用于矿井通风网络解算的通风机风压性能曲线自动识别方法

吴奉亮; 寇露

doi:10.13272/j.issn.1671-251x.2023100036

用于矿井通风网络解算的通风机风压性能曲线自动识别方法

doi: 10.13272/j.issn.1671-251x.2023100036

吴奉亮,
寇露

西安科技大学安全科学与工程学院，陕西西安　710054

基金项目: 国家自然科学基金资助项目（51974232）。

详细信息

作者简介:
吴奉亮（1977—），男，四川新都人，教授，博士，主要研究方向为矿井通风与安全，E-mail：15038537@qq.com

中图分类号: TD724
计量
- 文章访问数: 93
- HTML全文浏览量: 31
- PDF下载量: 15
- 被引次数: 0
出版历程
- 收稿日期: 2023-10-12
- 修回日期: 2024-03-15
- 网络出版日期: 2024-05-10

Automatic recognition method of ventilator wind pressure performance curve for mine ventilation network calculation

College of Safety Science and Engineering, Xi'an University of Science and Technology, Xi'an 710054, China

摘要

摘要: 从通风机性能曲线图像中采样并识别风压性能曲线，进而拟合出风压性能函数是矿井通风网络解算的关键技术。目前常用人工方式识别风压性能曲线，效率低且准确性不高。提出一种基于图像处理技术的通风机风压性能曲线自动识别方法。采用双边滤波、图像锐化和二值化技术对原始通风机风压性能曲线图像进行预处理，以提高图像质量。分别基于腐蚀算法、轮廓检测算法提取通风机性能曲线图像中的网格线和坐标文字，采用逻辑运算、中值滤波、轮廓检测和K3M算法提取风压性能曲线。以逐行像素识别方式识别风压性能曲线的像素坐标，采用模板匹配算法识别坐标数字，进而完成像素坐标到物理坐标的转换，实现风压性能曲线识别。将通风机风压性能曲线自动识别方法集成至通风网络解算软件，对通风机风压性能曲线进行识别试验，结果表明，该方法对单条风压性能曲线的采样速度为24 Samples/s，识别的风压性能曲线与原始曲线的重合度高，风压拟合值与原始值的最大误差仅为0.88%，较人工识别方法大大提高了通风网络解算效率和准确性。
- 矿井通风网络 /
- 通风网络解算 /
- 通风机 /
- 风压性能曲线 /
- 图像自动识别 /
- 风压性能函数拟合
Abstract: Sampling and recognizing the wind pressure performance curve from the wind performance curve image, and then fitting the wind pressure performance function, is a key technology for solving the mine ventilation network. Currently, manual methods are commonly used to recognize wind pressure performance curves, which have low efficiency and poor accuracy. This study proposes an automatic recognition method for the wind pressure performance curve of ventilator based on image processing technology. The method uses bilateral filtering, image sharpening, and binarization techniques to preprocess the original ventilator wind pressure performance curve image, in order to improve image quality. The method extracts grid lines and coordinate text from the performance curve image of the ventilator based on corrosion algorithm and contour detection algorithm. The method uses logical operation, median filtering, contour detection, and K3M algorithm to extract the wind pressure performance curve. The pixel coordinates of the wind pressure performance curve are recognized using a row by row pixel recognition method. The method uses template matching algorithm to recognize coordinate numbers, and then complete the conversion from pixel coordinates to physical coordinates, achieving wind pressure performance curve recognition. The automatic recognition method for the wind pressure performance curve of the ventilator is integrated into the ventilation network calculation software. The recognition experiment is conducted on the wind pressure performance curve of the ventilator. The results show that the sampling speed of the method for a wind pressure performance curve is 24 Samples/s. The recognized wind pressure performance curve has a high overlap with the original curve. The maximum error between the wind pressure fitting value and the original value is only 0.88%. Compared to manual recognition methods, the method greatly improves the efficiency and accuracy of the ventilation network calculation .
- mine ventilation network /
- ventilation network calculation /
- ventilator /
- wind pressure performance curve /
- automatic image recognition /
- wind pressure performance function fitting

