Digital twin measurement system and method for spatial straightness of fully-mechanized working face

Abstract

The present application relates to the field of straightness measurement, and discloses a digital twin measurement system and method for spatial straightness of a fully-mechanized working face. A knowledge-driven system performs motion mechanism analysis processing on coordinated advancement of equipment on the basis of spatial kinematics, and performs virtual physical relationship simulation on the basis of obtained parameterized operation rule data, so as to determine a mapping relationship between virtual equipment operation and real equipment operation. A data-driven system determines real-time pose data by means of virtual deduction and on the basis of the mapping relationship and information data, and performs pose calculation on the basis of the real-time pose data to obtain calculated pose data. A hybrid-driven system performs inversion and correction processing on the basis of the calculated pose data, then determines real-time working face straightness information, and on the basis of the mapping relationship and pose similarity, determines predicted working face straightness information on the basis of the working face straightness information and then performs straightness fusion processing to obtain the straightness of the fully-mechanized working face. The present application aims to realize real-time straightness measurement and improve the measurement accuracy.

Description

A digital twin detection system and method for spatial straightness of fully mechanized mining faces Technical Field

[0001] This application relates to the field of straightness detection, and in particular to a digital twin detection system and method for the spatial straightness of a fully mechanized mining face. Background Technology

[0002] In the operation and maintenance of fully mechanized mining face production systems, the straightness of the face affects mining efficiency and safety requirements. Good straightness is crucial for ensuring healthy equipment operation and efficient mining; therefore, reliable testing is one of the keys to ensuring safe and efficient production. The straightness of the fully mechanized mining face is reflected by the undulations of the laying trajectory of the hydraulic support group and the scraper conveyor, under the connection of the floating linkage mechanism. However, the following problems exist during the testing process:

[0003] Static straightness testing is one-sided. Static straightness refers to the straightness information obtained by detecting the initial fully mechanized mining face without considering changes in mining processes. However, in the process of straightness testing, the narrow and harsh working environment limits the application of some sensors in underground mining, and the influence of sensor errors makes it difficult to obtain reliable and comprehensive equipment position data. In particular, the positioning information of hydraulic supports is still missing. In addition, the complex connection relationships between equipment lead to a situation where "equipment position measurement is limited and straightness extrapolation is one-sided" in the current testing.

[0004] The lag in dynamic straightness detection. Dynamic straightness refers to the straightness of the working face after the equipment group's process has been switched, taking into account the switching of mining processes. As the cutting operation progresses, the timely adjustment of mining processes such as pushing conveyors and moving supports upgrades the measurement of working face straightness to a "four-dimensional spatiotemporal" measurement problem. However, the trajectory obtained by inverting straightness through the traction of the coal mining machine has a lag in the mining time sequence. In the process of straightness control, it is necessary to take into account the impact of the straightness of the working face after timely advancement. Therefore, it is necessary to detect the straightness of the working face after the advancement.

[0005] In some cases, the inertial carrier is integrated with an RTK-GNSS rover and wheel speed odometer. The trajectory of the rover is generated based on the measurement data as the test data, and the test data is evaluated based on the trajectory data to verify the inertial navigation test of the straightness of the working surface.

[0006] A strip-shaped diffuser is installed at the head of the hydraulic support push rod to identify the hydraulic support number; a video analysis server is used to acquire video images from multiple PTZ cameras, and a target detection model is trained based on the acquired video images; a working surface vector line is drawn using coordinate information, and the straightness of the working surface is calculated based on the working surface vector line.

[0007] By using a digital twin model of the hydraulic support in an intelligent fully mechanized mining face, the straightness of the hydraulic support under different bending conditions is simulated with high precision, enabling real-time data synchronization and iterative interaction between the physical entity and the digital twin of the hydraulic support.

[0008] However, the above research has two problems: (1) Insufficient coupling between sensor data and model, incomplete data and inaccurate model have resulted in low accuracy in monitoring the fully mechanized mining face; (2) The mechanism relationship between each level of the production system has only been partially made transparent, and the uncertainty factors of coal seam fluctuation, the connection mechanism of some equipment, and the coupling mechanism between equipment and mining environment are still black box problems.

[0009] Summary of the Invention

[0010] The purpose of this application is to provide a digital twin detection system and method for the straightness of a fully mechanized mining face, which can realize real-time detection of straightness and improve detection accuracy.

