Fjärrvärme är ett uppvärmningsalternativ framlyft som en  potentiellt  viktig beståndsdel för att lösa  EU:s satta klimatmål att minska koldioxidutsläppen, effektivisera energiförbrukningen och öka andelen  förnyelsebar  energi,  de  så  kallade  20-20-20-målen.  Detta  då  fjärrvärme  möjliggör tillvaratagandet av spillvärme från industrier och  centraliserade  förbränningar  vilka kan använda annars outnyttjade resurser såsom avfall och grot. Stora förbränningsanläggningar möjliggör även en förbättrad rening av rökgaser. Falun och Borlänge i Dalarna, Sverige, har  sedan 1984 och 1969 haft  fjärrvärmesystem  vilkadrivs av  Falu Energi & Vatten respektive  Borlänge Energi. I samband med att  de två bolagen skulle förbinda sina  vattensystem  togs även beslutet att bilda ett  gemensamt  fjärrvärmenät. Detta kommer förverkligas genom att Grundledningen  hösten 2014  kommer  att  färdigställas. Det gemensamma energisystemet Falun-Borlänge kommer försörjas av tre  kraftvärmeverk,  ett vilket förbränner avfall i Borlänge och två som förbränner biobränsle i Falun. Systemet försörjs även med spillvärme från Stora Enso Kvarnsvedens pappersbruk och stålverket SSAB i Borlänge.Genom att de både systemet i nuläget har olika  temperaturstyrkurvor  måste dessa samordnas till en gemensam körstrategi för att kunna tillvarata den potential det gemensamma energisystemetmedför,  varav  det  gemensamma  ägda  dotterbolaget  Grundledningen  HB  skapades,  vilken  ärsamordningsansvarig  av  driften.  Detta  examensarbete  har  genomförts  i  samarbete  med Grundledningen  HB  med  målet  att  ta  fram  en  ny  gemensam  körstrategi  för  det  kommande energisystemet Falun-Borlänge. Den  nya  körstrategin  har  framtagits  genom  att  beräkna  en  ekonomisk  optimal framledningstemperatur för olika utetemperaturer. Vilken bestämts genom att analysera systemet,genom  att  ställa  det  som  ett  linjärprogrammeringsproblem,  och  ta  hänsyn  till  olika flödesbegränsningar samt temperaturkrav som finns i systemet.Rapporten  tar  fram  totalt  tre  olika  körstrategier  beroende  på  möjlighet  att  kontrollera distributionsnätet,  resultaten har  validerats  genom en känslighetsanalys. De framtagna styrkurvorna är  i  vissa  lägen  högre  än  de  nuvarande  använda  vilket  är  en  konsekvens  av  samordningen  av systemet. / District heating is a heating alternative that deserves attention  as a  possible  key contributor in the  fulfilment  of  European  Union  energy  and  climate  goals,  the  20-20-20  goals,  to  reduce carbondioxide  emissions, improve the overall energy efficiency and increase the share of renewable energy. This is because district heating allows  industrial  excess heat recovery  and also because it presents the possibility to use otherwise unused resources such as waste and forest residue as fuel. The cities Falun and Borlänge in Dalarna, Sweden, has since 1984 and 1969 operational district heating systems, which are owned and driven by the  municipal corporations: Falu Energi &  Vatten (Falu Energy & Water) and Borlänge Energi (Borlänge Energy). When the two companies decided to collaborate  and connect their freshwater systems the decision was made to also connect the two district heating systems. The soon-to-be district heating system Falun-Borlänge will be supplied by cogeneration plants in Borlänge with waste as fuel and in Falun biofuels, but also with excess heat from the Stora Enso Kvarnsveden paper mill and the steel plant SSAB in Borlänge.Today the two systems  has different control strategies  which has to become one in order to coordinate the systems. To make this happen the two companies founded the subsidiary company Grundledningen  HB  which  is  to  be  responsible  for  the  coordination.  This  thesis  has  been conducted  with  cooperation  with  Grundledningen  HB  with  the  aim  to  develop  a  new  control strategy for the soon to be joint energy system Falun-Borlänge.The new  control strategy  was  developed  by calculating the  optimum supply temperature for different outdoor temperatures.  This was  determined by an operational analysis,  by making the problem  into  a  linear  programming  problem  and  thereafter  account  for  flow  restrictions  and temperature requirements in the system.The  result  concludes  three  different  optimal  control  strategies,  depending  on  the  ability  to control the distribution network. The  validity of the control strategies has been established through a sensitivity analysis. The developed control curves are in certain positions higher than the  ones currently used as a consequence of the coordination of the system.

