Tuesday, October 15th 2024
(Milan Time) 14:00-17:30
(Beijing Time) 20:00-23:30
Tencent ID : 425-418-722
Session Aims & Scope
The aim of this session is to explore how digital twin technology combined with artificial intelligence of the energy sector, including the design and manufacture, inspection and control, monitoring and maintenance of Intelligent equipment of coal mining and petroleum. At present, energy equipment is in a critical period from visual intervention to intelligent adaptive control. The capabilities of Intelligent monitoring, sensing and diagnosis are the technical problems in the energy industry. In recent years, digital twin technology has been deeply applied in the fields of intelligent production, comprehensive automatic control and full life cycle health maintenance in coal, oil and other fields. Participants will gain insights into how digital twins can enhance efficiency, safety, and sustainability while also discussing the challenges and opportunities for future development.
Session Chair(s)
Chair
Xuhui ZHANG
Professor
Xi’an University of Science and Technology (China)
Co-Chair
Xuewen WANG
Professor
Taiyuan University of Technology (China)
Co-Chair
Yinan GUO
Professor
China University of Mining & Technology (China)
Session Presentation
1.
Xuhui ZHANG
Professor
Xi’an University of Science and Technology (China)
Title: DT-driven Virtual tunning and decision control of underground mining
Abstract
The traditional automatic control method is difficult to adapt to the complex underground environment, and the debugging of the control system is difficult. With the help of digital twin, visual pose measurement, robot control and other technologies, the research on the virtual debugging and decision control method of the section forming process of mining equipment strongly supports the national strategic orientation of unmanned operation in coal mines. Through the construction of the driving equipment intelligence agent, the deep reinforcement learning algorithm is used to realize the autonomous decision-making of the driving equipment section forming in the unknown environment under complex working conditions, achieve the autonomous decision-making within the specified step size, and obtain the high-precision cutting trajectory of the section forming. A virtual debugging platform was built, and virtual debugging and application experiments were carried out. The experimental results showed the feasibility of the system and provided a new technical idea for the intelligent control of underground driving equipment.
2.
Yinan GUO
Professor
China University of Mining & Technology (China)
Title: Key technologies for digital twin in smart coal mines
Abstract
Digital twin plays a core role in improving the intelligence of a coal mine. Facing various mining processes, different decision-making and control problems are emerged in digital twin-based intelligent platforms for mining coal. Based on this, an architecture of digital twin-based intelligent platforms is first illustrated. Then, key technologies in mining coal and excavating a roadway, as well as the transportation system are analyzed, respectively. A five-dimensional digital twin model is built for the intelligent mining workface, and knowledge-guided autonomous virtual-real interaction method is put forward. Especially, three classification methods for data streams stemmed from cutting part in a shearer are presented to analyze the coupling relationship between the characteristics of cutting part in a shearer and rock seam. Digital twin-based control strategies are put forward to guarantee the reliable temporary and permanent support for the roof of a roadway. Finally, a digital twin-based transportation platform for open-pit mine is given and the planning method for its haulage systems is presented.
3.
Xusheng XUE
Associate Professor (PhD)
Xi’an University of Science and Technology(China)
Title: Virtual Testing Method for Explosion Proof Performance of Coal Mine Equipment Driven by Multi source Data
Abstract
The explosion-proof performance of coal mine equipment is an important premise to ensure coal mine safety production. At present, affected by the environmental conditions of coal mine equipment explosion-proof performance testing, the problems of single information, difficult to capture, and invisibility in the performance testing process lead to long detection cycle, low efficiency, and high cost of coal mine equipment explosion-proof performance testing. Therefore, it is urgent to study the virtual detection method of coal mine equipment explosion-proof performance driven by multi-source data. The mapping relationship between the characteristic parameters of coal mine explosion-proof equipment and the digital twin is studied, the multidimensional perception method of the physical entity and environmental characteristics of coal mine explosion-proof equipment is constructed, the correlation data set of the physical entity characteristic multidimensional data of coal mine explosion-proof equipment and the digital twin characteristic parameters is established, and the digital twin measurement system of coal mine explosion-proof equipment based on the data set is developed. This method effectively realizes the virtualization of the internal explosion propagation test measurement of coal mine equipment, improves the detection efficiency of the explosion-proof performance of coal mine equipment and reduces the cost, and provides the basis for the green design and manufacture of coal mine equipment.
4.
Yan WANG
Lecturer
Xi’an University of Science and Technology (China)
Title: Virtual-reality interaction control method for cantilever roadheader driven by digital twin
Abstract
Due to the time-delay caused by sensor data acquisition, signal processing, and network communication, the digital space and physical space cannot achieve absolute synchronization of virtual-real information decision-making. The accumulation of these time delays may result in a reduction in the timeliness of the decision-making in the digital space. Aiming at the current demand for real-time and reliability of intelligent control of cantilever roadheader, a virtual-real interactive control method for cantilever roadheader driven by digital twin is proposed, which plays the advantages of virtual control in decision-making security boundary and physical control in decision-making real-time. Firstly, the five-dimensional model architecture of virtual-real interaction control for cantilever roadheader is constructed, and the logical relationship between digital space and physical space is sorted out to realize the fusion of virtual-real interaction control decision-making; on this basis, the mathematical descriptive model of virtual-real interaction control is constructed, and the effect of virtual-real interaction time delay on the fusion control decision-making of the roadheader is analyzed; furtherly, the execution logic architecture of the virtual-real interaction fusion control model of the roadheader considering time-delay characteristics is established, and the inherent delay effect on the virtual-real interaction control of the roadheader is reduced through the processes of “state alignment”, “evolution prediction” and “decision fusion”. Finally, the shaping quality evaluation of the roadway is examined to verify the effectiveness of the proposed virtual-real interaction fusion control method.