[0011] To achieve the above objectives, this application provides the following solution:

[0012] In the first aspect, this application provides a digital twin detection system for the spatial straightness of a fully mechanized mining face, including: a knowledge-driven system, a data-driven system, and a hybrid-driven system;

[0013] Both the knowledge-driven system and the data-driven system are connected to the hybrid driving system;

[0014] The knowledge-driven system is used for:

[0015] Based on spatial kinematics, the motion mechanism of the equipment is obtained, and the coordinated motion of the equipment is analyzed and processed to obtain parameterized operation rule data. The motion mechanism is determined by analyzing the single-machine motion of the fully mechanized mining equipment and the positional relationship between the fully mechanized mining equipment. The parameterized operation rule data is the equipment motion constraint parameters determined according to the coordinated operation rules between the fully mechanized mining equipment.

[0016] Based on the parameterized operation rule data, a virtual physical relationship simulation is performed to determine the mapping relationship between the operation of virtual and real equipment;

[0017] The data-driven system is used for:

[0018] The real-time pose data is determined by using a virtual simulation method based on the mapping relationship and the collected information data; the information data includes: environmental monitoring information data, equipment performance data, and pose sensing data during the production process;

[0019] The pose data is calculated based on the real-time pose data to obtain the calculated pose data.

[0020] The hybrid drive system is used for:

[0021] Based on the calculated pose data, inversion and correction processes are performed to obtain the straightness information of the equipment group.

[0022] The real-time straightness information of the working face is determined based on the straightness information of the equipment group.

[0023] Based on the mapping relationship and pose similarity, the predicted straightness information of the working face is determined according to the straightness information of the working face; the pose similarity is determined based on historical pose data, the historical advance trajectory of the fully mechanized mining equipment, and the coupling relationship between the coal seam and the equipment pose during the advance process;

[0024] Based on the predicted straightness information of the working face and the real-time straightness information of the working face, straightness fusion processing is performed to obtain the straightness of the fully mechanized mining face; the straightness of the fully mechanized mining face is used to characterize the undulation of the laying trajectory of the fully mechanized mining equipment.

[0025] In one embodiment, the knowledge-driven system includes: a runtime mechanism analysis unit, a runtime rule parameterization unit, and a physical relationship simulation unit;

[0026] The operation mechanism analysis unit is connected to the operation rule parameterization unit; the operation rule parameterization unit is also connected to the physical relationship simulation unit;

[0027] The operation mechanism analysis unit is used to determine the operation mechanism of the fully mechanized mining equipment based on the individual machine motion of the fully mechanized mining equipment and the positional relationship between the fully mechanized mining equipment.

[0028] The operation rule parameterization unit is used to analyze and process the equipment's coordinated motion based on spatial kinematics to obtain parameterized operation rule data.

[0029] The physical relationship simulation unit is used to perform virtual physical relationship simulation based on the physical engine and the stress conditions of the fully mechanized mining equipment, and to determine the mapping relationship between the virtual and real equipment operation according to the parameterized operation rule data.

[0030] In one embodiment, the data-driven system includes: a pose data modeling unit, a virtual detection unit, and a data processing unit;

[0031] The pose data modeling unit is coupled to the virtual detection unit; the data processing unit is connected to both the pose data modeling unit and the virtual detection unit.

[0032] The pose data modeling unit is used to collect initial information data and classify the initial information data to obtain information data;

[0033] The virtual detection unit is used for:

[0034] Based on the detection principle of physical sensors and the set error, a virtual sensor is determined;

[0035] Based on the virtual sensor, virtual detection of the fully mechanized mining equipment is performed according to the mapping relationship to obtain virtual sensing data;

[0036] Real-time pose data is determined based on the virtual sensing data and the information data using a virtual simulation method.

[0037] The data processing unit is used to perform collaborative pose calculation on the real-time pose data using multidisciplinary software to obtain the calculated pose data.

[0038] The data processing unit is also used to transmit the calculated pose data to each unit and the hybrid driving system based on the active communication mechanism of data interaction.

[0039] In one embodiment, the hybrid drive system includes: a static straightness detection unit and a dynamic straightness detection unit;

[0040] The static straightness detection unit is connected to the knowledge-driven system, the data-driven system, and the dynamic straightness detection unit, respectively; the dynamic straightness detection unit is also connected to the knowledge-driven system and the data-driven system, respectively.

[0041] The static straightness detection unit is used for:

[0042] The straightness information of the scraper conveyor in the fully mechanized mining equipment is obtained by inversion based on the calculated coal mining machine posture.

[0043] The straightness information of the hydraulic support group in the fully mechanized mining equipment is determined based on the point cloud information; the point cloud data is obtained by non-contact measurement method, which is to collect the positioning and attitude information of the hydraulic support group through three-dimensional lidar.

