Research and application of polynomial models in the development of an automated geological and mine surveying support system for open-pit mining

О. Zelensky Oleksii Donchenko
Received 15.02.2025, Revised 27.05.2025, Accepted 30.06.2025
Abstract

The relevance of the research was determined by the necessity to improve the accuracy and efficiency of geological and mine surveying operations under conditions of complex geological structures and limited resources for field investigations. The development and implementation of automated software enabled the prompt analysis of large data volumes, optimisation of the testing network, and reduction of exploration costs. The aim of the research was the development of software for polynomial models of curves and surfaces with a proposed simplified algorithm for implementing the least squares method. The desktop application software was developed in Microsoft Visual Studio 2019 using the Microsoft Foundation Classes library. The mobile application was developed in Android Studio. For 3D graphics, the OpenGL library with shaders was employed, ensuring high operational performance. Parameters for evaluating the adequacy of the model were defined, as standard standalone packages provided insufficient objectivity. It was demonstrated in automated mode that the Lagrange polynomial represents a particular case of the developed polynomial model in curve construction. Furthermore, the interpolation method for surface construction was improved by incorporating spatial variability of indicators through the use of autocorrelation functions. During the construction of autocorrelation functions, spatial variability was assessed based on autocorrelation coefficients between values at adjacent control points, as well as the critical correlation radius, which made it possible to evaluate the predictive potential beyond the polynomial construction zone. The application of critical correlation radius in the automated geological and mine surveying system for analysing the testing network of all boreholes (exploratory and blasting) was described. It was noted that the network parameters should reflect the natural variability of the studied indicators. In cases where the deposit depth exceeded the bench height, the feasibility of predicting qualitative characteristics based on the results of blasting boreholes located above the mineral deposit was considered, allowing for a significant reduction in the scope of costly exploratory works. It was established that the multi-module automated geological and mine surveying system had been developed and implemented over many years in open-pit mines, particularly at the Erdenet enterprise (Mongolia), which develops a copper-molybdenum deposit. The results of the conducted research were utilised in the educational process and in the automated geological and mine surveying system for planning and managing mining operations in open-pit environments

Keywords:
regression model; autocorrelation function; least squares method; Lagrange method; frame model; critical correlation radius; indicatrix

Retrieved from Vol. 59, No. 1, 2025

Pages 10–21

References Suggested citation
  1. Bell, J.W. (2020). Polynomial least squares multiple-model estimation: Simple, optimal, adaptive, and practical. SN Applied Sciences, 2, article number 1964. doi: 10.1007/s42452-020-03439-x.
  2. Bidiuk, P.I., Demkivskyi, Y.O., & Demkivska, T.I. (2020). Method of development of heteroskedastic processes models. Bulletin of the Kyiv National University of Technology and Design. Technical Science Series, 142(1), 19-26. doi: 10.30857/1813-6796.2020.1.2.
  3. David, I.J., Adubisi, O.D., Ogbaji, O.E., Eghwerido, J.T., & Umar, Z.A. (2020). Resistant measures in assessing the adequacy of regression models. Scientific African, 8, article number e00437. doi: 10.1016/j.sciaf.2020.e00437.
  4. Horbachuk, V.M., & Kushlyk-Dyvulska, O.I. (2023). Probability theory and mathematical statistics. Kyiv: Igor Sikorsky Kyiv Polytechnic Institute.
  5. Khlivnyi, V.V., & Bazilo, C.V. (2023). Influence of operating parameters of hydroabrasive cutting of various materials on the accuracy of observing the cutting width. Bulletin of Cherkasy State Technological University, 28(2), 38-49. doi: 10.24025/2306-4412.2.2023.280719.
  6. Lapach, S.M. (2025). Conflict of user accuracy requirements and regression model indicators. Mathematical Machines and Systems, 2025(1), 91-102. doi: 10.34121/1028-9763-2025-1-91-102.
  7. Laverty, W.H., & Kelly, I.W. (2021). Exploring the effects of assumption violations on simple linear regression and correlation using Excel. American Journal of Theoretical and Applied Statistics, 10(4), 194-201. doi: 10.11648/jajtas.20211004.13.
  8. Manoukian, E.B. (2021). Mathematical nonparametric statistics. Hoboken: Taylor & Francis Group.
  9. Mountgomery, D.C., Peck, E.A., & Vining, G.G. (2021). Introduction to linear regression analysis (6th ed). Hoboken: John Wiley & Sons, Inc.
  10. Pardoe, I. (2021). Applied regression modeling (3d ed.). Hoboken: John Wiley & Sons, Inc.
  11. Prokaza, O.I., & Kuznetsova, O.V. (2022). Construction of mathematical models of evaporatiоn plant by experimental-statistical method. Visnik of the Volodymyr Dahl East Ukrainian National University, 271(1), 36-40. doi: 10.33216/1998-7927-2022-271-1-36-40.
  12. Romanyuk, O.N., Romanyuk, O.V., & Tsikhanovskaya, O.M. (2024). Information support for economic disciplines. In Proceedings of the XVII international scientific and practical conference of the information technologies and automation – 2024 (pp. 393-396). Odesa: Odesa National University of Technology.
  13. Sencer, B. (2024). On the pros and cons of using Excel for regression analysis. Turkish Journal of Agriculture – Food Science and Technology, 12(s2), 2234-2241. doi: 10.24925/turjaf.v12is2.2234-2241.6931.
  14. Srilakshmi, U., Manikandan, J., Thanmayee, V., Abhinav, G., Tharun, K., & Saideep, D. (2024). A new approach to computationally-successful linear and polynomial regression analytics of large data in medicine. Journal of Computer Allied Intelligence, 2(2), 35-48. doi: 10.69996/jcai.2024009.
  15. Tadashi, D., Siqiao, L., & Okamura, H. (2024). Local polynomial software reliability models and their application. Information and Software Technology, 166, article number 107366. doi: 10.1016/j.infsof.2023.107366.
  16. Tkachenko, O.M. (2021). Polynomial methods and tools for estimating regression parameters using Non-Gaussian error models. (Doctoral dissertation, Cherkasy State Technological University, Cherkasy, Ukraine).
  17. Vizhva, Z., Demydov, V., & Vizhva, A. (2023). Statistical modeling of random fields with Gaussian correlation function for magnetometry data analysis. Bulletin of Taras Shevchenko National University of Kyiv: Geology, 3(102), 81-87. doi: 10.17721/1728-2713.102.
  18. Zelensky, O. (2023). Construction surface using desktop and mobile applications. Scientific and Practical Journal “Economics and Technical Engineering”, 1, 87-102. doi: 10.62911/ete.2023.01.01.07.
  19. Zelensky, O.S., & Lysenko, V.S. (2022). Improvement of automated contouring of ore grades along the horizons of ore quarries. Journal of Kryvyi Rih National University, 55, 15-24. doi: 10.31721/2306-5451-2022-1-55-15-24.
  20. Zorin, O., & Palahin, V. (2023). Adaptive system for receiving data by bipolar discrete signals in information and measurement systems in asymmetric non-Gaussian noise. In International scientific conference of the information technologies and computer modelling (pp. 181-182). Ivano-Frankivsk: Vasyl Stefanyk Precarpathian National University.
Zelensky, О., & Donchenko, О. (2025). Research and application of polynomial models in the development of an automated geological and mine surveying support system for open-pit mining. Mining Journal of Kryvyi Rih National University, 59(1), 10-21. https://doi.org/10.31721/2306-5435-2025-1-10-21
en