矿井安全非线性动力学评价模型及应用研究

【作者】 施式亮；

【导师】 刘宝琛；

【作者基本信息】 中南大学 ， 安全技术及工程， 2000， 博士

【摘要】 本文首先分析了我国煤炭工业在国民经济发展中的重要基础地位，指出了煤矿安全是煤炭工业健康、可持续发展的关键问题。针对矿井安全管理及其评价中存在的问题，提出了矿井安全非线性动力学评价和预测的研究课题，并进行了研究，提出了作者独到的见解和方法，并在实际的应用中达到了预期的效果。 煤矿井下生产系统是一个由人-机-环境构成的、空间极其复杂的灾害系统。其中人工、自然因素共存，瓦斯、煤尘、水害、火灾、顶板以及机电事故等是存在的主要灾害形式。这些事故发生的机理各异，引发事故的因素却相互关联，在时间、空间上各种灾害随时随地发生，且相互影响。因此，根据煤矿井下灾害系统的结构特点，对系统的危险程度进行评价，事先获得事故的可能后果及对整个生产系统的影响，从而使矿井的技术和管理部门有针对性地采取措施，达到安全生产的目的。寻求并建立科学、合理的矿井安全评价模型，并与实际生产相结合，是矿井安全管理与控制的关键问题。 矿井安全系统中的许多问题都是非线性的，传统的、事先设定变化规律和特性的评价方法已经显现出其局限性，且难以很好地解决从因素到结果的定权和变权问题。本文在分析了矿井灾害系统所具有的非线性特征的基础上，论述了具有典型非线性动力学特征的灰色理论和人工神经网络与矿井安全性评价有机的结合适应性问题；系统地分析了矿井事故发生的机理和规律，建立了反映矿井事故特性的分析模型；在遵循可比性、科学性、统一性、适用性、拆分与合成性以及指标定量化处理等原则前提下，充分考虑了影响矿井安全的各种因素，构建了全面的矿井安全评价指标体系，使矿井安全生产过程中的各个重要影响因素在指标体系中得到体现，并配套编制了安全数据采集表和各指标分级定量化表；进行了矿井安全评价与非线性动力学模型的适应性研究，建立了非线性灰色安全评价模型和基于神经网络的非线性安全评价模型，对矿井安全状况的历史和现状进行动态评价，实现由历史到现在甚至未来的动态评价，并且解决了由安全状态参数与结果之间因素权值的智能化确定，即变权问题，实现了真正意义上的动态安全评价。实现了矿井安全状态指标数据采集的动态和实时性，改变了目前现有方法不能实现可操 中南大学博土学位论文：矿井亥全非线性动力学评价模型及应用研究 作性的状况，推动了安全评价方法的发展。本文将神经网络技术、灰色 系统理论与安全信息处理技术的结合为矿井安全管理和科学决策提供了 新的、更为有效的途径和方法。 预测就是依据历史寻求事物的未来发展趋势，是对事物未来发展趋 势的认识，目的就是根据事件的发展与变化趋势来采取相应的措施。矿 井安全的有效控制对生产和作业人员的安危具有重要意义。有效的管理 与控制，必须有完善、可靠的过程监测，而过程控制的成功与否，取决 于对煤矿安全性指标的超前把握，准确的预测是超前把握并采取有效技 术和管理措施的先决条件，矿井安全预测就是通过系统现有或过去的危 险信息来预测未来的系统安全状态。本文根据宏观与微观、静态与动态 的辨证关系，确定了矿井安全预测的基本原则，建立了非线性灰色预 测、神经网络预测数学模型。建立的灰色预测模型适应于历史数据不充 分的预测问题，而首次将神经网络应用于安全预测的数学模型擅长于解 决具有大量的历史数据的预测课题，使得殒测结果更具客观性和预见 性。 在本文理论和应用研究的同时，研制、开发了与非线性动力学评价 和预测模型相配套的计算机应用程序，包括安全评价和安全预测两大主 要功能，且在本系统程序开发中首次引入了科学计算可视化技术，实现 了安全评价和预测过程的可视化监控。更多还原

【Abstract】 In this paper, the important and basic position was analyzed about the coal industry to the national economy, and the key-point of the coal safety to the continuous development of the coal industry was pointed. According to the problems existing in the safety management and assessment, the research project of the safety assessment based on the non-linear dynamics in the coal mines was proposed and studied, and the author抯 original idea and methods was given out. At same time, the study results were applied in the practice samples, and the prospective results were attained.The production system underground in coal mine is a disaster system which consists of the human-machinery-environment and the extremely complex horizons, the main disaster types are methane, dust, water, fire, roof and machinery in it. The mechanism of the disasters is different, but the factors initiating the disasters were interrelated each other. The disasters described above may take place at any time and every where, and affected from one to others. According to the structure features of the calamity in coal mine, we have to take the assessment to the hazard degree of the system, and previously acquire the effect of the possible results of the accidents to the whole production system, so that the technology and management administrators can adopt the measures, and the aim of the safety production will be obtained. The key problem to the safety management and control is to find and establish the scientific and reasonable safety evaluation models, and to combine the models with the practice production.Many problems in the safety system in coal mine are non-linear. The traditional and the previously function-setting evaluation methods have appear their localization, and the problems of the fixing and changing weight could not also be solved perfectly. In this paper, based on the analysis to the non-linear characteristics of the disaster system in coal mine, the adaptability combining the mine safety assessment with the non-linear gray theory and the artificial neural network technology was discussed. The mechanism and disciplinarian of the disasters in the mine were by the numbers analyzed, and the model reflection the features of the accidents was established. Under the precondition and the principles following the comparison, science, oneness, applicability, synthesizing, split, and quantity of the indexes, and sufficiently thinking over the variety of the factors affecting the safety status of the mine, the system of the safety assessment indexes were completely constructed, and the everyviiimportant factor during the production processes was incarnated in the index system, at the same time, the original safety data table and the index quantity tables were designed and worked out. The non-linear safety assessment models were established based on the non-linear gray system theory and the artificial neural network, and the dynamic evaluation came true from the past to the present and the future. Furthermore, the intellectualized decision of the weights between the status and its factors was solved, and the actual dynamic assessment was realized. The dynamic and real-time safety data collection of the indexes was carried out in the mines, the not-easy maneuverability status of the traditional methods was changed, and the assessment method development was promoted. In the paper, the author combined the artificial neural network technology and the gray system theory with the safety information process, and supplied a new way and method for the safety management and scientific decision in the coal-mines. The prediction is to search the future developing trend according to the history, also is the recognition to the future developing tendency. The aim of the prediction is that the processing measures can be done according to the developing and changing trend. It is very important to control effectively the safety in coal-mine for the mine production and operators. The effective control and management relied on the perfected更多还原