Abstract 摘要

 

当计划生物质生物能源生产，规划者应该考虑到每个帐户，并沿生物质供应链的每一个利益相关者，如生物质资源供应，运输，转换和电力供应商。此外，规划者必须考虑社会的关注，以建立生物质系统和每个国家的具体困难有关的环境和经济的影响。为了克服以可持续的方式处理这些问题，需要一个强大的决策支持系统。为了这个目的，对于生物能源生产规划和实施两个层次的普通生物能源决策系统（gBEDS）的开发。该gBEDS的核心部分是信息的基础上，其包括基本的生物能源信息和详细信息作出决定。详细的决策信息认为该参数值数据库默认值，变量数据库，通过仿真和优化得到的数值。它还包括一个场景数据库，它是用于演示的新用户，也为规划者和executers案例推理。根据信息库，以下模块包括支持决策：根据本单位的过程（UP）的最优决策以及应用数据挖掘Matlab的模块定义和遗传算法（气）等方法的图形界面仿​​真模块方法（模糊C均值聚类和决策树）生物质收集点，定义基于生物能源发电的整个生命周期开发的模拟和优化模型存储和生物能源转化厂的位置。此外，Matlab中用于设置与关键生物量规划参数计算模型（例如成本，CO2排放量），在预定的生物质的供应链。基于GIS技术的接口，允许所有的生物能源转化厂和存储数据的可视化的基础上量化的输出，以支持用户的决定。随着图形界面的帮助下，用户可以方便地定义可行的生物质供应链，作出决定，并根据自己的兴趣进行评估他们，他们可能是环境，经济，社会或他人。因此，gBEDS支持在国家和地区层面生物质能规划师来评估生物质能利用和计划的选项（技术上和经济上）生物能源一代有能力和可持续的方式。 2006年保留爱思唯尔有限公司保留所有权利。When planning bioenergy production from biomass, planners should take into account each and every stakeholder along the biomass supply chains, e.g. biomass resources suppliers, transportation, conversion and electricity suppliers. Also, the planners have to consider social concerns, environmental and economical impacts related with establishing the biomass systems and the specific difficulties of each country. To overcome these problems in a sustainable manner, a robust decision support system is required. For that purpose, a two levels general Bioenergy Decision System (gBEDS) for bioenergy production planning and implementation was developed. The core part of the gBEDS is the information base, which includes the basic bioenergy information and the detailed decision information. Basic bioenergy information include, for instance, the geographical information system (GIS) database, the biomass materials’ database, the biomass logistic database and the biomass conversion database. The detailed decision information considers the parameters’ values database with their default values and the variables database, values obtained by simulation and optimization. It also includes a scenario database, which is used for demonstration to new users and also for case based reasoning by planners and executers. Based on the information base, the following modules are included to support decision making: the simulation module with graph interface based on the unit process (UP) definition and the genetic algorithms (GAs) methods for optimal decisions and the Matlab module for applying data mining methods (fuzzy C-means clustering and decision trees) to the biomass collection points, to define the location of storage and bioenergy conversion plants based on the simulation and optimization model developed of the whole life cycle of bioenergy generation. Furthermore, Matlab is used to set up a calculation model with crucial biomass planning parameters (e.g. costs, CO2 emissions), over the predefined biomass supply chains. The GIS based interface allows visualization of all bioenergy conversion plants and storage data to support the users’ decision based on quantifiable outputs. With the help of the graphical interface, the users can define easily the feasible biomass supply chains, take decisions and evaluate them according to their own interests, might they be environmental, economical, social or others. Therefore, the gBEDS supports biomass energy planners in both national and regional levels to assess the options (technically and economically) of biomass utilization and plan for bioenergy generation in a competent and sustainable way.  2006 Elsevier Ltd. All rights reserved. Keywords: Decision support systems; Geographical information system; Bioenergy; Data mining; Fuzzy C-means clustering; Biomass 

 

1. Introduction 介绍

 

