AP23489685 – Development of a digital method for monitoring defects and characteristics of intra-quarry technological roads during open-pit mining of mineral deposits
Objective of the project – Increasing the operating efficiency of mining dump trucks through the development and implementation of a digital method for monitoring defects, which considers the impact defects in quarry roads on the durability of metal structures of mining dump trucks, allowing management predictive measures to obtain the best performance.
Relevance: At the current stage of open-pit mining, the most labor-intensive and costly process is the transportation of rock mass, with up to 65% of haulage performed by truck transport. The increase in the load capacity of mining dump trucks expands their field of application; however, it is accompanied by high fuel consumption, tire wear, and increased transportation costs. Studies show that improving the design and maintenance of haul roads can reduce transportation costs by 20–60%, while road conditions account for 23–27% of the total cost of haulage. The key factors affecting efficiency are surface smoothness, pavement strength, and road construction and maintenance technologies. Road surface irregularities cause significant dynamic loads on dump trucks, accelerating equipment wear and tire damage, which accounts for 50–70% of tire failures. At the same time, dump trucks themselves contribute to road deterioration, increasing the number of defects. Existing road monitoring methods are expensive, complex, and dependent on external conditions, which limits their application. The lack of effective methods for detecting defects and their relationship with structural loads prevents accurate assessment of road conditions and prediction of service life. In this regard, the development and implementation of digital methods for monitoring haul roads, considering road characteristics and operating conditions, is a relevant task for improving the efficiency and safety of mining operations.
Scientific supervisor: Ph.D., Associate Professor, Utegenova Assem
Expected and achieved results: Based on calculations and computer modeling, the hardware and software configuration of a digital monitoring system for in-pit haul roads has been substantiated and selected, and its functionality has been tested and adjusted. An analysis of existing approaches to processing data on road conditions was carried out, and optimal neural network architectures for defect analysis were selected. Integration of data from pressure and acceleration sensors, loads on dump truck metal structures, and GPS systems was implemented. Algorithms for normalization, cleaning, and reconstruction of missing data, as well as neural network-based intelligent analytics models, were developed. A self-learning software algorithm was created with adaptation to different types of dump trucks and road defects, and an initial database was formed. Testing of the software systems was carried out, errors were identified and corrected, and stable operation under variable loads was ensured. Equipment for onboard data acquisition devices and positioning systems was substantiated and selected. A methodology was developed for identifying hidden dependencies between dump truck loads and road defects, as well as a method for intelligent data analysis with requirements for measurement equipment. Calibration and verification of measuring systems were performed, confirming their metrological reliability. A conceptual scheme of the hardware and software complex for digital monitoring was developed. A prototype software system for automated data collection, processing, and geospatial visualization was created. Preliminary analytical results were obtained, including classification of road surface defects and identification of key degradation factors. Methodological and technical documentation was prepared to support system implementation. The effectiveness of digital monitoring was justified based on analytical and patent studies, and the use of drones, laser scanning, and artificial intelligence was proposed. Experimental studies were conducted, and a patent for the invention was obtained. An analysis of the stress-strain state of dump trucks was carried out using numerical modeling and finite element methods. A method for monitoring road defects using onboard sensors and machine learning models was developed, enabling wear prediction and improving operational safety.
List of publications with links to them
- Кадыров Ж.Н., Шакенов А.Т. Переналаживаемый измерительный комплекс автоматической системы оперативного выявления и идентификации размеров и конфигурации дефектов полотна карьерных технологических дорог: патент на изобретение № 36969 от 04.10.2024. – Республика Казахстан, 2024.
- Акт о внедрении методики цифрового мониторинга состояния внутрикарьерных технологических дорог в филиале АО «ЕЭК» разрез «Восточный» от 29.05.2025. – Республика Казахстан, 2025.