation tools for network simulations, appropriate for both for teaching and also for the researching procedures of new internet devices and protocols. The basic package is intended for simulating communication networks and for protocols and devices development. There are also additional specific modules, such as a module assigned to the simulation of wireless networks, an ACE module for application analyzing, 3DNV module for visualizing networks on virtual terrain and the "System in the loop" module for simulating networks with real communication equipment in the loop, in real-time.
Keywords: network simulation, simulation tools, modelling, communication networks, OPNET
Published: 31.05.2012; Views: 1099; Downloads: 29
Link to full text
Prediction of surface roughness using a feed-forward neural networkJernej Šenveter
, Simon Klančnik
, Jože Balič
, Franc Čuš
, 2010, original scientific article
Abstract: This article presents the development of a system for predicting surface roughness, using a feed-forward neural network. The primary goal was to develop a system in order to predict with complex reliability and defined accuracy. However, this system is designed in such a way that it is also possible to use it for various other workpieces. The described system uses a neural network which receives signals at the input level. The signals then travel through all hidden levels to the output level, where the responses to input signals are received. Data are used which affects the selection of surface roughness regarding the input to the neural network. Three different inputs in total are used for the neural network. Data which represents the inputs to the neural network are encoded, so that they occupy values between 0 and 1. Adequate cutting speed, feed, and depth of cut, are selected in order to achieve an adequate surface roughness of the workpiece, using the trained neural network. This contributes to the optimisation and economy of machining, which is very important during the production of an individual product and also for an individual company or organisation when transferring the final product to the contracting authority or final customer.
Keywords: machining, turning, surface roughness, neural network
Published: 31.05.2012; Views: 1094; Downloads: 42
Link to full text
Intelligent cutting tool condition monitoring in millingUroš Župerl
, Franc Čuš
, Jože Balič
, 2011, original scientific article
Abstract: Purpose: of this paper is to present a tool condition monitoring (TCM) system that can detect tool breakage in real time by using a combination of neural decision system, ANFIS tool wear estimator and machining error compensation module. Design/methodology/approach: The principal presumption was that the force signals contain the most useful information for determining the tool condition. Therefore, ANFIS method is used to extract the features of tool states from cutting force signals. The trained ANFIS model of tool wear is then merged with a neural network for identifying tool wear condition (fresh, worn). Findings: The overall machining error is predicted with very high accuracy by using the deflection module and a large percentage of it is eliminated through the proposed error compensation process. Research limitations/implications: This study also briefly presents a compensation method in milling in order to take into account tool deflection during cutting condition optimization or tool-path generation. The results indicate that surface errors due to tool deflections can be reduced by 65-78%. Practical implications: The fundamental limitation of research was to develop a single-sensor monitoring system, reliable as commercially available system, but much cheaper than multi-sensor approach. Originality/value: A neural network is used in TCM as a decision making system to discriminate different malfunction states from measured signals.
Keywords: tool condition monitoring, TCM, wear, tool deflection, ANFIS, neural network, end-milling
Published: 01.06.2012; Views: 905; Downloads: 22
Link to full text