CC BY
1
2023;27(4):655-663
ISSN 2782-6341 (online)
Original article
MECHANICAL ENGINEERING
EDN: BMYSAA
DOI: 10.21285/1814-3520-2023-4-655-663
Particle swarm optimization support vector machine-based coal and rock cutting tool load spectrum identification method
Ruohan Liu1 , Lan Lyu2, Sunbao Wang3, Zhiqiang Chai 4
134Heilongjiang University of Science and Technology, Harbin, China 2Komsomolsk-na-Amure State University, Komsomolsk-on-Amur, Russia
Abstract. The goal of this research is to achieve safe and efficient excavation of coal and rock tunnels with complex geological structures, and to enhance the self-sensing ability of coal and rock cutting equipment and tools. Particle swarm optimization support vector machine is used to identify the cutting state of disc cutting tools. EDEM finite element analysis software is used to analyze cutting process characteristics of the disc cutting tool when used to cut through coal and rock with different compressive strengths. Empirical mode decomposition is used to decompose the load spectrum characteristics; for this purpose, the first-order and seventh-order intrinsic mode functions containing all the feature information of the original signal of the load spectrum are selected. The sample entropy is calculated as the feature input vector. The extracted feature vector is input into the trained support vector machine model and the particle swarm optimization support vector machine model. By extracting the sample entropy of the load spectrum of the disc cutter as the feature vector, the particle swarm optimization support vector model is used to identify the cutting state of the coal and rock. The recognition accuracy of the support vector machine model before and after the improvement is compared and analyzed. The results show that compared to the unoptimized support vector machine, the support vector machine optimized by particle swarm optimization can identify the load spectrum of the coal more quickly and accurately. The recognition accuracy is 96,82%, which verifies the effectiveness of the particle swarm optimization support vector machine model in identifying the load spectrum of the coal and rock disc cutter.
Keywords: disc cutter, coal and rock, discrete element modeling, feature vector, particle swarm optimization
Acknowledgments: This work was supported by the Chinese National Natural Science Foundation (52104130 and 51974111).
For citation: Liu Ruohan, Lyu Lan, Wang Sunbao, Chai Zhiqiang. Particle swarm optimization support vector machine-based coal and rock cutting tool load spectrum identification method. iPolytech Journal. 2023;27(4):655-663. https://doi. org/10.21285/1814-3520-2023-4-655-663. EDN: BMYSAA.
Метод распознавания диапазона нагрузок дискового резака при резке угля и горной породы на основе метода опорных векторов
оптимизации роя частиц
13-4Хэйлунцзянский университет науки и техники, г. Харбин, КНР
2Комсомольский-на-Амуре государственный университет, г. Комсомольск-на-Амуре, Россия
Резюме. Цель - выявление решений, направленных на обеспечение безопасной и эффективной экскавации сложных геологических формаций угольных и каменных месторождений в туннелях, а также на повышение чувствительности оборудования и инструментов для резки угля и породы. Для идентификации состояния реза дискового инструмента был использован метод оптимизации роя частиц, поддерживающий метод опорных векторов. Программный продукт конечно-элементного анализа EDEM был использован для анализа характеристик дисковых резаков, применяемых для резания угольных пород с различными пределами прочности на сжатие. Спектральные характеристики нагрузки на резак были разложены на эмпирические моды, при этом были выбраны компоненты внутренних колебаний первого и седьмого порядка, содержащие всю характерную информацию из исходного сигнала спектра на-
МАШИНОСТРОЕНИЕ
Научная статья УДК 622
Жохань Лю1: , Лань Люй2, Суньбао Ван3, Чжицян Чай4
© Liu Ruohan, Lyu Lan, Wang Sunbao, Chai Zhiqiang, 2023 https://ipolytech.elpub.ru -
грузки. Вычисленная энтропия сигнала была использована в качестве входного вектора признаков. Извлеченные векторы признаков были введены в модель опорных векторов и модель оптимизации роя частиц с поддержкой метода опорных векторов. В результате проведенных исследований на основе спектра нагрузки дискового резака была испытана модель оптимизации роя частиц с поддержкой модели опорных векторов для распознавания состояния резания и сравнивалась ее точность с неоптимизированной моделью опорных векторов. Полученные результаты указывают, что по сравнению с неоптимизированной моделью метода опорных векторов модель оптимизации роя частиц с поддержкой модели опорных векторов может быстрее и точнее идентифицировать спектр нагрузки режущего диска для резания угольной породы. Точность распознавания составляет 96,82%, что подтверждает эффективность данной модели при определении спектра нагрузок режущего диска, применяемого для работы с угольными породами.
