煤矿物联网环境下的瓦斯爆炸灾害预测预警模型研究

发布时间：2017-12-27 06:08

本文关键词：煤矿物联网环境下的瓦斯爆炸灾害预测预警模型研究　出处：《河北工程大学》2017年硕士论文　论文类型：学位论文

【摘要】：瓦斯灾害预测预警技术是防止煤矿重、特大事故发生,改善煤矿安全生产问题的有效方法。煤矿物联网系统是将物联网技术应用于煤矿生产中,是煤矿井下多类型数据实现灵活、动态、实时采集,可有效支撑应用层信息处理与识别技术的系统。煤矿物联网的应用可实现煤矿生产信息实时化、控制智能化、预测预警精准化的目标,有效降低我国煤矿事故死亡人数,全面提高煤矿生产安全。论文对煤矿物联网环境下的瓦斯爆炸灾害预测、预警技术展开深入研究。主要研究内容如下:首先,阐述了煤矿物联网的技术现状、系统应用现状和瓦斯爆炸灾害预测、预警技术研究现状。煤矿物联网的应用可解决传统瓦斯监测系统中存在的数据类型单一、实时性差的问题,实现对煤矿井下多类型数据的实时采集。论文较为详细地介绍了煤矿物联网在山不拉煤矿中的应用及煤矿物联网信息处理与识别技术。其次,对瓦斯爆炸灾害预测技术展开研究。论文建立了基于PSO-MEA混合优化算法的BP神经网络瓦斯爆炸灾害预测模型。采用提出的PSO-MEA混合优化算法优化BP神经网络模型的权值和阈值。其结合了粒子群优化算法(Paticle Swarm Optimization,PSO)和思维进化算法(Mind Evolutionary Algorithm,MEA)的优点,提高了模型的准确度。进而,对瓦斯爆炸灾害预警技术展开研究。论文建立了一种具有非线性特征的模糊综合评价模型,实现了对瓦斯爆炸灾害的危险等级评价,形成了瓦斯爆炸灾害的预警模型。该模型可实现危险等级的准确评价,提高了评价精度,并且具有适用范围广,充分提取、利用已有信息,信息损失小的特点。最后,采用山不拉煤矿的实时监测数据分别对瓦斯爆炸灾害预测模型和预警模型进行了MATLAB仿真分析。通过与PSO-BP和MEA-BP两种模型的对比分析,本文所建立的瓦斯爆炸灾害预测模型的最大输出平均误差为0.0174,最小输出平均误差为0.0078,最小输出精度提高了3.59%和3.27%。预警模型较现有预警模型实现了更为准确、精细的评价。
[Abstract]:The prediction and early warning technology of gas disaster is an effective method to prevent the occurrence of heavy and large accidents in coal mines and improve the safety of coal mine production. Coal mine Internet of things system is the application of Internet of things technology in coal mine production. It is a flexible, dynamic and real-time acquisition of multi type data in coal mine, and it can effectively support the application layer information processing and recognition technology. The application of coal mine Internet of things can achieve the goal of real-time production, intelligent control, precise prediction and early warning, effectively reduce the number of deaths in coal mines and improve the safety of coal mine production. This paper studies the prediction and early warning technology of gas explosion disaster under the environment of coal mineral networking. The main contents are as follows: first, the present situation of the technology, the application status of the system, the prediction of the gas explosion disaster and the research status of the early warning technology are expounded. The application of coal mine Internet of things can solve the problem of single data type and poor real-time performance in traditional gas monitoring system, and achieve real-time collection of multi type data in coal mine. The paper introduces the application of the coal mine network in the coal mine and the technology of information processing and recognition in the coal mine. Secondly, the research on the prediction technology of gas explosion disaster is carried out. In this paper, a BP neural network gas explosion disaster prediction model based on PSO-MEA hybrid optimization algorithm is established. The proposed PSO-MEA hybrid optimization algorithm is used to optimize the weights and thresholds of the BP neural network model. It combines the advantages of Paticle Swarm Optimization (PSO) and Mind Evolutionary Algorithm (MEA), and improves the accuracy of the model. Then, the early warning technology of gas explosion disaster is studied. In this paper, a fuzzy comprehensive evaluation model with nonlinear characteristics is established, which realizes the risk rank evaluation of gas explosion hazard and forms an early warning model for gas explosion disaster. This model can achieve accurate evaluation of risk level, improve the accuracy of evaluation, and has a wide range of application, fully extract and utilize existing information, and the information loss is small. Finally, the MATLAB simulation analysis of the gas explosion disaster prediction model and early warning model is carried out by using the real-time monitoring data of the coal mine. Compared with the two models of PSO-BP and MEA-BP, the maximum output average error of the gas explosion prediction model established by this paper is 0.0174, the minimum output mean error is 0.0078, and the minimum output accuracy is increased by 3.59% and 3.27%. The early warning model is more accurate and precise than the existing early warning model.
【学位授予单位】：河北工程大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP391.44;TN929.5;TD712.7

【相似文献】