A quota-based integrated commercial fishery owns fishing trawlers, processing plants, and fish quotas. Such a fishery must decide how to schedule trawlers for fishing and landing, how to schedule processing of products, how to schedule labour for processing, and how to plan inventory of raw materials and products. This problem is of great economic significance to New Zealand, whose economy depends to a large extent on the fishery industry. To assist the fishery manager, we develop a mixed integer linear program (MILP) for optimal scheduling of fishing trawlers, production planning (processing) and labour allocation for a quota-based integrated fishery of New Zealand. The model decides when and where each trawler should go for fishing, how much fish each trawler should land, and how much product to produce in each period. Since the fishery is a private farm, its main objective will be profit maximization (or cost minimization if its demand is on contract). The government manages the conservation of fish through the quota allocation. In this thesis the objective of the fishery model is to maximise the total profit. We demonstrate our model with examples based on data from a major New Zealand fishery. We investigate ways to manage the uncertainties involved in trawler scheduling and production planning of the fishery. To manage end-of-planning-horizon effects in the fishery, we develop a simple safety stock approach. We also analyse the workability of a rolling horizon approach to solve the longer planning horizon models and to deal with the end-of-planning horizon effects. We investigate the effect of initial and final position of the trawlers on the profit. We also investigated many different challenging data sets to observe the impact on the effectiveness of our IFPM. The second objective of this thesis is to develop an efficient solution procedure for the MILP, named integrated fishery planning model (IFPM). The IFPM consists of a fishing subproblem, a processing subproblem, and complicating side constraints. We have tried techniques including LP relaxation, Lagrangean relaxation (LR), Dantzig-Wolfe decomposition (DWD) and decomposition-based pricing (DBP). We develop a new DBP method to solve the IFPM. It gives excellent computation times. We also develop a decomposition-based O'Neill pricing (DBONP) method to improve the solution obtained from DBP procedure. It improves the DBP solutions but takes longer time to solve the IFPM. Finally, we develop a simple and efficient reduced cost-based pricing (RCBP) method. It takes less time to solve the IFPM and yields excellent results. The initial formulations for several planning horizons are solved using the AMPL modelling language and CPLEX with branch and bound. Relevant results and computational difficulties are reported.

Steam boilers are used to generate steam in order to meet cold process heat requirements. The most common heat exchanger network layout associated with the boiler is a completely parallel design. This observation suggests that the flowrate of steam needed for the system can be reduced, while maintaining the required duty, simply by changing the layout of the network. Phase change of saturated steam to saturated liquid plays a vital role in the targeting method as well as the design of the network layout. A hybrid graphical and mathematical technique for targeting and network synthesis is presented. The main objective of the technique is to reduce steam flowrate without compromising duty requirements of the process heat exchangers. In order to assess the advantage of a hybrid technique, a case study is used where steam savings of 29.6% is obtained. A complete mathematical technique to demonstrate the advantage of the graphical targeting concept in solution time improvement is further presented. / Dissertation (MEng)--University of Pretoria, 2008. / Chemical Engineering / unrestricted