Ключевые слова: дисковая фреза, уголь и горная порода, моделирование методом дискретных элементов, вектор признаков, оптимизация роя частиц, метод опорных векторов оптимизации роя частиц
Благодарности: Исследования выполнены при поддержке Китайского национального фонда естественных наук (52104130 и 51974111).
Для цитирования: Лю Жохань, Люй Лань, Ван Суньбао, Чай Чжицян. Метод распознавания состояния резки дискового резака на основе метода опорных векторов оптимизации роя частиц // iPolytech Journal. 2023. Т. 27. № 4. С. 655-663. (In Eng.). https://doi.org/10.21285/1814-3520-2023-4-655-663. EDN: BMYSAA.
ВВЕДЕНИЕ
Cantilever roadheader is the main equipment in coal mines, and its working environment is complicated. The load spectrum of the working mechanism of the tool cutting coal and rock, which contains rich information, and the characteristics of the load spectrum under different working conditions can help identify the cutting state of the tool, which has certain practical significance to improve the intelligent cutting of roadheader [1, 2].
The load spectrum of coal and rock cut by disc tool is nonlinear and has certain randomness, so there are some difficulties in extract its frequency features. The variational modal decomposition (VMD) needs to determine the number of modes and their boundary effects artificially in advance when extracting the sample entropy as a feature vector, which limits its application. The wavelet transform algorithm is used to extract the features of the disc tool load spectrum with partial energy signal loss and the limitation of basis function selection, which is not suitable for its decomposition [3]. For the characteristics of the research subject, the sample entropy of intrinsic mode function (IMF) component is obtained by empirical mode decomposition (EMD), and the characteristic information of the disc tool load is extracted effectively by solving the problems of preset modal number and basis function selection [4-6]. Compared to neural networks, support vector machines (SVMs) are suitable for solving high-dimensional, small-sample, nonlinear problems and can effectively classify small-sample data, but it is particularly important to select appropriate SVM parameters, otherwise they are prone to fall into local optima and poor state recognition rates [7-9].
For this research, the disc tool load spectrum is taken as the research subject to address the above problems, EMD decomposition is used to extract the sample entropy as the feature vector, particle swarm optimization (PSO) algorithm is used to adaptively select the parameters of the support vector machine, algorithm validation is performed on the disc tool cut coal and rock load spectrum dataset that is classified and identified by support vector machine (SVM) and PSO-SVM. Its accuracy on the load spectrum recognition is explored, and the application of PSO-SVM model for coal and rock-cutting disc tool recognition based on the load spectrum is realized.
FEATURE EXTRACTION OF DISC TOOL LOAD SPECTRUM
Simulation analysis of coal and rock cut by disc tool using discrete element modeling software EDEM was conducted with the following parameters: the density of coal and rock is 1280, 2460, and 2630 kg/m3; Poisson's ratio of coal and rock is 0,28, 0,24, 0,23, and 0,31; elastic modulus of coal and rock is 2010, 3260, and 12100 MPa; normal contact stiffness of coal and rock is 1,108*108, 9,744*108, and 7,472*107 N/m3; and tangential contact stiffness of coal and rock is 8,514*107, 1,068*109, and 8,545*108 N/m3; the density of disc tool is 17850 kg/m3; Poisson's ratio of disc tool is 0,31. According to the discrete element method theory, the coal and rock is regarded as composed of discrete particles [10, 11]. In the discrete element software EDEM, the coal and rock particle model is established as a spherical particle with a diameter of 4 mm and a contact radius of 4.6 mm. The wedge angle of the disc tool is 55°, and the simulation model of the disc tool cutting coal and rock is shown in Fig. 1.
2023;27(4):655-663
The stress cloud of the disc tool is shown in Fig. 2 for the situation when the simulated disc tool is used to cut the broken coal and rock. The derived
ISSN 2782-6341 (online)
load data can be obtained under three cutting states, and the data is then input into Matlab to obtain the corresponding load spectrum.
