巨灾应对任务的不确定规划模型研究

发布时间：2018-06-29 12:25

本文选题：巨灾 + 情景-应对　；参考：《哈尔滨工业大学》2013年博士论文

【摘要】：有史以来，人类在生存、生活和发展的同时，必须与各类灾害威胁进行不懈地抗争，尤其是巨灾事件。巨灾事件对人类社会造成巨大的经济损失、大量人员伤亡和大规模设施受损，一直是人类面临的社会性难题。有效的巨灾应对能够减轻损失，减少人员伤亡，保护财产。因此，巨灾应对是政界、学界以及社会各界关注的问题，期望巨灾应对研究工作能够取得实质性进展。巨灾事件是“小概率大影响”事件，较少发生，一旦爆发即造成灾难性的破坏。巨灾事件极少出现或者不会出现预兆，灾情报告时间和应对决策时间极短；巨灾应对缺乏历史案例参考，预备行动方案策划适应性不足，常规应急决策机制不适用于巨灾应对任务；巨灾应对面临极度不确定性，巨灾应对人员必须在不确定环境下进行应急决策；现有应对资源、保障能力和应对人员不能满足应对需求。因此，巨灾应对组织队伍面临了巨大挑战。本文的主要研究对象是自然巨灾事件，主要研究内容是巨灾事件爆发后应急响应阶段的应对任务规划问题。在对国内外现有研究成果梳理和总结的基础上，本文研究了巨灾应对任务分解问题，在巨灾应对的领域任务和基层任务的框架下，借鉴数学领域的不确定理论及不确定规划理论，讨论了巨灾应对任务的不确定属性；采用价值函数度量巨灾应对任务目标；在此基础上，建立了巨灾应对领域任务的不确定规划模型和基层任务的不确定规划模型，开发了不确定双层任务规划模型。针对巨灾应对任务中应急资源需求缺口大、应急保障能力严重不足的特征，本文采用过载规划模型予以解决。借鉴军事资源规划领域的过载规划研究成果，本文建立了过载规划的数学模型，将其推广到巨灾应对的不确定领域，形成过载规划的不确定规划模型。首先，建立巨灾应对任务体系。本文借鉴了国内外的灾害应对任务体系，构建了巨灾应对的双层任务体系，包含领域任务层面和基层任务层面。领域任务是针对上层或应对指挥部的应对任务，指上层的、中观的、行业的应对任务，主要内容包括统筹策划任务、资源调配任务和执行协调任务等；基层任务是针对基层应对执行人员的应对任务，指底层的、现场的、一线的应对任务，主要内容包括紧急拯救任务、事态控制任务和民生恢复任务。本文还讨论了各应对任务之间的逻辑关系和时间承接关系。其次，构造巨灾应对任务的价值函数。研究了巨灾应对任务的不确定属性，根据不确定变量生成方法，建立了巨灾应对任务的不确定变量。针对其不确定特性，构造了巨灾应对任务的价值函数，并以极大价值函数作为任务规划的目标。本文选取社会各界对巨灾应急资源和保障能力的最低期望水平作为巨灾应对任务的价值函数的参照点，依此定义了巨灾应对的“收益”和“损失”；为了消除量纲的影响，定义了巨灾应对任务的达成率，将其代入前景理论的价值函数即得到巨灾应对任务的价值函数。再次，建立巨灾应对任务规划模型。根据不确定规划理论和过载规划模型的建模机理，建立了巨灾应对任务规划的不确定规划模型。统筹策划规划模型、资源调配规划模型和运输支持任务规划模型，构成了巨灾应对的领域任务规划模型；本文以电力恢复任务规划模型为例，建立了巨灾应对的临时供电任务规划模型、电力枢纽抢修任务规划模型和电力网络抢修任务规划模型，作为巨灾应对的基层任务规划模型的范例。本文采用了过程集成方法，将巨灾应对的领域任务规划模型和基层任务规划模型集成为巨灾应对的双层任务规划模型，还讨论双层任务规划模型的Stachelberg-Nash均衡解的算法。最后，本文以巨灾事件中大面积电网受损为依赖情景，尝试设计一个巨灾应对任务规划的用例，开展相关试算，以验证本文所提出的巨灾应对不确定规划模型的理论方法和使用过程，证实模型的有效性和实用性。
[Abstract]:At the same time, human beings, while living, living and developing, must struggle unremittingly with all kinds of disaster threats, especially catastrophic events. Catastrophe events have caused huge economic losses to human society, a large number of casualties and damage to large-scale facilities. It has always been a social problem facing human beings. Effective catastrophe response can reduce the loss. Loss, reduction of casualties and protection of property. Therefore, the response to catastrophe is a matter of concern to the political, academic and social circles, and it is expected that substantial progress can be made in the study of catastrophe.
Catastrophe event is a "small probability and big impact" event, less occurrence, once the outbreak is the damage of disaster. Catastrophe events rarely appear or do not appear omen, disaster reporting time and decision time are very short; catastrophe response to the lack of historical case reference, preparation action scheme planning adaptability insufficient, conventional emergency decision-making mechanism It is not suitable for the catastrophe response task; the catastrophe response faces extreme uncertainty, and the catastrophe responders must make emergency decisions in the uncertain environment; the existing coping resources, the security ability and the response personnel cannot meet the needs. Therefore, the contingency of the catastrophe response organization faces great challenges.
The main research object of this paper is natural catastrophe. The main research content is the planning of coping tasks in the emergency response stage after catastrophic events.
On the basis of combing and summarizing the existing research results at home and abroad, this paper studies the problem of catastrophe response task decomposition. Under the framework of the field task and basic task of the catastrophe response, we use the uncertainty theory of mathematics and the uncertain planning theory to discuss the uncertainty attribute of the catastrophe response task, and use the value function measure. On the basis of the catastrophe, the uncertain planning model and the uncertain planning model of the task in the catastrophe response field are set up, and the uncertain two-layer task planning model is developed.