[0044] The straightness information of the scraper conveyor is corrected using point cloud information to obtain corrected straightness information;

[0045] Based on the straightness information of the equipment group and the parameterized operation rule data, the straightness information of the working face is determined; the straightness information of the equipment group includes: the straightness information and the corrected straightness information of the hydraulic support group;

[0046] The dynamic straightness detection unit is used for:

[0047] Based on the mapping relationship and the pose similarity, the predicted straightness information of the working surface is determined according to the straightness information of the working surface;

[0048] Based on the predicted straightness information of the working face and the real-time straightness information of the working face, straightness fusion processing is performed to obtain the straightness of the fully mechanized mining working face.

[0049] Secondly, this application provides a digital twin detection method for the spatial straightness of a fully mechanized mining face, characterized in that the method is implemented using the aforementioned digital twin detection system for the spatial straightness of a fully mechanized mining face; the method includes:

[0050] Acquire information data; the information data includes: environmental monitoring information data, equipment performance data, and posture sensing data during the production process;

[0051] The real-time pose data is determined using a virtual simulation approach based on the mapping relationship and the aforementioned information data. The mapping relationship is a one-to-one correspondence between the actual working face and the virtual physical relationship established by the parameterized operation rule data. The parameterized operation rule data is obtained by analyzing and processing the coordinated motion of the equipment based on spatial kinematics and motion mechanisms. The motion mechanism is determined by analyzing and processing the individual motion of the fully mechanized mining equipment and the pose correlation between the fully mechanized mining equipment. The parameterized operation rule data consists of equipment motion constraint parameters determined according to the coordinated operation rules between the fully mechanized mining equipment.

[0052] The pose data is calculated based on the real-time pose data to obtain the calculated pose data.

[0053] Based on the calculated pose data, inversion and correction processes are performed to obtain the straightness information of the equipment group.

[0054] The real-time straightness information of the working face is determined based on the straightness information of the equipment group.

[0055] Based on the mapping relationship and pose similarity, the predicted straightness information of the working face is determined according to the straightness information of the working face; the pose similarity is determined based on historical pose data, the historical advance trajectory of the fully mechanized mining equipment, and the coupling relationship between the coal seam and the equipment pose during the advance process;

[0056] Based on the predicted straightness information of the working face and the real-time straightness information of the working face, straightness fusion processing is performed to obtain the straightness of the fully mechanized mining face; the straightness of the fully mechanized mining face is used to characterize the undulation of the laying trajectory of the fully mechanized mining equipment group during the advancement process.

[0057] In one embodiment, a virtual simulation method is used to determine real-time pose data based on the mapping relationship and the information data, specifically including:

[0058] Based on the detection principle of physical sensors and the set error, a virtual sensor is determined;

[0059] Based on the virtual sensor, virtual detection of the fully mechanized mining equipment is performed according to the mapping relationship to obtain virtual sensing data;

[0060] Real-time pose data is determined by using virtual simulation based on the virtual sensing data and the information data.

[0061] In one embodiment, the straightness information of the equipment group is obtained by inversion and correction processing based on the calculated pose data, specifically including:

[0062] The straightness information of the scraper conveyor in the fully mechanized mining equipment is obtained by inversion based on the calculated coal mining machine posture.

[0063] The straightness information of the hydraulic support group in the fully mechanized mining equipment is determined based on the point cloud information; the point cloud data is obtained by non-contact measurement method, which is to collect the positioning and attitude information of the hydraulic support group through three-dimensional lidar.

[0064] The straightness information of the scraper conveyor is corrected using point cloud information to obtain corrected straightness information;

[0065] Based on the straightness information of the equipment group and the parameterized operation rule data, the straightness information of the working face is determined; the straightness information of the equipment group includes: the straightness information and the corrected straightness information of the hydraulic support group.

[0066] In one embodiment, the straightness of the fully mechanized mining face is obtained by performing straightness fusion processing using a Kalman filter method based on the predicted straightness information and the real-time straightness information of the working face.

[0067] In one embodiment, the digital twin detection method for the spatial straightness of the fully mechanized mining face further includes:

[0068] Data transmission is performed using a proactive data interaction communication mechanism.

[0069] In one embodiment, the data interaction active communication mechanism transmits data in the form of a data element interface.