Fig. 1. Discrete element modeling of disc cutting tools for crushing coal and rock Рис. 1. Конечно-элементное моделирование дисковых резцов для дробления угля и горной породы методом дискретных элементов
Fig. 2. Simulation stress cloud map of disc cutting tools Рис. 2. Карта облака напряжений при моделировании дисковых резаков
In EDEM, 30 samples were collected for each class of states, and divided into training and test sets with a ratio of 7:3. The disc tool starts to enter at the coordinate zero point, and
12 MPa
PSO-SVM STATE IDENTIFICATION MODEL, SUPPORT VECTOR MACHINE PRINCIPLE
Support vector machine (SVM) has improved the generalization ability in finding the process that minimizes the overall structured risk, so that good regression ability can still be maintained in small samples of the truncated load spectrum. The disc tool truncated coal and rock load spectrum has nonlinearity, and there exist sample sets (x1, y1), (x2, y2), ---, (xn, y) --- , xn e R,, yn e R,
more than 1500 valid data samples are collected for each state, and the three truncated coal and rock state load spectra are shown in Fig. 3.
MM
0 0.5 I 1.5
1/ s
42 MPa
(n = 1, 2, ---, i), and the basic idea is to use the nonlinear mapping p(x) to map into the high-dimensional space and perform linear regression to obtain the function as follows:
where u is the regression function weight vector; b is the bias with non-unique dimension.
After removing the idiosyncratic points in the
30 MPa
Fig. 3. Three load state curves Рис. 3. Три кривые нагрузки различных состояний
nonlinear case, the remaining part is linearly inseparable, implying that this point is less than 1. The relaxation variable Z is introduced to construct the optimal hyperplane with the constraint:
e>4tx)+b>\-4i (2)
The optimization function is ' expressed as follows:
mm —\\co\
.»л? 2
(3)
where C is a constant that constrains the degree of penalty:
s.t y,«(x)+b) >1 -£,1 = 1,2,-, JV (4)
l>0, 1=1,2>--- N. ■
Using Lagrange multipliers, the Lagrange function is established, while the kernel function k(x., xj) is introduced, and the SVM regression function expression is obtained as:
/О) = 2 - a*)K(x,,Xj ) + b
(5)
where a, a*is Lagrange multipliers, a.; > 0, a*> 0, a I. x a* = 0.
The artificially selected parameters of the kernel function g, and the constant C, which constrains the degree of penalty, have poor generalization ability. Thus, it is necessary to optimize the penalty coefficient C in the SVM model and the parameters of the selected kernel function when using SVM for load spectrum classification, and to optimize the hyperparameters of the SVM using the PSO algorithm to improve the accuracy of the SVM in recognizing the load spectrum.
PSO ALGORITHM
PSO is initialized as a population of random particles (random solutions), and the optimal solution is found by iteration. In each iteration [12-14], the particle updates itself by tracking two "extremes" (pbest, gbest). pbest and gbest are the individual and population optimums, respectively, and after finding these two optimums, the particle updates its velocity and position as:
vi=ywi + c/\ (pbest, - x, ) + c2r2(gbest - xf)
x;=xi+vi, (6)
where /' = 1, 2, ..., A/; N is the total number of particles in this cluster; vi is this cluster; Vi is the velocity of the particles; r1 and r2 are random
number between 0 and 1; x i is the current position of the particle; c1 and c2 are the learning factors (usually c1 = c2 = 2); w is the inertia factor; and the two listed e quations are the standard forms of PSO.
The larger the value of the inertia factor, the stronger the global search ability and the weaker the local search ability; the smaller the value, the weaker the global search ability and the stronger the local search ability. In order to get a better inertia factor, a linear decreasing weight (LDW) strategy is used, as shown in the following equation, which will become smaller and smaller as the number of iterations increases:
w,=(M>i-wt)(Gk-g)iGK+We
(7)
where ws is the initial inertia factor, we is the inertia factor at the maximum number of iterations, GK is the number of iterations, g is the global optimum.
For different search problems, the global and local search capabilities can be adjusted so that the PSO algorithm can optimize the parameters of the SVM.