HTML全文

图 1 矿井通风机性能曲线图像

Figure 1. Performance curve image of mine ventilator

下载: 全尺寸图片幻灯片

图 2 简化矿井通风网络

Figure 2. Simplified mine ventilation network

下载: 全尺寸图片幻灯片

图 3 不同滤波效果对比

Figure 3. Comparison of different filtering effects

下载: 全尺寸图片幻灯片

图 4 通风机性能曲线图像预处理结果

Figure 4. Preprocessed result of ventilator performance curve image

下载: 全尺寸图片幻灯片

图 5 通风机性能曲线图像分割

Figure 5. Segmentation of ventilator performance curve image

下载: 全尺寸图片幻灯片

图 6 网格线提取过程

Figure 6. Process of extracting grid lines

下载: 全尺寸图片幻灯片

图 7 坐标数字提取过程

Figure 7. Process of extracting coordinate numbers

下载: 全尺寸图片幻灯片

图 8 曲线提取过程

Figure 8. Process of extracting curves

下载: 全尺寸图片幻灯片

图 9 图像细化结果

Figure 9. Image after skeletonization

下载: 全尺寸图片幻灯片

图 10 像素坐标系中识别出的曲线段

Figure 10. Curve segments recognized in pixel coordinate system

下载: 全尺寸图片幻灯片

图 11 模板图像

Figure 11. Template image

下载: 全尺寸图片幻灯片

图 12 待匹配数字8图像

Figure 12. The image of the number 8 to be matched

下载: 全尺寸图片幻灯片

图 13 程序界面

Figure 13. User interface

下载: 全尺寸图片幻灯片

图 14 通风机风压性能曲线识别结果

Figure 14. Recognition results of ventilator wind-pressure performance curve

下载: 全尺寸图片幻灯片

图 15 通风网络解算工况点对比

Figure 15. Comparison of operating points in ventilation network solution

下载: 全尺寸图片幻灯片

表 1 通风机风压性能曲线识别使用的OpenCV库函数

Table 1. The OpenCV functions used in recognizing ventilator wind-pressure performance curve

函数名称	作用	函数名称	作用
cvtColor（）	颜色模型转换	erode（）	图像腐蚀
threshold（）	图像二值化	dilate（）	图像膨胀
resize（）	图像尺寸调整	findContours（）	轮廓检测
rectangle（）	矩形绘制	drawContours（）	轮廓绘制
mediaBlur（）	中值滤波	contourArea（）	轮廓面积计算
bilateralFilter（）	双边滤波	matchTemplate（）	图像模板匹配
bitwise_or（）	逻辑“或”运算	boundingRect（）	轮廓外接矩形
bitwise_xor（）	逻辑“异或”运算	getStructuringElement（）	结构元素生成

下载: 导出CSV

表 2 风压性能曲线拟合函数

Table 2. The fitting function for wind-pressure performance curve

叶片角/（°）	拟合函数
叶片角/（°）	人工识别	自动识别
33/30	4666.9−33.29365Q+ 0.47619Q²−0.0037Q³	7500.8−125.3284Q+ 1.43854Q²−0.007Q³
36/33	−1131.7+134.32482Q− 1.0688Q²+0.00161Q³	2693.3+30.13515Q− 0.14524Q²−0.0011Q³
39/36	1304.9+63.9791Q− 0.39354Q²−0.00013Q³	−1120.4+123.62072Q− 0.85917Q²+0.00102Q³
42/39	9744.5−108.19048Q+ 0.75681Q²−0.00235Q³	11328.1−139.91739Q+ 0.96312Q²−0.0028Q³
45/42	−3181.9+156.33247Q− 0.98338Q²+0.00151Q³	5439.6−7.95758Q+ 0.03772Q²−0.00057Q³

