地下铀矿山多智能体安全生产系统可靠性优化控制研究

发布时间：2018-02-27 02:16

本文关键词： 系统可靠性多智能体系统地下铀矿山安全生产系统优化控制　出处：《南华大学》2016年硕士论文　论文类型：学位论文

【摘要】：地下铀矿山安全生产系统是一个涉及多因素、多层次的、动态变化的柔性系统,涉及到矿山穿孔爆破、矿井通风、氡及氡子体析出、矿井运输与提升等诸多安全生产子系统。虽然其发生机理各异,但引发事故的因素即相互关联、相互影响的。因此,针对铀矿开采作业环境的复杂性、安全生产系统的网络性、事故诱因的多样性,开展铀矿山安全生产系统的可靠性研究与安全生产系统的优化控制研究成为当前解决地下铀矿山安全系统优化控制的紧迫任务。为此应用Multi-Agent技术、可靠性理论构建地下铀矿山安全生产多智能体系统可靠性模型结构及协商机制,应用免疫遗传优化算法实现地下铀矿山安全生产多智能体系统的最优控制。本文主要的研究工作如下:(1)构建地下铀矿山安全生产多智能体系统的可靠性结构体系。通过深入了解地下铀矿山安全生产系统的生产工艺及生产特性,结合多智能体技术,将地下铀矿山安全生产系统分为人机交互界面、主控Agent、爆破Agent、采装(放矿)Agent、运提Agent、通风Agent、放射性污染物控制Agent,每个单智能体都有其自身独立的决策、计算以及通信能力,但是自身的功能又有限,需要根据其他Agent信息作出判断。分别针对每一Agent模型的功能、内部结构进行设计。(2)地下铀矿山安全生产多智能体系统协商机制研究。根据系统的内在机制(串联、并联、混联)的方式与多智能体的协商机制相结合的方法来解决矿山安全生产系统的内部协商问题。在MAS运行过程中,需要多个具有不同能力的Agent以使整个系统可靠性达到最高为目标以协商的方式来合作完成复杂的任务。(3)安全生产多智能体系统免疫遗传优化控制研究。融合智能优化算法的免疫算法和遗传算法,构建性能优良的免疫遗传优化算法,运用免疫遗传优化算法分别对多智能体结构模型进行多次迭代优化求解,以能使多智能体结构多目标系统全局优化需求为目的,解决地下铀矿山安全生产多智能体系统的全局最优控制问题。(4)铀矿山安全生产多智能体系统优化及实例分析。以某大型地下铀矿山为例进行多智能体安全生产系统的建模与集成,根据矿山生产实际,设定各智能体之间的约束条件及运作流程。应用多智能体技术及可靠性理论、免疫遗传优化算法实现该地下铀矿山安全系统的协调优化控制问题,并进行计算机程序设计与调试。综上所述,本文应用多智能体技术和可靠性理论构建的地下铀矿山安全生产多智能系统可靠性模型及免疫遗传优化控制研究,解决了复杂系统可靠性研究中建模困难、难以得出切合实际的可靠性指标以及系统全局最优控制等问题,对于实现地下铀矿山维修策略的优化、合理安排生产计划具有重要的指导作用。实例分析验证与检测这一研究成果的有效性及适用范围。
[Abstract]:The underground uranium mine safety production system is a multi-factor, multi-level, dynamic and flexible system involving mine perforation blasting, mine ventilation, radon and radon daughter exhalation. Although the mechanism of mine transportation and hoisting is different, the factors that cause accidents are interrelated and influence each other. Therefore, in view of the complexity of the working environment of uranium mining, the network of safety production system, Because of the diversity of accident inducement, it is an urgent task to study the reliability of the safety production system and the optimal control of the safety production system in uranium mines. Therefore, Multi-Agent technology is applied to solve the problem of optimal control of the safety system in underground uranium mines. Reliability theory is used to construct the reliability model and negotiation mechanism of multi-agent system in underground uranium mine safety production. The immune genetic optimization algorithm is used to realize the optimal control of the multi-agent system for the safe production of underground uranium mines. The main research work of this paper is as follows: 1) the reliability structure system of the multi-agent system for the safe production of underground uranium mines is constructed. ... Through an in-depth understanding of the production process and production characteristics of the underground uranium mine safety production system, Combined with multi-agent technology, the safety production system of underground uranium mine is divided into man-machine interface, main control agent, blasting agent, mining (drawing agent, transporting agent, ventilating agent, radioactive pollutant controlling agent). Each single agent has its own independent decision. Computing and communication capabilities, but their own functions are limited, need to be judged according to other Agent information. Respectively for the function of each Agent model, The internal structure of the underground uranium mine safety production multi-agent system negotiation mechanism. According to the internal mechanism of the system (series, parallel, In order to solve the problem of internal negotiation of mine safety production system, the method of combining the method of mixing with the negotiation mechanism of multi-agent is adopted. In the process of MAS operation, Multiple Agent with different capabilities are needed to achieve maximum reliability of the whole system. The goal is to cooperate to complete complex tasks in a negotiated manner. Research on immune genetic optimization control of multi-agent systems in safety production. Fusion intelligence. Immune algorithm and genetic algorithm for optimizing algorithm, The immune genetic optimization algorithm with good performance is constructed. The immune genetic optimization algorithm is used to solve the multi-agent structure model iteratively and iteratively. The purpose of this algorithm is to make the multi-agent structure multi-objective system global optimization. To solve the global optimal control problem of multi-agent system for underground uranium mine safety production. Optimization of multi-agent system for safety production of uranium mine and case analysis. Taking a large underground uranium mine as an example to carry out multi-agent safety production system. Modeling and integration, According to the practice of mine production, the constraint conditions and operation flow among agents are set up. Using multi-agent technology and reliability theory, the immune genetic optimization algorithm is used to realize the coordinated optimal control of the underground uranium mine safety system. To sum up, the reliability model and immune genetic optimization control of multi-intelligence system for underground uranium mine safety production are constructed by using multi-agent technology and reliability theory. It solves the difficulty of modeling in the reliability research of complex system, and it is difficult to obtain the practical reliability index and the global optimal control of the system, so as to optimize the maintenance strategy of underground uranium mine. Reasonable arrangement of production plan has important guiding function. The validity and application scope of this research result are verified and tested by case analysis.
【学位授予单位】：南华大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TD868

【参考文献】