基于BP神经网络的采空区激光探测环境误差修正

来源期刊:中南大学学报(自然科学版)2020年第3期

论文作者:徐帅 侯朋远 梁瑞余 纪晓飞

文章页码:758 - 767

关键词:复杂环境采空区;三维激光探测;正交试验;BP神经网络;误差分析;误差修正

Key words:complex goaf; three-dimensional laser detection; orthogonal test; BP neural network; error analysis; error correction

摘    要:为了研究地下金属矿山高温-高湿-多尘复杂环境对采空区三维激光探测精度的影响,研发采空区复杂环境模拟装置,设计64组正交试验方案,分析温度、相对湿度和粉尘质量浓度对探测误差的影响并进行修正。为进一步量化探测误差,建立BP神经网络预测模型,并应用于某铅锌矿采空区的探测中。研究结果表明:随相对湿度升高,点云平均误差比呈“S”型曲线增长,当相对湿度介于75.0%~85.0%时,增长速率明显加快;随粉尘质量浓度增大,点云平均误差比以40 mg/m3和100 mg/m3为节点呈三段式线性增长,当粉尘质量浓度介于40~100 mg/m3时,增长速率最快,当粉尘质量浓度大于100 mg/m3时,增长速率次之,当粉尘质量浓度小于40 mg/m3时,增长速率最低;温度对点云平均误差比影响较小。模型预测值与实测值的平均相对误差为1.80%,校正决定系数为0.993;修正后的采空区边界标高、体积和顶板暴露面积等参数更加符合实际情况。

Abstract: A simulation device of environment for high-temperature, high-humidity and dusty goaf in underground metal mine was developed to study the distortion for 3D laser detection results. Multi-factor experiments, involving temperature, relative humidity, and dust concentration, were carried out for error analysis and correction based on 64 orthogonal tests. The BP neural network model was built and used for predicting detection errors. The results show that with the increase of relative humidity, the average error ratio of point cloud shows an S-shape curve increase, and the growth rate obviously accelerates as the relative humidity is between 75.0% and 85.0%. Furthermore, the error ratio also increases linearly with the increase of dust concentration. The growth rate is the largest when a dust concentration is between 40 mg/m3 and 100 mg/m3, the lowest rate occurs when the dust concentration is less than 40 mg/m3 and the modest rate occurs with the dust concentration greater than 100 mg/m3. However, the temperature has little effect on the average error ratio of point cloud. The average relative error is 1.80% and correction coefficient is 0.993. The modified parameters, such as boundary height, volume and roof exposed area, in the complex goafs of a lead-zinc mine, are more in line with the actual situation.

相关论文

  • 暂无!

相关知识点

  • 暂无!

有色金属在线官网  |   会议  |   在线投稿  |   购买纸书  |   科技图书馆

中南大学出版社 技术支持 版权声明   电话:0731-88830515 88830516   传真:0731-88710482   Email:administrator@cnnmol.com

互联网出版许可证:(署)网出证(京)字第342号   京ICP备17050991号-6      京公网安备11010802042557号