A preference recognition method for coal-rock interface fusion of multi-sensor information based on hardness characteristics

Abstract

The invention discloses a coal-rock interface multi-sensing information fusion preference identification method based on hardness characteristics, comprising the following steps: classifying the coal and rock based on the hardness, pouring the classified coal-rock to obtain the coal-rock to be tested; The sensor signal of the coal machine, and feature extraction is performed to construct a signal feature database; the weight preference value of the sensor signal is obtained by performing data fitting and weight distribution on the feature database; based on CNN neural network and D‑S evidence theory Weight optimization is performed on the weight preference value to obtain a target weight coefficient, and based on the target weight coefficient, the coal-rock interface of the coal rock to be tested is identified. The present invention significantly improves coal rock identification accuracy and cutting efficiency by studying the preference characteristics of each signal under different hardnesses and different cutting ratios, and at the same time, performing double weight optimization on the obtained weight coefficients.

Description

technical field [0001] The invention belongs to the field of coal-rock interface trajectory identification, and in particular relates to a coal-rock interface multi-sensing information fusion preference identification method based on hardness characteristics. Background technique [0002] Trajectory identification of coal-rock interface is a high-tech and difficult subject recognized at home and abroad, and it is also the key and prerequisite for realizing automatic height adjustment and automation of fully mechanized mining face. In the process of coal mining, once the cut rock is mixed into the coal seam, it is easy to cause the coal quality to decline, and the different hardness of coal and rock will also seriously affect the service life of the shearer pick. The force of the machine drum, the loss of mechanical load and the power of the cutting motor will be greatly affected, which will reduce the quality and efficiency of coal mining, and cannot guarantee high-quality a...

AI Technical Summary

Problems solved by technology

Once the cut rock is mixed into the coal seam during the coal mining process, it is easy to cause a decline in coal quality. The difference in hardness between coal and rock will also seriously affect the service life of the shearer pick. The force of the machine drum, mechanical load loss and cutting motor power will be greatly affected, which will reduce the quality and efficiency of coal mining, and cannot guarantee high-quality and high-volume coal mining. Therefore, the accurate identification of the trajectory of the coal-rock interface It is the bottleneck problem to realize the automation and intelligence of coal mine fully mechanized mining face

[0003] At present, there are single-sensor identification methods and multi-sensor information fusion identification methods for coal-rock interface identification. The single-sensor identification method mainly collects a single vibration signal or acoustic emission signal during cutting, or uses the collected data The energy spectrum is analyzed in the time domain or frequency domain, but the recognition accuracy is low and the fluctuation is large due to environmental interference; while the multi-sensor information fusion method can improve the recognition accuracy and integrate data samples from various sources, but the fusion process There are often data conflicts between the information collected by multiple sensors, and the dependence and reliability of coal rock identification under different coal rock hardness conditions on different types of sensor data are very different, and the coal rock identification results are often not accurate enough

Images