IMPROVED PSO-SVM ALGORITHM
In this paper, the particle swarm algorithm is used to optimize the SVM by determining the number of populations and the maximum number of iterations [15-17], and adjusting its speed and position according to the current individual extremum it finds and the current global optimal solution shared by the whole particle swarm [18-20]. The principle of the disc tool cutting state recognition model of the roadheader mentioned in the paper is to perform EMD decomposition on the collected disc tool load spectra, extract the sample entropy of each IMF component as a feature vector, divide it into training set and test set, input to the state recognition model of particle swarm algorithm optimized SVM, and obtain the recognition results.
IDENTIFICATION OF TRUNCATED STATE LOAD SPECTRUM, EMD DECOMPOSITION
The EMD decomposition is calculated as follows.
According to the signal x(t), the local maxima and local minima are connected by three spline curves, so that the whole signal sequence is between the upper and lower envelopes, Find the mean value of the two m1(t), to obtain fy(t) as follows:
2023;27(4):655-663
x{t)-mi(t) = h\(t).
(8)
If ^(t) does not satisfy the precondition, consider h1(t) as a new signal sequence and apply the previous equation to find one that satisfies the condition, denoted as c1(t). c1(t) represents the EMD decomposition of the obtained signal sequence.
Separate the signal sequences and obtain a new signal sequence as follows:
r\{t) = Xl(t) -ci(t).
(9)
Consider r1(t) as a new signal sequence x(t), process it according to step 1 and step 2, and record the remaining signal sequence as i, where i is taken as 1, 2, ... m and repeat equation in step 1 m times, and stop when the termination condition is satisfied, resulting in the following:
?](0-c2(0 = r2(0
r2 (0-^,(0-^(0
Vi (0-^(0 = ^,(0
(10)
The signal sequence is decomposed from equations from steps 1 and 2 as follows:
S(0=i>(0 + r.(0, (11)
¡==11
where r (t) is the residual term.
m w
To ensure that the IMF component is meaningful, the standard deviation is used to determine when the «sieving» is over.
(12)
where T is the total time of the discrete signal sequence.
To ensure the stability and linearity of the IMF component and to ensure that the eigenmodal function has the corresponding physical significance, Z is taken as 0,2.
SAMPLE ENTROPY
The sample entropy (SE) is a measure of the complexity of the time series and is calculated as follows.
1 .For the time series {*(')"" -the reconstructed time series is obtained as
follows:
Ml
Mz)} = {x(j),x{i+l),---,x{t + m-\)}—--
r'.[n-r)] '
ISSN 2782-6341 (online)
where m is the embedding dimension and N represents the time duration.
2. Calculate the distance in the new time series with the formula
dy = d[x(i),x(j)] = max[| x(i+k),x(j+k) |] (14)
where dij is the one with the largest difference of the corresponding element.
3. Calculate the amount of distance dij less than the similarity tolerance rto obtain B™(r) , which is given by the following formula:
du < r ..
«;"(/•) - —-,/e|i,.v in| (15)
N -m
4. Find the mean of B"'(r) using the formula:
B"\r) =
N-m +1
.
(16)
5. For the m+7-dimension, repeat steps 1 to
5 to obtain Bm+\r)
The SE definition formula for the original sequence is expressed as follows:
Bm+I(r)1
SampEn{m,r)
limT-ln
■v^'- Bm(r)
]
(17)
When it is a finite number, its formula is expressed as follows:
SampEn{m,r,N) = \iiB"\r)-B"n\r) . (18)
FEATURE EXTRACTION AND RESULT ANALYSIS
The results of using the EMD algorithm to adaptively decompose the load spectrum are shown in Fig. 4.
Because of the different number of IMF components obtained from EMD decomposition of different types of load spectra, in order to maintain the same length of feature vectors in different states and include all information of the original load spectrum, the first 7 IMF components are selected to represent the original load spectrum. The extracted sample entropy values are calculated. 21 sets of data from each cutting state load spectrum were selected as the training set, and the remaining 9 sets of data as the test set. To verify the advantages of the PSO-SVM model proposed in the article, it was compared with the unoptimized SVM model with an initial population of 1000 and a maximum number of iterations of 300. The experimental results are
_ 40000 ¡>20000
. 100000 i 50000
-]ljjnljrlr
■^ыш^щшШ/ПЩЦщ
•1CCOO , 5000
2 °
- -SC00 юооо
л 5000 i 0
— -5000
a 10000
ll 5000
3 0
" -5000
N 5000
u. 0
s .scoo
■1COOO
1 60000
2 5 ê 40000 X
k aÉaÉià rt flàlitei ™r " ' ! p™" * " ™ ™ "
MfMflll II
#
-лЛЛ/ \r>f УЛ Д/Wv^N J
Л/ \д A/4 \ :
/:ч
. юоо:о
I
i, 50o:o
é I 'il' I .....