Bioenergy production from biomass resources is of great importance to keep the level of CO2 emissions under control. The decisions of planning and establishing biomass utilization projects have to overcome satisfactorily wide* Corresponding author. Tel.: +81 45 924 5258; fax: +81 45 924 5270. E-mail address: nasser@pse.res.titech.ac.jp (N. Ayoub). 0196-8904/$ -see front matter  2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.enconman.2006.09.012 spread types of problems such as: economical and technical feasibilities of bioenergy production; low energy values of biomass and complicated supply chain designs [1–3].In the last two decades, many research works have been directed towards estimating biomass potential as a source of energy at all levels of evaluation, i.e. global, national and regional [4–7]. At the same time, many attempts were conducted in the form of product and process development trying to overcome the different hindrances and provide N. Ayoub et al. / Energy Conversion and Management 48 (2007) 709–723 possible promotion methods for biomass utilization [8–12] or as decision support systems that deal with a set of complicated problems to provide acceptable solutions [13–16]. Mitchell [13], reviewed a large number of decision support models and systems, but those were mostly built and designedfor supporting the technicaland economicalfeasibilities, while some others were established for environmental impacts estimations and only a few handle the social impact of biomass use. Some decision support systems are developed considering both the environmental and economical sides and benefits from the geographical information system (GIS) capabilities [17,18,14,16]. However, the previously proposed decision support systems and models in the bioenergy field are mostly constructed for one type of biomass from a specific point of view, i.e. economical, environmental or social, and focus on one decision level, either regional or national [19,20]. This exemplifies the need for a system that handles all bioenergy production stakeholders’ objectives in both national and regional decision levels and involves different possible types of biomass with effective potential for energy production. Decision support systems (DSS) are computer technology systems that can be used to support complex decision making and problem solving [21], providing the user with an accessible computer interface where the results are presented in a readily understandable form. DSS link the information processing capabilities of a management information system with modeling techniques and the judgment of managers to support decision making in unstructured situations. As the social concern is slowly moving towards seeking not only economical but also sustainable energy production methods as well as social benefits [22], a user oriented decision support system is needed to support the energy planners in meeting such changes in interests. In this respect, a general bioenergy decision system (gBEDS) with knowledge discovery capabilities, i.e. data mining has been proposed to assist in planning andimplementing bioenergy production systems. In the following paragraphs, some of the aforementioned DSSs will be analyzed. 

 

1.1. DSS for economical and environmental decisions 

The development of an economical DSS is very popular in all applied science fields to deal with the entangled set of production and marketing problems to reach an optimal or near optimal solution, e.g. production planning, risk management [23] and recently in environmental decision making [24,25]. However, little work is found in the bioenergy planning field for economical and environmental decision making. This might be explained by the low economical attractiveness of biomass. The bioenergy assessment model (BEAM), a spreadsheet based decision support system has been developed [19] for technoeconomic assessment of biomass to electricity schemes, including investigation on the interfacing issues. The model considers biomass supply from conventional forestry and short rotation coppice and has the possibility of handling different conversion routes, combustion, pyrolysis, gasification and integrated gasification combined cycle (IGCC) to electricity as well as two biological routes to ethanol. A module for investigating the collection and generation of electricity from municipal solidwaste (MSW) was also presented. Additionally, different scenarios of electricity and heat production from biomass were proposed and dynamically evaluated for economic efficiency by a mixed integer linear optimization model [26]. The model aimed at determination of the economic energy supply structure considering different kinds of users and investors to reduce biomass fuel prices.Adecision support framework for choosing efficient, environmentally sound and economically feasible greenhouse gas (GHG) emission reduction scenariosin forestry is also constructed and tested for realistic scenarios includingforest conservation, forest expansion, forest management, forest products usability andsubstitution of fossil fuels with woody biomass [20]. The system consists of three evaluation modules: the carbon assessment module for estimating emission reduction per hectare per year; the environmental assessment module for assessment of environmental impacts of carbon projects and the economic assessment module to obtain the net cost of 1ton emission reduction. 

 

1.2. DSS integrated with GIS 

The idea of using the geographical information system (GIS) in decision support systems is not new and is found in the development of many decision support systems [27– 30]. However, the integration between the GIS and the DSS technology is comparatively less extensive in the bioenergy planning field. An optimal planning decision support system of biomass as a source of energy for a consortium of municipalities in an Italian mountain region was developed [15]. That study suggests actions and policies for biomass exploitation in the region, the sizing of plants and the verification of the performance over the entire system. The authors applied mathematical optimization technologies integrated with the GIS and databases, whose information was gathered through experimental tests andinteraction with experts. Voivontas [18] presented a method for estimating the potential of bioenergy production from agriculture residues at the regional level through a GIS decision support system. This DSS provides tools to identify the geographic distribution of the economical biomass potential in four exploitation levels, i.e. theoretical, available, technological and economical, relying on the modelingcapabilities of the GIS environment. The authors used electricity production cost as the main criterion in the identification of the sites of economically exploited biomass potential. A two part method for estimating potential fuel-wood resources from forests based on forest inventory data was also presented [14]. The first part considers the assessment of fuelwood resources potential, and the second part deals with the economical availability of resources. The biomass assessment data, location of conversion facilities