Pour atteindre une haute efficacité énergétique dans l'industrie, des méthodes d'intégration énergétique ont été développées. La plupart de ces travaux traitent des procédés continus. Or 50 % des procédés industriels sont des procédés discontinus ou variables dans le temps. D'où la problématique : comment effectuer l'intégration énergétique des procédés discontinus ou variables dans le temps en incluant le stockage thermique et les systèmes de conversion d'énergie tout en minimisant l'exergie consommée ? Et comment dimensionner les utilités dans ce type de procédés pour répondre aux contraintes imposées par le réseau et aux obligations d'effacement ? La première partie présente le modèle de conception de l'architecture d'intégration énergétique des procédés discontinus en incluant les stockages thermiques et des systèmes de conversions d'énergie (pompes à chaleur, ORC, machines trithermes) dans la phase de conception. Ce modèle permet le dimensionnement et la détermination de la température des stockages thermiques ainsi que des systèmes de conversion d'énergie en minimisant l'exergie consommée dans le procédé. La deuxième partie présente le modèle de dimensionnement des utilités, en prenant en compte leur fonctionnement réel à charge partielle ou soumises à des contraintes imposées par le réseau. C'est une optimisation économique permettant de déterminer le nombre, le type et la capacité des utilités permettant de satisfaire les besoins du procédé en l'intégrant énergétiquement. / Many heat integration methods were developed to achieve high-energy efficiency in industry. Most of them are dedicated to continuous processes. However, statistics show that 50 % of processes are variable or batch. Hence: how to energetically integrate variable or batch processes including heat storages and energy conversion systems while minimizing the consumed exergy in the process? How to design utilities while considering the constraints on the energy network like the fluctuation of energy prices (imposed to reduce the energy consumption on peak hours) and the energy availability (resulting from the use of renewable energies)?The first part of the thesis presents the model to design the optimal architecture of heat integration in discontinuous processes including heat storages and energy conversion systems (heat pumps, tri-thermal machines, ORC etc.). This model allows the design of heat storages. The capacity and temperatures of energy conversion systems are also determined using this model. The objective is to reduce the consumed exergy in the process.The second part present the model for utilities design taking into account the real behavior of those under part load conditions or under constraints imposed by the network. It is an economic optimization.

This thesis is concerned with the implementation of a proposed Energy Hub Management System (EHMS) mathematical model for residential appliances under Time of Use (TOU) based electricity rate structure. The objective is to shift the residential electrical energy consumption during periods of high grid energy demand to low demand periods subject to operational constraints. The customer benefits from reduced daily energy consumption and consequent reduction to its cost with minimal effect on the comfort level. Two scheduling periods are considered simultaneously. The first is a 24-interval schedule with one-hour time steps for appliances such as a dishwasher, clothes washer and dryer. The second is a 96-interval schedule with 15-minute time steps for other appliances such as refrigerator, freezer and water heater. Each appliance has been modeled as a discrete time linear dynamic system and the objective of this thesis is to make these models work in a real world situation by determining realistic estimations of the model parameters and constants. It is vital to properly calculate the mathematical model parameters as they have direct impact on the results. Minor modifications to some domestic appliance models were proposed to make the practical implementation easier. It was found that while some parameters in the mathematical model can be easily calculated based on thermodynamic equations, other parameters are hard to be calculated; therefore, a practical procedure was proposed to measure these parameters. An experiment on a small refrigerator was carried out to validate the refrigerator mathematical model and parameters measurement procedure. The resulting model is a mixed integer linear problem (MILP) and was solved using GNU Linear Programming Kit (GLPK) freeware solver. The performance of GLPK was found to be satisfactory as compared to the commercial solver CPLEX, and was particularly suitable for practical and commercial implementations.

If renewable energy and biofuels are to attain success in the market place, each step of their production and the system as a whole must be optimized to increase material and energy efficiency, reduce production cost and create a competitive alternative to fossil fuels. Systems optimization techniques may be applied to product selection, process design and integration, feedstock procurement and supply chain management to improve performance. This work addresses two problems facing a biorefinery: technology selection and feedstock scheduling in the face of varying feedstock supply and cost. Also addressed is the optimization of a biorefinery supply chain with respect to distributed processing of biomass to bio-products via preprocessing hubs versus centralized processing and facility location selection. Two formulations are proposed that present a systematic approach to address each problem. Case studies are included to demonstrate model capabilities for both formulations. The scheduling model results display model sensitivity to feedstock price and transport distance penalized through carbon dioxide emissions. The distributed model shows that hubs may be used to extend the operating radius of a biorefinery and thereby increase profits.

