Identification of Rock Properties of Rock Wall Cut by Roadheader Based on PSO-VMD-LSSVM

Abstract

The problem of low digging efficiency and mining imbalance due to outdated digging technology and low degree of equipment intelligence has long existed in coal mine roadway excavation work. Lithology identification is the key to the intelligence of roadheading equipment. Accurate lithology identification significantly affects the automatic control of roadheader cutting conditions. Completing the identification of lithology in the process of rock wall cutting by a roadheader involved the following steps: building a tunneling experiment platform, making four rock specimens with different lithologies, completing the tunneling simulation experiments on four lithologies, obtaining current sensor data of four lithologies cutting, and finally proposing an intelligent lithology identification method of PSO-VMD-LSSVM. The research results show that the particle swarm algorithm (PSO) optimized the variational modal decomposition (VMD) with minimum envelope information entropy as the fitness function can realize the adaptive decomposition of the current signal of truncated motors. The signal reconstruction can increase the signal-to-noise ratio of the current signal by selecting the eigenmodal components according to the energy density and correlation coefficient criterion. The multi-scale fuzzy entropy is used as the eigenvector of the reconstructed current signal as the fuzzy entropy of different lithology cut-off motor currents has better differentiation at different scales. The least-squares support vector machine (LSSVM) is used to classify the feature vectors processed by custom decomposition parameter VMD and gives a recognition rate of 87.5%. The recognition rate increases to 97.5% for the feature vectors processed by PSO-VMD. The particle swarm algorithm optimizes the noise reduction via VMD to effectively improve the lithology recognition rate. The research results can provide a methodological reference for rock property recognition during rock cutting by a roadheading machine.

Research on Path Planning and Control Method for Secondary Autonomous Cutting of Cantilever Roadheader in a Large-Section Coal Roadway

“…It could be seen from the analysis of Equation ( 5) that each variable in the pose transformation matrix of Equation ( 3) should be equal to the corresponding element in the matrix composed of joint variables of Equation (5). From these equations, the relationship between joint variables and n, o, a, p vectors could be obtained.…”

“…It can be said that fossil energy such as coal will remain the main energy source in China for a long time to come [1][2][3]. Coal mining mainly adopts the underground mining method, and a large number of roadways need to be excavated to ensure the efficient production of coal, so it can be said that roadway tunneling is the premise of coal mining [4,5]. The main corollary equipment for roadway excavation in coal mines in China is an anchor drilling machine and cantilever roadheader, and the cantilever roadheader integrates the functions of cutting, walking, transportation and dust reduction [6].…”

Comparative Study SVM and Random Forest Algorithms for the Classification of Terrestrial Visual Rock Types

“…Aligned with the progressions in time and contemporary technology, the utilization of image processing technology and artificial intelligence has become indispensable across various domains, including geology and natural resource mapping [1,2]. An essential aspect of understanding the geological environment is the categorization of terrestrial visual rock types.…”