WAV-IVMV^'
i ° ■10000 1 SOCO
г о
- -5o:o л 10
200 <100 600 800 1OB0 1200 14C0 1600 л
12 MPa a
200 400 600 GOO 1000 1200 МС0 16И
It
30 MPa b
1200 1400
42 MPa c
Fig. 4. EMD decomposition of different cutting load spectra (a, b, c) Рис. 4. Эмпирическая модальная декомпозиция различных спектров нагрузки при резке (a, b, c)
Comparison of training stl^rcdiclion results SVM classification ассигасу^85Л 143% ^
2.8 2.6 14
I 2.2
E ? 2
о
о
S.18 1.6
true value
SVMpredictive vuluc
3 2.8 2.6 2.4
I 2.2
S
£ 1-8
Comparison oflraining sel prediction results PSO-SVM clossification accuracv=96.S254%
-1-]- -]-]-»(»Miit»»!*"»»111
• in I с rallie о PSO-SVM predict!« value
I *********************
10 20
SVM a
PSO-SVM b
Fig. 5. SVM and PSO-SVM recognition results (a, b) Рис. 5. Результаты распознавания метода опорных векторов и модели оптимизации роя частиц
с поддержкой метода опорных векторов (a, b)
shown in Fig. 5.
The classification accuracy of SVM is 85,71%, and the recognition accuracy of PSO-SVM is 96,82%. Compared with unoptimized, particle swarm optimization algorithm has shorter optimization time and higher optimization accuracy.
CONCLUSION
The recognition algorithm of disc cutting tool load spectrum based on PSO-SVM utilizes PSO adaptive optimization of SVM parameters. The disc cutting tool cutting coal and rock load
spectrum is selected as the test set, and the accuracy of support vector machine model recognition before and after improvement is compared and analyzed. The results show that by extracting the sample entropy of the cutting state load spectrum of the disc cutter as the feature vector and using the PSO-SVM model to identify the cutting state of coal and rock, the time and accuracy of the tool cutting are better than those of the unoptimized SVM, with less time consumption and a recognition accuracy of 96,82%. Based on the PSO-SVM model, an analysis was conducted on the dataset of coal and
rock load spectra for disc cutting tools, verifying the effectiveness of the PSO-SVM model in identifying coal and rock load spectra for disc cutting tools.
References
1. Wang Hong. The 40 years developmental review of the fully mechanized mine roadway heading technology in China. Journal of China Coal Society. 2010;35(11):1815-1820.
2. Liu Chunsheng, Li Degen, Yuan Hao. Theoretical mechanics model and load characteristics of coal and rock vibration cutting with disc cutter. Journal of China Coal Society. 2020;45(8):3015-3023.
3. Zhao Yan, Dang Kangjia, Sun Jiangshan. Fault diagnosis method of high-voltage circuit breaker based on SVM optimized firefly algorithm. Journal of Heilongjiang University of Science and Technology. 2023;33(1):123-128.
4. Sun Jingyun, Zhao Panpan, Ding Yi. RMB Exchange Rate Prediction Based on Sample Entropy Reconstruction and PSO Optimization Algorithm. Mathematics in Practice and Theory. 2021;51(19):70-83.
5. Liu Chunsheng, Liu Yanting, Li Degen, Liu Ruohan, Ren Chunping, Yuan Hao. The mechanism and load model of the interaction between the disc blade and coal and rock under axial vibration cutting. Journal of China Coal Society. 2023;48(1):484-496.
6. Saini M., Sinwar D., Swarith A.M., Kumar A. Reliability and maintainability optimization of load haul dump machines using genetic algorithm and particle swarm optimization. Journal of Quality in Maintenance Engineering. 2023;29(2):234-240. https://doi.org/10.1108/JQME-11-2021-0088.
7. Meng Fannian, Du Wenliao, Gong Xiaoyun, Li Hao, Xie Guizhong. Fault Recognition of Rolling Bearings Based on LSSVM Optimized by Particle Swarm Optimization. Bearing. 2020;12(493):43-50.