[0070] According to the specific embodiments provided in this application, the following technical effects are disclosed:

[0071] This application provides a digital twin detection system and method for the spatial straightness of a fully mechanized mining face. By analyzing the complex connections between the structures of the fully mechanized mining equipment and considering spatiotemporal characteristics under the constraints of the mining process, the straightness of the fully mechanized mining face is determined. This application establishes corresponding knowledge-driven and data-driven systems from the perspectives of motion mechanism / rule parsing virtualization and reliable data detection. Based on the characteristics of the work process and the motion characteristics of the equipment and between equipment, the similarity of the pose changes of the fully mechanized mining equipment in the data and knowledge dimensions is obtained. A hybrid driving system is further established to achieve accurate detection of the positioning and attitude determination of the fully mechanized mining equipment and its dynamic pose during overall advancement. Therefore, this application can achieve real-time straightness detection and improve detection accuracy. Attached Figure Description

[0072] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0073] Figure 1 is a structural framework diagram of a digital twin detection system for the spatial straightness of a fully mechanized mining face corresponding to one or more embodiments;

[0074] Figure 2 is a schematic diagram of the hybrid drive process during straightness detection;

[0075] Figure 3 is the architecture diagram of Hybrid Driver Module 1;

[0076] Figure 4 is the architecture diagram of Hybrid Driver Module 2;

[0077] Figure 5 shows the architecture of Hybrid Driver Module 3;

[0078] Figure 6 is the architecture diagram of Hybrid Driver Module 4;

[0079] Figure 7 is a schematic diagram of a fully mechanized mining face;

[0080] Figure 8 is a schematic diagram of the first analysis results of the straightness information of the scraper conveyor;

[0081] Figure 9 is a schematic diagram of the second analysis results of the straightness information of the scraper conveyor;

[0082] Figure 10 is a schematic diagram of the first analysis results of the straightness of the hydraulic support group;

[0083] Figure 11 is a schematic diagram of the second analysis results of the straightness of the hydraulic support group. Detailed Implementation

[0084] To ensure safe production and provide guidance for future production processes, this invention proposes a digital twin detection system and method for spatial straightness of fully mechanized mining faces. This is achieved by using virtual-real fusion to acquire and extrapolate the spatiotemporal trajectory of the fully mechanized mining face in real time, thereby obtaining the straightness of the face in the next cut.

[0085] In one exemplary embodiment, a digital twin detection system for the spatial straightness of a fully mechanized mining face is provided. This system includes a knowledge-driven system, a data-driven system, and a hybrid-driven system. Both the knowledge-driven system and the data-driven system are connected to the hybrid-driven system.

[0086] The knowledge-driven system is used to analyze and process the coordinated motion of equipment based on spatial kinematics to obtain parameterized operation rule data. The motion mechanism is determined by analyzing the single-machine motion of the fully mechanized mining equipment and the positional relationship between the fully mechanized mining equipment. The parameterized operation rule data consists of equipment motion constraint parameters determined according to the coordinated operation rules between the fully mechanized mining equipment.

[0087] The knowledge-driven system is also used to simulate virtual physical relationships based on parameterized operation rule data, and to determine the mapping relationship between virtual and real equipment operation.

[0088] In one embodiment, the knowledge-driven system includes: a runtime mechanism analysis unit, a runtime rule parameterization unit, and a physical relationship simulation unit. The runtime mechanism analysis unit is connected to the runtime rule parameterization unit; the runtime rule parameterization unit is also connected to the physical relationship simulation unit.

[0089] The operation mechanism analysis unit is used to determine the operation mechanism of the fully mechanized mining equipment based on the individual machine motions and the positional relationships between the equipment. The operation rule parameterization unit is used to perform analytical processing of the equipment's coordinated motion based on spatial kinematics to obtain parameterized operation rule data.

[0090] The physical relationship simulation unit is used to perform virtual physical relationship simulation based on the physical engine and the stress conditions of the fully mechanized mining equipment, and to determine the mapping relationship between the virtual and real equipment operation according to parameterized operation rule data.

[0091] The data-driven system is used to determine real-time pose data based on mapping relationships and collected information data using a virtual simulation approach. The information data includes: environmental monitoring information data, equipment performance data, and pose sensing data during the production process.

[0092] The data-driven system is also used to perform pose calculations based on real-time pose data to obtain the calculated pose data.

[0093] Specifically, the data-driven system includes: a pose data modeling unit, a virtual detection unit, and a data processing unit; the pose data modeling unit and the virtual detection unit are coupled together; the data processing unit is connected to both the pose data modeling unit and the virtual detection unit.

[0094] The pose data modeling unit is used to collect initial information data and classify the initial information data to obtain information data.