5.
Haoqi WANG
Professor
Zhengzhou University of Light Industry (China)
Title: Key technologies and application exploration of industrial digital twin system for grinding and mining Equipment
Abstract
High-end equipment, such as large-scale mineral grinding and mining equipment, is the key equipment for Chinese economic pillar industries of mining and coal, and has the characteristics of high value, multiple equipment coordination, many control and operation points, and harsh operating environment. Among them, large-scale grinding equipment occupies the core link in the industrial chain and its investment and energy consumption account for more than 70% of the mineral processing process. Energy saving optimization is the key to achieving the intelligent transformation of coal mining equipment, and digital twin provides an effective approach. Digital twin is expanding from local digital twin scenarios of mechanical equipment to the industrial digital twin system (iDTS) of “human-machine-environment” integration. This presentation first introduces the characteristics and reference architecture of the industrial digital twin system (iDTS) for coal mining equipment. Then, it introduces relevant key technologies, including multi-physical field simulation, multi-modal perception and intelligent interaction methods, and AI agent-based system self-learning and optimization control methods. Finally, it shares practical exploration cases, , including the virtual-real mapping of the scraper conveyor S-bend based on digital twin, the energy-saving regulation of the large vertical mill based on multi-physical field simulation, and the optimization design of the impact crusher for dust and energy saving.
6.
Shuguang LIU
Ph.D. Student
Taiyuan University of Technology (China)
Title: XR intelligent operation and maintenance system for complex human-robot collaboration tasks in coal mines
Abstract
With the development of coal mine intelligence and the development and application of coal mine robots, efficient collaboration between coal mine operators and coal mine robots plays a vital role in complex underground tasks. Aiming at the complex scenarios of multi-person and multi-robot collaboration in coal mine auxiliary operation, a coal mine XR intelligent operation and maintenance system was established to empower the collaboration between two types of operators (including central control operators and field control operators) and two types of robots (including detection robots and operation robots). The VR operation and maintenance subsystem, AR operation and maintenance subsystem, and communication network were constructed respectively. Through the collaborative operation of the VR/AR operation and maintenance subsystems with coal mine operators and coal mine robots, coordinated perception, decision-making, and control in both virtual and physical spaces can be achieved. This allows for the simulation, iteration, optimization, and verification of complex tasks in the physical space within the virtual space, forming an intelligent operation model of human-human, human-machine, and machine-machine interaction and collaboration. It not only provides a new approach and paradigm for human-robot collaboration in complex mining tasks, but also serves as a general reference for similar application scenarios in other industries. It not only provides a new approach and paradigm for human-robot collaboration in complex mining tasks, but also serves as a general reference for similar application scenarios in other industries.
7.
Suhua LI
PhD student
Taiyuan University of Technology (China)
Title: A Digital Twin-based Bi-directional Deduction Method for the Full-pose of the Floating Connection Mechanism
Abstract
The pose of the equipment connected using the Floating Connection Mechanism (FCM) is closely related to the straightness of the fully-mechanized mining face. However, the current monitoring of equipment posture still exists in the blind zone, unreliable sensing data, etc. Therefore, We proposes a digital twin-based bi-directional deduction method for the pose of starting/end point of the FCM. Firstly, considering the pose of the connection equipment, an optimization of the motion law of the FCM based on Conformal Geometry Algebra (CGA) is realized; secondly, based on the motion law and the structural character of the equipment, both the real-time solution of the end pose and the virtual application of optimization motion laws of the FCM are achieved by marking key points and extracting key information; then, referring to the information enhancement scheme obtained by adding digital twin sensors enhancement and a bi-directional deduction method is proposed for the virtual-real fusion of the pose of the starting/end point of the FCM; finally, the experimental validation shows that the motion law of the FCM proposed in this paper is in line with the actual motion and generates better analytical results.
8.
Jie SUN
Professor
Northeastern University (China)
Title: Application of Digital Twin for industrial process control: A case study of gauge-looper-tension optimized control in strip hot rolling
Abstract
During the hot rolling process, the performance of most control systems gradually degrades due to equipment aging and changing process conditions. However, existing gauge-looper-tension control method remain restricted by a lack of intelligent parameter maintenance strategies. To address this challenge and enhance the smart manufacturing capabilities of strip hot rolling, based on the digital twin method, this paper proposes a data-driven optimized control method for the gauge-looper-tension system of strip hot rolling. First, a hot rolling process model is constructed based on a digital twin method to serve as an evaluation and optimization platform. Subsequently, relevant data are collected to calculate the Hurst index for identifying the performance of the controller during the rolling process. Additionally, for controllers with poor Hurst index values, the crayfish optimization algorithm is employed for adjusting controller parameters to maximum the Hurst index. Experimental results demonstrate that the evaluation method effectively recognized the control state of gauge-looper-tension system and the optimization method successfully enhances the performance of the control system. Therefore, based on the digital twin platform, the proposed method can effectively maintain performance-degraded control systems.
