管理工程学报
管理工程學報
관리공정학보
Journal of Industrial Engineering and Engineering Management
2013年
1期
99~107
,共null页
彭张林 濮剑鹏 张登洲 胡笑旋 张强 詹明明
彭張林 濮劍鵬 張登洲 鬍笑鏇 張彊 詹明明
팽장림 복검붕 장등주 호소선 장강 첨명명
证据推理 效能评估 多任务规划 临近空间
證據推理 效能評估 多任務規劃 臨近空間
증거추리 효능평고 다임무규화 림근공간
evidential reasoning; effectiveness evaluation ; multi-mission planning; near space system
在临近空间系统多任务规划过程中,对不同的多任务规划场景下,临近空间系统综合效能及其变化规律的研究,可以帮助决策者寻找到相对最优的临近空间多任务规划方案。由于临近空间系统涉及多资源、多任务和复杂的内外部环境,影响其效能的因子较多且动态多变,因此,其效能评估将遇到多种确定性和不确定性因素共生环境下的评价问题,传统的方法难以独立处理该问题。鉴于此,将证据推理的方法引入效能评估中,设计了基于证据推理的效能评估模型,该模型综合考虑了无知、模糊、不准确、区间数和区间信念度等多种不确定性因素,并实现了多源证据的有效合成。最后,在实现的2组面向点目标的临近空间多任务规划仿真实验中,运用该模型分别计算出了不同场景下临近空间系统综合效能,计算结果显示资源数和任务数的变化对临近空间系统综合效能的影响符合客观规律,表明该模型具有一定的可用性。
在臨近空間繫統多任務規劃過程中,對不同的多任務規劃場景下,臨近空間繫統綜閤效能及其變化規律的研究,可以幫助決策者尋找到相對最優的臨近空間多任務規劃方案。由于臨近空間繫統涉及多資源、多任務和複雜的內外部環境,影響其效能的因子較多且動態多變,因此,其效能評估將遇到多種確定性和不確定性因素共生環境下的評價問題,傳統的方法難以獨立處理該問題。鑒于此,將證據推理的方法引入效能評估中,設計瞭基于證據推理的效能評估模型,該模型綜閤攷慮瞭無知、模糊、不準確、區間數和區間信唸度等多種不確定性因素,併實現瞭多源證據的有效閤成。最後,在實現的2組麵嚮點目標的臨近空間多任務規劃倣真實驗中,運用該模型分彆計算齣瞭不同場景下臨近空間繫統綜閤效能,計算結果顯示資源數和任務數的變化對臨近空間繫統綜閤效能的影響符閤客觀規律,錶明該模型具有一定的可用性。
재림근공간계통다임무규화과정중,대불동적다임무규화장경하,림근공간계통종합효능급기변화규률적연구,가이방조결책자심조도상대최우적림근공간다임무규화방안。유우림근공간계통섭급다자원、다임무화복잡적내외부배경,영향기효능적인자교다차동태다변,인차,기효능평고장우도다충학정성화불학정성인소공생배경하적평개문제,전통적방법난이독립처리해문제。감우차,장증거추리적방법인입효능평고중,설계료기우증거추리적효능평고모형,해모형종합고필료무지、모호、불준학、구간수화구간신념도등다충불학정성인소,병실현료다원증거적유효합성。최후,재실현적2조면향점목표적림근공간다임무규화방진실험중,운용해모형분별계산출료불동장경하림근공간계통종합효능,계산결과현시자원수화임무수적변화대림근공간계통종합효능적영향부합객관규률,표명해모형구유일정적가용성。
As military and commercial values of near space exploration increase, air vehicles (e. g. hypersonic vehicles and aerostats) installed in the near space are carrying more responsibility for military missions. Near space systems may have multiple military missions at the same time, and demand multi-mission planning. In the multi-mission planning, multiple alternatives may exist in order to meet the mission demand and resource attributes. However, what is the best planning alternative? This paper intends to evaluate the effectiveness of near space systems under different planning situations and decide on a multi-mission planning solution to maximize or optimize the effectiveness of near space system evaluation. One central problem of this study is to evaluate the effectiveness of near space systems. Near space systems involve dynamic variables, such as multiple resources, multiple missions, complex internal and external conditions. Therefore, traditional assessment methods will face the problem of evaluating the effectiveness of the coexistence of various certainties and uncertainties because they often deal with each variable independently. Evidential reasoning method considers various kinds of uncertainties, such as ignorance, fuzziness, interval data, and interval belief degrees, and can implement multiple sources of evidence. As such, the method can overcome the limitation of uncertain evaluations. In contrast, the traditional method deals with complex conditions and can effectively solve problems associated with the effectiveness evaluation of near space system to perform multiple missions. In view of this, this paper constructs an effectiveness evaluation model based on the evidential reasoning method, and applies the model in planning for performing multiple missions in the near space system. This paper is composed of two parts as follows: Firstly, this paper constructs an effectiveness evaluation model based on the evidential reasoning theory. This model is constructed by following seven steps: (1) constructing a multi-stage distributed effectiveness evaluation system; (2) determining the weight of each indicator; (3) selecting the discernment frame for each leaf node and its corresponding utility and determining a consolidation frame and its utility; (4) collecting the original evaluation data, and transforming them into Basic Belief Assignment (BBA) using a predetermined formula; (5) using the transformation rule to transform the BBA of different frames to the BBA of the consolidation frame ; (6) using the evidence combination calculating method to combine data from the bottom up and collecting the final evaluation of the root node; and (7) using the maximum subordinated degree or the overall utility principle to analyze the effectiveness evaluation results. Secondly, this paper evaluates the effectiveness under two multi-mission planning situations: ( 1 ) increasing the missions while maintaining the numbers of resources and (2) increasing resources while maintaining the numbers of missions. The evaluation results show that under the condition of maintaining the numbers of resources ( 1 ) if the planning missions can be carried out, increasing missions can improve the effectiveness of the air vehicle resources, reduce the cost of resources, and promote the comprehensive effectiveness of the multi-missions in the near space system. (2) if planning missions cannot be carried out, increasing the numbers of air vehicle resources can improve the mission implementation rate and promote the effectiveness of multi-missions in the near space system. Both of these situations can help optimize the multi-mission planning effectiveness in near space systems. To sum up, the effectiveness evaluation model based on evidential reasoning is a valid method. The model can solve the problem of effectiveness evaluation in a complex system having a plurality of uncertainties. The results of applying the model to the multi-mission planning in the near space system also verify the applicability of the model.