This thesis is concerned with the implementation of a proposed Energy Hub Management System (EHMS) mathematical model for residential appliances under Time of Use (TOU) based electricity rate structure. The objective is to shift the residential electrical energy consumption during periods of high grid energy demand to low demand periods subject to operational constraints. The customer benefits from reduced daily energy consumption and consequent reduction to its cost with minimal effect on the comfort level. Two scheduling periods are considered simultaneously. The first is a 24-interval schedule with one-hour time steps for appliances such as a dishwasher, clothes washer and dryer. The second is a 96-interval schedule with 15-minute time steps for other appliances such as refrigerator, freezer and water heater. Each appliance has been modeled as a discrete time linear dynamic system and the objective of this thesis is to make these models work in a real world situation by determining realistic estimations of the model parameters and constants. It is vital to properly calculate the mathematical model parameters as they have direct impact on the results. Minor modifications to some domestic appliance models were proposed to make the practical implementation easier. It was found that while some parameters in the mathematical model can be easily calculated based on thermodynamic equations, other parameters are hard to be calculated; therefore, a practical procedure was proposed to measure these parameters. An experiment on a small refrigerator was carried out to validate the refrigerator mathematical model and parameters measurement procedure. The resulting model is a mixed integer linear problem (MILP) and was solved using GNU Linear Programming Kit (GLPK) freeware solver. The performance of GLPK was found to be satisfactory as compared to the commercial solver CPLEX, and was particularly suitable for practical and commercial implementations.

Passive optical networks (PONs) are point-to-multipoint networks where a single Central Office (CO) is connected to a number of downstream Optical Network Units (ONUs) via a single optical fiber by splitting the optical signal with passive splitters. Due to technology advances and increasing bandwidth requirements, these networks have moved to last mile deployment, also known as fiber-to-the-home (FTTH). The planning of these PONs are traditionally done by hand, but automated methods can be used to decrease deployment costs and planning time. Even though a number of methods have been proposed to address this problem through the solving of integer linear programming (ILP) models, they suffer from limited availability, inaccuracies and limited scalability due to the problem complexity. This dissertation focusses on improving the accuracy of these models as well as improving scalability to a point where large-scale problems can be solved feasibly. To address this, a basic model is implemented to capture the network structure and verified accordingly. Results show this model can be solved quickly, but has large discrepancies with real-world plans. Refinements in the form of fiber duct sharing, network constraints, multiple splitter types and economies of scale among others are then incorporated into a refined model and solved. Analysis of the experimental results indicates improved accuracy and lower deployment costs, at the expense of increasing computation effort considerably. Heuristic techniques are then examined to improve computational performance, including an elementary heuristic (ELEM), the Branch Contracting Algorithm (BCA) and problem decomposition. It is demonstrated that through the use of k-means clustering, the refined model can be solved in a fraction of the time while keeping deployment costs comparably low. / Thesis (MIng (Computer and Electronic Engineering))--North-West University, Potchefstroom Campus, 2013

Passive optical networks (PONs) are point-to-multipoint networks where a single Central Office (CO) is connected to a number of downstream Optical Network Units (ONUs) via a single optical fiber by splitting the optical signal with passive splitters. Due to technology advances and increasing bandwidth requirements, these networks have moved to last mile deployment, also known as fiber-to-the-home (FTTH). The planning of these PONs are traditionally done by hand, but automated methods can be used to decrease deployment costs and planning time. Even though a number of methods have been proposed to address this problem through the solving of integer linear programming (ILP) models, they suffer from limited availability, inaccuracies and limited scalability due to the problem complexity. This dissertation focusses on improving the accuracy of these models as well as improving scalability to a point where large-scale problems can be solved feasibly. To address this, a basic model is implemented to capture the network structure and verified accordingly. Results show this model can be solved quickly, but has large discrepancies with real-world plans. Refinements in the form of fiber duct sharing, network constraints, multiple splitter types and economies of scale among others are then incorporated into a refined model and solved. Analysis of the experimental results indicates improved accuracy and lower deployment costs, at the expense of increasing computation effort considerably. Heuristic techniques are then examined to improve computational performance, including an elementary heuristic (ELEM), the Branch Contracting Algorithm (BCA) and problem decomposition. It is demonstrated that through the use of k-means clustering, the refined model can be solved in a fraction of the time while keeping deployment costs comparably low. / Thesis (MIng (Computer and Electronic Engineering))--North-West University, Potchefstroom Campus, 2013