8. Jin Dalong, Yuan Dajun, Li Xinggao, Su Weilin. Probabilistic analysis of the disc cutter failure during TBM tunneling in hard rock. Tunneling and Underground Space Technology Incorporating Trenchless Technology Research. 2021;109:103744. https://doi.org/10.1016/j.tust.2020.103744.
9. Moodi M., Ghazvini M., Moodi H. A hybrid intelligent approach to detect android botnet using smart self-adaptive learning-based PSO-SVM. Knowledge-based Systems. 2021;222:106988. https://doi.org/10.1016Zj.knosys.2021.106988.
10. KochetkovaA.S. Handling equipment in the mining industry: analysis and comparison. Izvestiya Tul'skogo gosudarstvennogo universiteta. Tekhnicheskie nauki = Proceedings of the Tula State University. Series: Technical Sciences. 2021;6:218-221. (In Russ.). https://orcid.org/10.24412/2071-6168-2021-6-218-221.
11. Naeimipour A., Rostami J., Buyuksagis I.S., Frough O. Estimation of rock strength using scratch test by a miniature disc cutter on rock cores or inside boreholes. International Journal of Rock Mechanics and Mining Sciences. 2018;107:9-18. https://doi.org/10.1016/j.ijrmms.2018.03.020.
12. Kromsky E.I., Kondakov S.V., Tilloev K.Z. Cone rolling to crawler excavator. Vestnik Yuzhno-Ural'skogo gosudarstvennogo universiteta. Seriya: Mashinostroenie = Bulletin of the South Ural State University. Series: Mechanical Engineering Industry. 2018;18(1):34-39. (In Russ.). https://doi.org/10.14529/engin180104.
13. Flora M., Jäger M., Teuscher P. Selection and specification of a tunnel boring machine as the leading factor for the achievement of project goals. Geomechanics and Tunnelling. 2018;11(5):450-455. https://doi.org/10.1002/geot.201800044.
14. Stopka G. Modelling of rock cutting with asymmetrical disc tool using discrete-element method (DEM). Rock Mechanics and Rock Engineering. 2021;54:6265-6279. https://doi.org/10.1007/s00603-021-02611-y.
15. Bulut B., Gunduz O., Baydogan M., Kayali E.S. Determination of matrix composition for diamond cutting tools according to the hardness and abrasivity properties of rocks to be cut. International Journal of Refractory Metals and Hard Materials. 2021;95:105466. https://doi.org/10.1016/j.ijrmhm.2020.105466.
16. Johansson D., Hrechuk A., Bushkya V., Martensson M., Can A., Stahl J.-E. Small scale testing of PCD and WC-Co tooling in rock cutting using longitudinal turning. Wear. 2019;426-427(B):1515-1522. https://doi.org/10.1016/j.wear.2018.11.036.
17. Alehossein H., Li Xingsheng, Boland J. Towards improving rock cutting tools using thermally stable diamond composites. Advanced Materials Research. 2009;76-78:585-590. https://doi.org10.1016/j.wear.2023.204848.
18. Zhou Xiao-Ping, Zhai Shu-Fang, Bi Jing. Two-dimensional numerical simulation of rock fragmentation by TBM cutting tools in mixed-face ground. International Journal of Geomechanics. 2018;18(3):06018004. https://doi.org/10.1061/(asce) gm.1943-5622.0001081.
19. Sarwary E., Hagap C.P. The effect of changes in tool tip angle on the cutting performance of a pointed pick. Mining Technology. 2016;125(3):184-190. https://doi.org/10.1080/14749009.2016.1159056.
20. Rostamsowlat I., Richard T., Evans B. Experimental investigation on the effect of wear flat inclination on the cutting response of a blunt tool in rock cutting. Acta Geotechnica. 2019;14(2):519-534. https://doi.org/10.1007/s11440-018-0674-1.
Список источников
1. Wang Hong. The 40 years developmental review of the fully mechanized mine roadway heading technology in China // Journal of China Coal Society. 2010. Vol. 35. Iss. 11. P 1815-1820.
2. Liu Chunsheng, Li Degen, Yuan Hao. Theoretical mechanics model and load characteristics of coal and rock vibration https://ipolytech.elpub.ru -661
cutting with disc cutter // Journal of China Coal Society. 2020. Vol. 45. Iss. 8. Р. 3015-3023.
