بهینه سازی سیستم تولید و توزیع روستاهای بیوانرژی
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|2852||2012||11 صفحه PDF||سفارش دهید||9700 کلمه|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : International Journal of Production Economics, Available online 13 October 2012
In bioenergy villages, local bioenergy plants are installed to supply electricity, which is fed into the national grid, and to heat households through a local heat distribution network. In this paper, a linear mathematical model, which economically optimizes local bioenergy production and distribution systems based on a given set of system components, is presented. The model simultaneously determines the optimal capacity of the system, the objects that should be connected to the heating network and the course of the network. Additionally, a combined heat and power (CHP) biogas plant builds the production system. The problem is modeled as a mixed integer linear program (MILP) and is applied to a village with n potential heat customers. This model offers the possibility of economically assessing various scenarios concerning different planning situations and optimizing the capacity planning for the biogas plant and the course of the district heating network.
The increasing shortage of resources and climate change have led to a new energy policy in Germany during the last decade. A further development of renewable energy sources and higher energy efficiency, for example, by insulating older buildings, is expected to decrease the dependency of this country on fossil fuels and reduce the emissions of greenhouse gases (Hennicke and Bodach, 2010). To reach these goals, several different laws have been passed and include the Combined Heat and Power Act (BMU, 2002) and the Renewable Energy Act (BMU, 2000). These laws set monetary incentives to use renewable energy sources and install plants with combined heat and power generators (CHP), which should realize energy efficiencies of greater than 80% (Nowak and Arthkamp, 2010). Combined heat and power generation in local heating networks can be optimized by Mixed Integer Linear Programs (MILPs) to find the optimal operating strategy (Casisi et al., 2008). By taking into account the set-up of microturbines and the lay-out of the heating network, Casisi et al. (2008) applied their model to a real, city-center situation and demonstrated the wide scope of optimizing the operation of such systems. Biogas plants can incorporate the use of renewable energy sources, producing methane from biomass, and the combination of heat and power generation, providing the fuel for a CHP facility. However, a sufficient supply of biomass must be obtained, to run the plant effectively. Additionally, for this type of decision problem, MILPs, which model biomass locations, capacities, the logistics of transportation, and various biofuel conversion technologies, can be formulated and implemented (Kim et al., 2011). The model from this study optimizes all decisions regarding different processing plants, biomass allocation, the final products, and their transport, and considers the objective function of the overall profit. Likewise, Gronalt and Rauch (2007) proposed an evaluation method for forest fuel supply networks by comparing centralized to local approaches. This group configured wood biomass supply networks for potential heating and energy plants, and considered the overall system cost of alternative configurations. With a primary energy potential of greater than 70,000 GWh per year, which is the energy demand of approximately 3.5 million average households, biogas is a potentially important renewable energy source in Germany, especially for decentralized, local energy concepts, because it is capable of providing baseload power and heat (Vogt, 2008). Nevertheless, biogas plants have recently become the subject of substantial criticism, and issues such as the competition for land use, rising leasing rates for arable land, mono-cropping and its negative consequences on the natural scenery and biodiversity have been discussed. To mitigate and solve some of these problems, new concepts for cultivating energy crops and higher energy efficiencies of technical facilities are needed. Some first steps may involve the use of combined heat and power technologies that are supplemented by location-specific heat concepts and the use of crop rotations in the place of mono-cropping. In Germany, the first resource-efficient energy concept of this type was realized in the bioenergy village of Jühnde in 2004. Electricity and heat are produced from biogas in a combined heat and power generator. Liquid manure and crops, which are cultivated from the agricultural land around the village, are the feedstock for the generation of biogas in an anaerobic digestion plant. The resulting electricity is then fed into the national electricity grid. To transport heat to villagers, a local hot water grid is installed (Ruppert, 2008).1 In this context, the following questions on production planning and logistics arise: What is the optimal course of the heating network, which potential heat customers should be connected to this network, and what capacity for the energy station should be installed? To answer these questions, an optimization model has been developed, that is composed of the CHP biogas plant as a production system and the local pipeline network as the distribution system that supplies the heat to customers. Building legislation, physical restrictions, and other political regulations are not taken into consideration. However, the costs of installation, maintenance and operation, and the expected revenues from selling heat to local customers and feeding electricity into the national grid are incorporated into the model. The model presented in this paper can be widely adjusted and allows for the representation of many different planning situations. It can be used to support decision-making during the strategic planning of an investment in local heating systems and offers the possibility to calculate an economic benchmark for any biogas plant, which is combined with a heating network, optimizing its course and the capacity of the plant. Furthermore, the model can compare various scenarios regarding, for example, the biomass availability or the willingness of households to be connected to the local heating grid. It thereby offers valuable information for potential investors and relevant stakeholders during the process of planning the installation of such a facility. The next section presents the general set-up of a district heating system based on the biomass. In Section 3, the cost for biomass allocation is estimated. In 4 and 5, the energy production and distribution systems are explained and economically assessed. In 6 and 7, the optimization model is shown and applied to a village with n=44 potential heat customers. In Section 8, various conclusions are drawn and further steps for improving the model are discussed. The last section summarizes the results.
