采煤机煤岩界面多传感融合的偏好动态识别研究

Unmanned mechanized mining is an urgent problem to be solved in coal mining, the dynamic identification of shearer coal-rock interface, which is based on multi-sensor information fusion and preference decision, it is the key technical of unmanned mechanized mining. In this project, the multiple methods are used, which combine similarity simulation experiment, the theory of elastoplastic for rock, numerical simulation and comprehensive analysis, Acoustic emission time series analysis, spatial distribution, amplitude distribution, it is the establishment of coal cutting sound information database and fuzzy rule set and feature library; it use EDEM discrete element software to simulation analysis of 3 typical coal rock medium and 7 kinds of cutting ratio under cutting response law;The minimum fuzzy entropy theory to establish the optimal cutting characteristic information of the section membership function model, and analysis of each evidence uncertainty; study on the identify preference of cutting signals for different characteristics of coal and rock, analyze the dynamic relationship between coal-rock medium and the weight of cutting preference, the coupled model of coal-rock preference, signal recognition and fusion decision is established, building the architecture for overall perception and emotional decision fusion, which are based on the physical characteristics and mechanical characteristics of coal-rock cutting, to elucidate the theory and mechanism of dynamic identification according to the cutting preference of coal-rock, it has an important theoretical value and significance to advance unmanned mining technology and disaster preventing.

井下无人机器化采煤是煤炭开采亟需解决的难题，采煤机煤岩界面多传感信息融合-偏好动态识别是实现井下无人化开采的关键。项目采用相似模拟实验，分析截割煤岩过程中声发射时间序列、空间分布、振幅分布规律，建立煤岩截割声信息模糊集及特征数据库与规则库；采用EDEM离散元软件模拟，分析3种典型煤岩介质、7种截割比例下截割响应规律，分析不同参数对截割阻力、截割振动、电参数信号的影响；采用最小模糊熵理论建立截割特征信息的最优隶属度函数模型，并分析各证据体的不确定性；采用数值计算、工作面验证和综合分析相结合方法，研究不同煤岩介质截割信号的识别偏好，分析煤岩介质与截割偏好的权重动态关系，建立煤岩偏好-信号识辨-融合决策耦合模型，构建基于煤岩截割物理特征和截割机械特征的全面感知与情感融合决策的体系，阐明依据煤岩截割偏好条件的煤岩界面动态识别理论与机制，对推进井下无人化采掘和灾害防治具有重要理论价值和意义。

井下无人机器化采煤是煤炭开采亟需解决的难题，采煤机煤岩界面多传感信息融合-偏好动态识别是实现井下无人化开采的关键。项目测试和提取不同煤岩比例条件下截齿截割的红外热图像,分析截齿表面温度场分布规律与截割闪温特征,得到不同煤岩比例条件下截齿截割温度特征数据库,建立基于最小模糊度优化的采煤机煤岩界面动态识别模型。研究发现:截齿在截割不同比例煤岩试件时,齿尖均产生突兀的瞬时闪温区,岩石比例越大,其闪温值越高；分析3种典型煤岩介质、7种截割比例下截割响应规律，分析不同参数对截割力、截割振动、电参数信号的影响，研究发现:煤、岩两种物质在相同的激励下的温度场特性有明显差异，当存在煤岩混合情况时，检测区域内的温差范围和标准方差会明显增大，可以以此作为煤岩混合的一种预测识别依据；对煤岩试件进行主动激励时，检测区域前10min内温升变化最为明显，温差范围和标准方差增大较快，10~30min内温度缓慢上升，温差范围和标准方差缓慢增大，30min后温度相对稳定，温差范围和标准方差无明显变化；采用最小模糊熵理论建立截割特征信息的最优隶属度函数模型，并分析各证据体的不确定性；研究不同煤岩介质截割信号的识别偏好，分析煤岩介质与截割偏好的权重动态关系，建立煤岩偏好-信号识辨-融合决策耦合模型，利用灰度处理的方式采集煤炭、岩石与矸石的图像特征，并通过其图像特征对其进行详细的划分，构建基于煤岩截割物理特征和截割机械特征的全面感知与情感融合决策的体系，阐明依据煤岩截割偏好条件的煤岩界面动态识别理论与机制，对推进井下无人化采掘和灾害防治具有重要理论价值和意义。