Submitted by Fernanda Rodrigues de Lima (fernanda.rlima@ufpe.br) on 2018-09-27T19:57:29Z No. of bitstreams: 2 license_rdf: 811 bytes, checksum: e39d27027a6cc9cb039ad269a5db8e34 (MD5) DISSERTAÇÃO Stevie Bruno de Lima Souza.pdf: 2102769 bytes, checksum: 2b8bd21a084e0ae1bbc33c188cd450bd (MD5) / Approved for entry into archive by Alice Araujo (alice.caraujo@ufpe.br) on 2018-11-19T20:57:04Z (GMT) No. of bitstreams: 2 license_rdf: 811 bytes, checksum: e39d27027a6cc9cb039ad269a5db8e34 (MD5) DISSERTAÇÃO Stevie Bruno de Lima Souza.pdf: 2102769 bytes, checksum: 2b8bd21a084e0ae1bbc33c188cd450bd (MD5) / Made available in DSpace on 2018-11-19T20:57:04Z (GMT). No. of bitstreams: 2 license_rdf: 811 bytes, checksum: e39d27027a6cc9cb039ad269a5db8e34 (MD5) DISSERTAÇÃO Stevie Bruno de Lima Souza.pdf: 2102769 bytes, checksum: 2b8bd21a084e0ae1bbc33c188cd450bd (MD5) Previous issue date: 2017-08-31 / O dimensionamento de sistemas térmicos que consigam atender as demandas ener- géticas de energia elétrica (E), vapor (S) e água gelada (W) com o melhor retorno financeiro para o investidor é uma tarefa difícil, pois existem diversas configurações de equipamentos possíveis, com diversos modos de operação. A otimização desses sistemas é normalmente feita utilizando métodos que se baseiam no estudo numérico. O presente trabalho vem propor uma nova metodologia de otimização híbrida baseada na programação linear inteira mista (MILP), com foco no aumento da velocidade de otimização em relação aos métodos que utilizam o processo híbrido de programação linear e busca exaustiva (BE/PL). O procedimento de otimização busca maximizar o VPL (Valor Presente Líquido) do sistema e consiste em 5 passos: proposta do sistema genérico, pré-seleção dos equipamentos, aplicação do MILP, busca selecionada e programação linear. A fim de verificar a performance do novo método, foram realizados estudos paramétricos variando a tarifa de energia elétrica entre 40% e 200%. O novo método obteve uma redução de 95,8% no tempo de otimização se comparado ao método híbrido que utiliza a busca exaustiva (BE/PL) e obteve os mesmos resultados de potência e VPL. / The design of thermal systems that meet the energy demands of eletricity (E), steam (S) and cold water (W) with the best financial return to the investor is a difficult task, since there are several configurations of possible equipment, with different operating modes. Optimization of these systems is usually done using methods that are based on numerical study. The present work proposes a new optimization methodology based on mixed integer linear programming (MILP), focusing on the increase in the optimization speed in relation to the methods that use the hybrid process of linear programming and exhaustive search (BS/PL). The optimization procedure seeks to optimize the VPL (Net Present Value) of the system and consists of 5 steps: generic system proposal, pre-selection of equipment, application of MILP, selected search and linear programming. In order to verify the performance of the new method, parametric studies were carried out varying the electricity tariff between 40% and 200%. The new method obtained a reduction of 95.8% in the optimization time when compared to the hybrid method (BE/PL) and obtained the same results of power and NPV.