應用分析網路程序法建立車用動力電池材料評選準則

Abstract

石化燃料排放溫室氣體，引發全球氣候變遷，是現代社會不容忽視的問題。相較於內燃機引擎車輛，電動汽車的能源轉換效率高、二氧化碳排放量低，已被國際間視為重要的解決方案。本研究目的在於利用多屬性的多準則評量模式，掌握經濟、節能、環境（3E）三者之衝突特性，針對電動車動力電池材料選擇規劃進行評估與探討。其中，在經濟類別中，篩選出「接口設置成本」、「材料成本」、「價格」及「聯盟對象」四項作為評判子準則；在環境類別中，篩選出「有毒物質」、「稀有物質」、「空氣汙染」、「水汙染」、「易拆解」及「使用壽命」六項作為子評判準則；在能源類別中，篩選出「石化能源替代」、「電能種類」、「續航力」及「散熱」四項作為子評判準則。為滿足不同屬性的三大主要準則與十四項次準則，本研究發展並提出「修正分析網路程序法(Modified Analytic Network Process，簡稱修正後ANP)」，利用倒推求解的網路架構，輔以正規化隨機矩陣極限收斂的數學理論基礎，以同時滿足質性與量性的評判準則，且令準則間具上下層級、自我回饋與交互作用等特性。就車用動力電池材料選擇問題的而言本研究發現主準則方面，以「環境保護」為最重要準則；次準則方面，在交互作用與回饋的影響下，分別以電能種類、接口設置成本、材料成本、水汙染與易拆解與為優先考慮準則。

The object of this research provides vehicle-used power-cell industry a framework to grasp the feature of conflict upon economic, energy and environment (3E). Electric vehicle has been regard as a solution of resources depletion and greenhouse emission problem. Power cell performs the key part component of electric vehicles. Among three main objects, interface setting cost, material cost, price and strategic alliances had refined out as the sub-criterion in the cluster of economic; Restricted substance, rare substance, air pollution, water pollution, disassembility and life time had refined out as the sub-criterion in the cluster of environment; Finally, for energy concern, substitute rate of fossil-fuel, fuel source, endurance and thermal issue are chosen as the sub-criteria. For the selection process of power cells, in order to integrate the three main objects and fourteen sub-criteria, which include both qualitative and quantitative attributes, also taking into consideration the feedback effective, interaction and hierarchy related. A Modified Analytic Network Process (Modified ANP) has been developed in this research. Modified Analytic Network Process applies backward induction and stochastic matrix as the fundamental theorem. Through the new Multi-Criteria Decision Problems (MCDP) methodology analysis, environment is found to be the top priority in main objects and fuel source, interface setting cost, material cost, water pollution and disassembility are found to be the most influential sub-criteria.