3. Zhao Yan, Dang Kangjia, Sun Jiangshan. Fault diagnosis method of high-voltage circuit breaker based on SVM optimized firefly algorithm // Journal of Heilongjiang University of Science and Technology. 2023. Vol. 33. Iss. 1. Р 123-128.
4. Sun Jingyun, Zhao Panpan, Ding Yi. RMB Exchange Rate Prediction Based on Sample Entropy Reconstruction and PSO Optimization Algorithm // Mathematics in Practice and Theory. 2021. Vol. 51. Iss. 19. Р. 70-83.
5. Liu Chunsheng, Liu Yanting, Li Degen, Liu Ruohan, Ren Chunping, Yuan Hao. The mechanism and load model of the interaction between the disc blade and coal and rock under axial vibration cutting // Journal of China Coal Society. 2023. Vol. 48. Iss. 1. Р 484-496.
6. Saini M., Sinwar D., Swarith A.M., Kumar A. Reliability and maintainability optimization of load haul dump machines using genetic algorithm and particle swarm optimization // Journal of Quality in Maintenance Engineering. 2023. Vol. 29. Iss. 2. Р. 234-240. https://doi.org/10.1108/JQME-11-2021-0088.
7. Meng Fannian, Du Wenliao, Gong Xiaoyun, Li Hao, Xie Guizhong. Fault Recognition of Rolling Bearings Based on LSSVM Optimized by Particle Swarm Optimization // Bearing. 2020. Vol. 12. Iss. 493. Р 43-50.
8. Jin Dalong, Yuan Dajun, Li Xinggao, Su Weilin. Probabilistic analysis of the disc cutter failure during TBM tunneling in hard rock // Tunneling and Underground Space Technology Incorporating Trenchless Technology Research. 2021. Vol. 109. Р. 103744. https://doi.org/10.1016/j.tust.2020.103744.
9. Moodi M., Ghazvini M., Moodi H. A hybrid intelligent approach to detect android botnet using smart self-adaptive learning-based PSO-SVM // Knowledge-based Systems. 2021. Vol. 222. Р 106988. https://doi.org/10.1016/j.knosys.2021.106988.
10. Кочеткова А.С. Подъемно-транспортное оборудование в горной промышленности: анализ и сравнение // Известия ТулГУ Технические науки. 2021. № 6. С. 218-221. https://orcid.org/10.24412/2071-6168-2021-6-218-221.
11. Naeimipour A., Rostami J., Buyuksagis I.S., Frough O. Estimation of rock strength using scratch test by a miniature disc cutter on rock cores or inside boreholes // International Journal of Rock Mechanics and Mining Sciences. 2018. Vol. 107. Р. 9-18. https://doi.org/10.1016/j.ijrmms.2018.03.020.
12. Кромский Е.И., Кондаков С.В., Тиллоев К.З. Конусный раскатчик к гусеничному экскаватору // Вестник Южно-Уральского государственного университета. Серия: Машиностроение. 2018. Т. 18. № 1. С. 34-39. https://doi.org/10.14529/ engin180104.
13. Flora M., Jäger M., Teuscher P. Selection and specification of a tunnel boring machine as the leading factor for the achievement of project goals // Geomechanics and Tunnelling. 2018. Vol. 11. Iss. 5. Р. 450-455. https://doi.org/10.1002/ geot.201800044.
14. Stopka G. Modelling of rock cutting with asymmetrical disc tool using discrete-element method (DEM) // Rock Mechanics and Rock Engineering. 2021. Vol. 54. Р 6265-6279. https://doi.org/10.1007/s00603-021-02611-y.
15. Bulut B., Gunduz O., Baydogan M., Kayali E.S. Determination of matrix composition for diamond cutting tools according to the hardness and abrasivity properties of rocks to be cut // International Journal of Refractory Metals and Hard Materials. 2021. Vol. 95. Р. 105466. https://doi.org/10.1016/j.ijrmhm.2020.105466.
16. Johansson D., Hrechuk A., Bushkya V., Martensson M., Can A., Stahl J.-E. Small scale testing of PCD and WC-Co tooling in rock cutting using longitudinal turning // Wear. 2019. Vol. 426-427. Iss. B. Р. 1515-1522. https://doi.org/10.1016/j. wear.2018.11.036.