نتیجه گیری انگلیسی
A linear optimization model has been generated to economically optimize the production and distribution systems of bioenergy villages. Optimal solutions for the capacity of a biogas facility, for the course of the heating grid, and regarding profitable heat customers can be simultaneously generated. The willingness of people to use locally generated bioenergy was included in the optimization process. In this way, some social aspects can be taken into account. To avoid mono-cropping and crop failure, a mix of different energy crops was used, and the process of crop rotation was applied. Although the idea of crop rotation has been reflected upon, the impact of cultivating energy crops on the climate and on biodiversity must be further considered when calculating the location-specific availability of biomass. The flexibility of the model allows for a direct application to various different settings and regions. It is possible to take political and social factors into account, such as the existence of governmental grants and allowances or the refusal of a portion of the population to participate in the project. Furthermore, various aspects of different landscapes can be incorporated, such as higher investments in the heating network in hilly regions or restricted land use in the outskirts of a larger city with a higher number of potential heat customers. The model even allows for the incorporation of existing structures into the process of optimization. By adjusting the adequate binary variables and net present values of the existing facilities in the constraints, the model is able to appropriately handle these existing assets. In this way, potential expansions of the decentralized heating system can also be optimized. Therefore, the model is applicable to a wide range of planning situations and offers valuable information to all potential investment stakeholders in a district heating system that is based on the biomass. 8.1. Sensitivity analysis Because the composition and costs of the substrate for biogas plants have a major impact on the solution of the optimization model, sensitivity analyses are performed to estimate the influence of changing parameters on the NPV of biogas plants. This analysis determines the effect of different substrate costs, allowances, capacity, efficiency, and the used percentage of liquid manure on the NPV of a biogas plant with an installed power of 500 kW that began operation in the year 2010. These results are shown in Fig. 5. The original value of each parameter is interpreted as 0%, the different functions display the effect of the altered parameters on the NPV, and all other variables remain constant. Figure options The substrate costs per ton BCkW[€/t]BCkW[€/t] can be written as a function of the installed power xkW by View the MathML sourceBCkW(xkW)=36.17+0.0333·xkW[€/t]. Because the plant capacities only range from 100 kW to 2 MW, the variable substrate costs per ton for the various plants are between 0.333€/t0.333€/t and 1.489€/t1.489€/t. Hence, the percentage of variable costs of the biomass compared to the total cost per ton amounts to only 0.9% to 4%. This finding also explains the progression of the two substrate cost functions in Fig. 5. Both curves have a negative slope because the NPV of the plant decreases as the substrate costs increase. However, the function of fixed substrate costs is much steeper because it accounts for 98% of the total substrate costs. An increase in allowances leads to a significant rise in the NPV, and because the revenue of the biogas plant is calculated by multiplying the allowances per kWh by the effective electricity output of the plant, the curve in Fig. 5 shows a linear progression. The slope of this curve depends on the total revenue, and it will be steeper for larger plants that generate a higher revenue. An interesting point on this curve is its root at −9% because it indicates the political margin for a decrease in the allowances. Thus, a decrease in the allowances of 9% represents the threshold above which this plant, with a capacity of View the MathML source500kWel, would still be economically profitable. A potential increase in the electrical efficiency also has a positive effect on the NPV because it effectively increases the electricity output and, therefore, the generated revenue of a plant without increasing the operating costs. However, because the benchmark in this analysis is a View the MathML source500kWel biogas plant, the curve actually does not express the increase in power output but represents the decrease in substrate input that is needed to generate the same amount of electricity. For this reason, the progression of the function is not linear but has a positive and slightly decreasing slope. The last parameter to be examined in this analysis is the percentage of liquid manure that is fed into the fermenter of a biogas plant. In contrast to the other variables, the alteration of this factor is given in percentage points and is not represented by the percentage change of the variable. Because the original value of liquid manure is set to 30%, it is impossible to calculate net present values for a higher decrease than −30%. The curve of this parameter is broken at 0%, which can be explained by the granted bonus allowance for using 30% or more liquid manure in the substrate mixture to run the plant. Beyond this point the curve has an increasing slope due to the fact that each additional percentage point of liquid manure replaces a larger amount of biomass and because the total amount of used biomass increases. This displacement leads to lower substrate costs and a higher NPV when it is assumed that the liquid manure is allocated free of charge. 