In view of the large demand gap of emergency resources and serious shortage of emergency support ability in the catastrophe response task, this paper adopts the overload planning model to solve the problem. In this paper, a mathematical model of overload planning is established for reference of the research results of overload planning in the field of military resources planning, which is extended to the uncertain field of catastrophe response and forms the overload. An uncertain programming model for planning.
First, the catastrophe response task system is set up. This paper draws on the disaster response task system at home and abroad, and constructs a double task system of catastrophe response, including the field task level and the basic task level. The field task is aimed at the upper level or the response task of the command department, which refers to the upper level, the middle view, the industry response task and the main content. It includes planning tasks, resource allocation tasks and implementation coordination tasks, and so on. Grass-roots task is to respond to the task of executive personnel at the grass-roots level, refers to the underlying, on-site, and front-line coping tasks. The main contents include emergency rescue task, state of affairs control task and people's livelihood restoration task. This paper also discusses the logic between the various tasks. The relationship between the edits and the time.
Secondly, the value function of the catastrophe response task is constructed. The uncertain attributes of the catastrophe response task are studied. According to the uncertainty variable generation method, the uncertain variables of the catastrophe response task are established. The value function of the catastrophe response task is constructed according to its uncertainty characteristics, and the maximum value function is used as the goal of the task planning. In order to eliminate the influence of the catastrophe, this paper defines the "income" and "loss" of catastrophe response by selecting the minimum expectation level of catastrophe emergency resources and security ability as the reference point of the value function of the catastrophe response task, and defines the rate of the catastrophe response to eliminate the dimensionless influence, and substituting it into the value function of the prospect theory. The value function of a catastrophe response task.
Thirdly, the catastrophe response task planning model is set up. According to the modeling mechanism of the uncertain planning theory and the overload planning model, the uncertain planning model of catastrophe response planning is established. The planning model of planning, the planning model of resource allocation and the transportation support task planning model constitute the domain task planning model of the catastrophe response. Taking the power recovery task planning model as an example, this paper sets up a temporary power supply task planning model for catastrophic response, the task planning model of the electric power hub and the task planning model of the power network, as an example of the grass-roots task planning model of the catastrophe response. This paper adopts the process integration method to deal with the field task of catastrophe response. The planning model and the grass-roots task planning model are integrated into the bilevel task planning model for catastrophic response, and the Stachelberg-Nash equilibrium solution of the bilevel task planning model is also discussed.
In the end, this paper tries to design a case of catastrophe response task planning with the damage of large area power grid in catastrophic event as the dependent scenario, and carries out the related trial calculation to verify the theoretical method and use process of the catastrophe response uncertainty planning model proposed in this paper, and verify the validity and practicability of the model.
【学位授予单位】：哈尔滨工业大学
【学位级别】：博士
【学位授予年份】：2013
【分类号】：X4

【参考文献】