下载: 导出CSV

表 3 人工识别与自动识别曲线上工况点对比

Table 3. Comparison between manual and automatic recognition of operating points on curves

叶片角/（°）	风量/ （m³·s⁻¹）	风压/Pa			风压识别误差/%
叶片角/（°）	风量/ （m³·s⁻¹）	原图像	人工识别	自动识别	人工识别	自动识别
33/30	75	3 230	3 287.5	3 239.8	1.78	0.30
33/30	85	2 900	3 005.2	2 942.5	2.92	0.77
36/33	100	3 150	3 005.2	3 154.4	2.31	0.14
36/33	140	650	782.0	655.7	8.87	0.27
39/36	105	3 600	3 533.4	3 579.8	1.85	0.88
39/36	155	1 240	1 282.8	1 234.8	6.90	0.20
42/39	155	2 350	2 294.5	2 353.0	2.39	0.13
42/39	180	1 020	1 085.0	1 018.5	6.44	0.15
45/42	160	2 800	2 841.7	2 797.0	1.49	0.10
45/42	200	800	829.3	796.7	3.67	0.41

下载: 导出CSV

参考文献(21)

[1]	李伟宏,魏志丹. 矿井智能通风控制系统研究及应用[J]. 工矿自动化,2021,47(增刊1):72-74,84. LI Weihong,WEI Zhidan. Research of mine intelligent ventilation control system and its application[J]. Industry and Mine Automation,2021,47(S1):72-74,84.
[2]	王凯,裴晓东,杨涛,等. 矿井智能通风联动调控理论与供需匹配实验研究[J]. 工程科学学报,2023,45(7):1214-1224. WANG Kai,PEI Xiaodong,YANG Tao,et al. Study on intelligent ventilation linkage control theory and supply-demand matching experiment in mines[J]. Chinese Journal of Engineering,2023,45(7):1214-1224.
[3]	任子晖,李昂,吴新忠,等. 矿井通风网络风量智能调控研究[J]. 工矿自动化,2022,48(11):110-118. REN Zihui,LI Ang,WU Xinzhong,et al. Research on intelligent control of air volume of mine ventilation network[J]. Journal of Mine Automation,2022,48(11):110-118.
[4]	范京道,李川,闫振国. 融合5G技术生态的智能煤矿总体架构及核心场景[J]. 煤炭学报,2020,45(6):1949-1958. FAN Jingdao,LI Chuan,YAN Zhenguo. Overall architecture and core scenario of a smart coal mine in-corporating 5G technology ecology[J]. Journal of China Coal Society,2020,45(6):1949-1958.
[5]	谈国文. 复杂矿井通风网络可视化动态解算及预警技术[J]. 工矿自动化,2020,46(2):6-11. TAN Guowen. Visualized dynamic solution and early warning technology for ventilation network of complex mine[J]. Industry and Mine Automation,2020,46(2):6-11.
[6]	周福宝,辛海会,魏连江,等. 矿井智能通风理论与技术研究进展[J]. 煤炭科学技术,2023,51(1):313-328. ZHOU Fubao,XIN Haihui,WEI Lianjiang,et al. Research progress of mine intelligent ventilation theory and technology[J]. Coal Science and Technology,2023,51(1):313-328.
[7]	张浪,刘彦青. 矿井智能通风与关键技术研究[J]. 煤炭科学技术,2024,52(1):178-195. ZHANG Lang,LIU Yanqing. Research on technology of key steps of intelligent ventilation in mines[J]. Coal Science and Technology,2024,52(1):178-195.
[8]	刘剑. 矿井智能通风关键科学技术问题综述[J]. 煤矿安全,2020,51(10):108-111,117. LIU Jian. Overview on key scientific and technical issues of mine intelligent ventilation[J]. Safety in Coal Mines,2020,51(10):108-111,117.
[9]	魏连江,周福宝,夏同强,等. 矿井智能通风与灾变应急决策平台[J]. 中国安全科学学报,2022,32(9):158-167. WEI Lianjiang,ZHOU Fubao,XIA Tongqiang,et al. Mine intelligent ventilation and disaster emergency decision platform[J]. China Safety Science Journal,2022,32(9):158-167.