8.2. Amendment to the Renewable Energy Act in 2012 The Renewable Energy Act is being amended in 2012 to adjust the law to a shifting policy away from an unconditional support of renewable energies. The major goals that are to be achieved by the amendment include further development of renewable energies, an increase in cost efficiency, and a faster integration into the national energy-system and market structures (BMELV, 2011) by several different mechanisms. The first important point is the simplified allowance system. These allowances are mainly paid for the use of different substrates (see Table 5). The degression rate has been increased to 2% but is limited to the substrate specific allowances. Substrates in class I include several energy crops, whole crop silage and scrap wood. Class II includes liquid and other manure, straw, and biomass from landscape conservation. The basic allowance is paid for nearly all biomass substrates according to BMU (2001). The main requirement for all biogas plants to receive any allowance is a substrate mixture that contains at least 60% manure or at least 60% heat usage from the CHP in houses or stables or for internal consumption. This requirement shall increase the energy efficiency of newly installed plants and decrease the amount of methane and other greenhouse gases that are emitted into the atmosphere by conventional methods of manure disposal. To reach this goal, small plants of up to an equivalent size of View the MathML source75kWel that are run using at least 80% liquid manure have a fixed allowance of 0.25€0.25€ for every kWh; however, the possibility to combine with other allowances is not permitted. To increase biodiversity and reduce the monocropping of maize and other profitable energy crops, the mass percentage of ensilaged maize and grain that is used in the substrate mixture is limited to 60%.Another important aspect of the amendment is the incentive for the operator of a biogas plant to market the energy alone. It is possible for these individuals to sell electricity at the national energy exchange in Leipzig or through delivery contracts with private customers. If the total revenue is below the granted allowances, they will be paid the difference so that they are not subjected to worse conditions than other companies. This economic incentive also increases the willingness to establish a demand-driven production of electricity, through higher revenues for the operator. This approach for reducing the dependency on the national allowances is supported by a flexibility allowance, which is granted for biogas storage and greater generators allowing for a postponement of the production of up to 12 h. In this way, the generation of electricity can be regulated and shifted with the demand peaks of each day. These different aspects of the amendment to the Renewable Energy Act all aim for the above-mentioned goals, namely, a further development of renewable energies, higher cost efficiency, and integration into the national electricity market. To include these changes in the presented model, it must be adapted in various ways, in particular the interdependencies and the meeting of all constraints pose a challenge to the modeling. 8.3. Outlook and further research To facilitate data collection for the issue of local biomass availability, the optimization model will be connected with a Geographic Information System (GIS). One suitable GIS for the region of Lower Saxony in Germany is BioSTAR (Bauböck, 2010). BioSTAR offers the possibility to determine location-specific harvest yields based on soil types and climate data, which are available through the German Weather Service.2 By predicting potential harvest yields in certain regions, it will also be possible to analyze various scenarios concerning climate change over the next 20 years. In this way, the development of climate conditions and its effects on harvest yields and substrate costs can be taken into consideration beforehand and can be included in the decision-making process. Currently, a CHP biogas plant is modeled as the only source of heat. In future research, a second heat source and a peak-load boiler will also be considered in the optimization model. As of now, additional investment costs have been mentioned but have not been integrated into the model. By doing this, the heat supply is secured for extremely cold days of the year, and a location-specific optimal mix of different heat facilities can be calculated. In the absence of a peak-load boiler and without an additional heating station, the capacity of the biogas facility is probably too large for periods of the year with less heat demand. As a consequence, a substantial amount of heat is wasted, and a substantial amount of biomass is needed to operate at a full-load capacity throughout the year to generate enough revenue from feeding electricity into the national grid. This occurrence causes problems in terms of energy efficiency and land-use competition. However, the plants are still profitable for the operating companies, and the revenues from selling electricity exceed the opportunity costs of the wasted heat. Nevertheless, the optimization model must be improved to be able to calculate an optimal mix of base and peak load energy stations with a high overall energy efficiency. In a second step, the model can be enlarged to optimize several plants simultaneously. By taking several plants into consideration, the model optimizes the set-up of the heating system regarding installed power and the location of the plants. Location planning could be achieved using binary variables for every potential location in a discrete set, which would determine the number of plants to be built and their specific locations. Nevertheless, various problems regarding the complexity of the model must be considered. By augmenting the model to depict multiple plants, the number of constraints is also multiplied by the number of plants. Furthermore, the availability of liquid manure and other biomass must be determined, the allocation of these substrates to the various locations must be taken into consideration, and every single plant must be equipped with a suitable peak-load boiler or other heat source as a backup facility. All of these model-specific adjustments will most likely increase the runtime of the algorithm, and how the model can be scaled for regional bioenergy projects must be investigated. The mathematical implementation could pose a great challenge due to the complexity of the problem and further research will have to examine under which constraints the model is still applicable to larger problems, providing solutions in a reasonable running time.