17. Alehossein H., Li Xingsheng, Boland J. Towards improving rock cutting tools using thermally stable diamond composites // Advanced Materials Research. 2009. Vol. 76-78. Р 585-590. https://doi.org10.1016/j.wear.2023.204848.
18. Zhou Xiao-Ping, Zhai Shu-Fang, Bi Jing. Two-dimensional numerical simulation of rock fragmentation by TBM cutting tools in mixed-face ground // International Journal of Geomechanics. 2018. Vol. 18. Iss. 3. Р. 06018004. https://doi.org/10.1061/ (asce)gm.1943-5622.0001081.
19. Sarwary E., Hagap C.P. The effect of changes in tool tip angle on the cutting performance of a pointed pick // Mining Technology. 2016. Vol. 125. Iss. 3. Р 184-190. https://doi.org/10.1080/14749009.2016.1159056.
20. Rostamsowlat I., Richard T., Evans B. Experimental investigation on the effect of wear flat inclination on the cutting response of a blunt tool in rock cutting // Acta Geotechnica. 2019. Vol. 14. Iss. 2. Р. 519-534. https://doi.org/10.1007/s11440-018-0674-1.
INFORMATION ABOUT THE AUTHORS ИНФОРМАЦИЯ ОБ АВТОРАХ
Ruohan Liu, Лю Жохань,
Postgraduate Student, Lecturer, аспирант, преподаватель,
Heilongjiang University of Science and Technology, ^S™™™ университет науки и технологии,
2468, PuYuan Road, SongBei District, 1500р22и дорога Пуюань, район Сонгбэй,
Harbin 150022, China г. Харбин, 2468, Хэйлунцзян, Китай
[email protected] [email protected]
https://orcid.org/0009-0005-2875-5821 https://orcid.org/0009-0005-2875-5821
Lan Lyu,
Postgraduate Student, Komsomolsk-on-Amur State University, 27, Lenin Prospect, Komsomolsk-on-Amur 681013, Russia
Deputy director for international affairs,
Heilongjiang University of Science and Technology,
2468, PuYuan Road, SongBei District,
Harbin 150022, China
lvlan@[email protected]
https://orcid.org/0009-0003-0874-3810
Sunbao Wang,
Master's Degree Student,
Heilongjiang University of Science and Technology,
2468, PuYuan Road, SongBei District,
Harbin 150022, China
https://orcid.org/0009-0004-3227-6746
Zhiqiang Chai,
Master's Degree Student,
Heilongjiang University of Science and Technology,
2468, PuYuan Road, SongBei District,
Harbin 150022, China
https://orcid.org/0009-0000-2868-5222
Contribution of the authors
The authors contributed equally to this article.
Conflict of interests
The authors declare no conflicts of interests.
The final manuscript has been read and approved by all the co-authors.
Information about the article
The article was submitted 01.09.2023; approved after reviewing 13.11.2023; accepted for publication 20.11.2023 .
Люй Лань,
аспирант,
Комсомольский-на-Амуре государственный университет,
681013, г. Комсомольск-на-Амуре, пр. Ленина, 27, Россия
заместитель начальника международного отдела, Хэйлунцзянский научно-технический университет, г. Харбин,
150022, дорога Пуюань, район Сонгбэй, г. Харбин, 2468, Хэйлунцзян, Китай Мап@1980@163.сот https://orcid.org/0009-0003-0874-3810
Ван Суньбао,
магистрант,
Хэйлунцзянский университет науки и технологии, г. Харбин,
150022, дорога Пуюань, район Сонгбэй, г. Харбин, 2468, Хэйлунцзян, Китай [email protected] https://orcid.org/0009-0004-3227-6746
Чай Чжицян,
магистрант,
Хэйлунцзянский университет науки и технологии, г. Харбин,
150022, дорога Пуюань, район Сонгбэй, г. Харбин, 2468, Хэйлунцзян, Китай [email protected] https://orcid.org/0009-0000-2868-5222
Вклад авторов
Все авторы сделали эквивалентный вклад в подготовку публикации.
Конфликт интересов
Авторы заявляют об отсутствии конфликта интересов.
Все авторы прочитали и одобрили окончательный вариант рукописи.
Информация о статье
Статья поступила в редакцию 01.09.2023 г; одобрена после рецензирования 13.11.2023 г.; принята к публикации 20.11.2023 г.