[10]	唐辉雄,赵伏军,张柏,等. 基于最小一乘准则求解矿井通风机性能曲线[J]. 中国安全生产科学技术,2014,10(10):16-21. TANG Huixiong,ZHAO Fujun,ZHANG Bai,et al. Solution on performance curves of mine ventilator based on least absolute deviation criteria[J]. Journal of Safety Science and Technology,2014,10(10):16-21.
[11]	尹昌胜,刘剑,邓立军. 正反相切抛物线方法拟合通风机特性曲线[J]. 辽宁工程技术大学学报(自然科学版),2015,34(1):77-81. YIN Changsheng,LIU Jian,DENG Lijun. Characteristic curve of fan fitted with two positive and negative tangent parabolic curve method[J]. Journal of Liaoning Technical University(Natural Science),2015,34(1):77-81.
[12]	叶敏,杨应迪. 轴流式风机特性曲线的图形数字化及优化研究[J]. 煤炭工程,2013,45(增刊1):110-111,115. YE Min,YANG Yingdi. The digitalization and optimization study of characteristic curves for axial flow fans[J]. Coal Engineering,2013,45(S1):110-111,115.
[13]	贾腾, 王海桥, 陈世强, 等. 矿用通风机性能曲线图的函数化再现[J]. 矿业工程研究,2014,29(2):70-74. JIA Teng, WANG Haiqiao, CHEN Shiqiang, et al. A functional representation of the mine ventilator's performance chart[J]. Mineral Engineering Research,2014,29(2):70-74.
[14]	吴奉亮. 矿井主要通风机性能曲线的最优多项式拟合[J]. 西安科技大学学报,2011,31(1):14-18. WU Fengliang. Optimal polynomial fitting for mine main fan characteristic curves[J]. Journal of Xi'an University of Science and Technology,2011,31(1):14-18.
[15]	RONGDI L. An optimal algorithm for bilateral filtering[J]. Journal of Physics: Conference Series,2023,2634(1). DOI: 10.1088/1742-6596/2634/1/012030.
[16]	CAPOBIANCO G, CERRONE C, PLACIDO D A, et al. Image convolution: a linear programming approach for filters design[J]. Soft Computing,2021,25(14):1-16.
[17]	ZHANG Weiguo,SHI Dan,YANG Xiaoqiang. An improved edge detection algorithm based on mathematical morphology and directional wavelet transform[C]. The 8th International Congress on Image and Signal Processing,Shenyang,2015:335-339.
[18]	LIU Hongquan,ZHANG Weijin,WANG Fushun,et al. Application of an improved watershed algorithm based on distance map reconstruction in bean image segmentation[J]. Heliyon,2023,9(4). DOI:10.1016/J.HELIYON.2023.E15097.
[19]	TABEDZKI M,SAEED K,SZCZEPAńSKI A. A modified K3M thinning algorithm[J]. International Journal of Applied Mathematics and Computer Science,2016,26(2):439-450. doi: 10.1515/amcs-2016-0031
[20]	CUI Zhongjie,QI Wenfa,LIU Yuxin. A fast image template matching algorithm based on normalized cross correlation[J]. Journal of Physics:Conference Series,2020,1693 (1). DOI: 10.1088/1742-6596/1693/1/012163.
[21]	吴奉亮,周澎,李晖,等. 基于智能对象的矿井通风CAD模型研究[J]. 煤炭科学技术,2009,37(5):54-57. WU Fengliang,ZHOU Peng,LI Hui,et al. Research on mine ventilation CAD model based on intelligent object[J]. Coal Science and Technology,2009,37(5):54-57.