[0095] The virtual detection unit is used to determine virtual sensors based on the detection principle of physical sensors and the set error; based on the virtual sensors, virtual detection of the fully mechanized mining equipment is performed according to the mapping relationship to obtain virtual data; and real-time pose data is determined based on virtual data and information data using a virtual simulation method.

[0096] The data processing unit is used to perform collaborative pose calculation on real-time pose data using multiple software programs to obtain the calculated pose data. The data processing unit is also used to transmit the calculated pose data to the hybrid drive system based on a data interaction active communication mechanism.

[0097] The hybrid drive system is used to perform inversion and correction processing based on the calculated pose data to obtain the straightness information of the equipment group; to determine the real-time straightness information of the working face based on the straightness information of the equipment group; and to determine the predicted straightness information of the working face based on the mapping relationship and pose similarity. The pose similarity is determined based on historical pose data, the historical advancement trajectory of the fully mechanized mining equipment, and the coupling relationship between the coal seam and the pose of the equipment during the advancement process.

[0098] The hybrid drive system is also used to perform straightness fusion processing based on the predicted straightness information and the real-time straightness information of the working face to obtain the straightness of the fully mechanized mining face; the straightness of the fully mechanized mining face is used to characterize the undulation of the laying trajectory of the fully mechanized mining equipment group during the advancement process.

[0099] Specifically, the hybrid drive system includes a static straightness detection unit and a dynamic straightness detection unit.

[0100] The static straightness detection unit is connected to the knowledge-driven system, the data-driven system, and the dynamic straightness detection unit, respectively; the dynamic straightness detection unit is also connected to the knowledge-driven system and the data-driven system, respectively.

[0101] The static straightness detection unit is used to invert the calculated coal mining machine pose to obtain the straightness information of the scraper conveyor in the fully mechanized mining equipment; and to determine the straightness information of the hydraulic support group in the fully mechanized mining equipment based on the point cloud information; the point cloud data is obtained by non-contact measurement method, through the acquisition of the positioning and pose information of the hydraulic support group by three-dimensional lidar.

[0102] The static straightness detection unit is also used to correct the straightness information of the scraper conveyor using point cloud information to obtain the corrected straightness information of the scraper conveyor; and to determine the straightness information of the working face based on the straightness information of the equipment group and the parameterized operation rule data; the straightness information of the equipment group includes: the straightness information of the hydraulic support group and the corrected straightness information of the scraper conveyor.

[0103] The dynamic straightness detection unit is used to determine the predicted straightness information of the working face based on the mapping relationship and pose similarity, and to perform straightness fusion processing based on the predicted straightness information and the real-time straightness information of the working face to obtain the straightness of the fully mechanized mining face.

[0104] In practical applications, as shown in Figure 1, the system provided in this application consists of three subsystems: a knowledge-driven system, a data-driven system, and a hybrid-driven system. Specifically, it includes a mechanism analysis unit, a running rule parameterization unit, a physical relationship simulation unit, a pose data modeling unit, a virtual detection unit, a data processing unit, a static straightness detection unit, and a dynamic straightness detection unit.

[0105] The knowledge-driven system aims to achieve "physical integration" of fully mechanized mining equipment operation in virtual scenarios. Based on the analysis of equipment motion mechanisms, it parameterizes the operation rules of complex industrial systems to realize the execution and updating of motion rules based on parameter changes. Furthermore, it establishes a mapping relationship between virtual and real equipment in digital twin operation to ensure the consistency of the physical relationship between virtual and real equipment movements.

[0106] The data-driven system dynamically classifies data generated during the operation and maintenance of complex industrial systems through a pose data modeling unit. This enables the extraction of data relationships, the classification of monitoring targets, and the acquisition of key pose data. Based on a virtual detection unit, it achieves comprehensive detection through virtual-real fusion, ensuring high detection accuracy and high-reliability monitoring. The data processing unit supports real-time data transmission and fusion computation processing of the above processes, with high computational efficiency ensuring efficient data transmission and processing.

[0107] Based on the problems existing in the static pose detection (static straightness detection unit) and dynamic pose detection (dynamic straightness detection unit) of the equipment, the hybrid drive system, under the premise of realizing equipment operation virtualization, utilizes virtual detection technology and, with the support of multi-source sensor information fusion and virtual-real fusion inference technology, to realize static pose detection of the equipment during operation; and uses a knowledge-driven system to extract the similarity pattern of pose changes during equipment movement and the temporal characteristics in historical pose data to reliably infer and predict the actions of the next process.

[0108] The operation mechanism analysis unit can analyze the individual motion of the fully mechanized mining equipment and the positional relationships between equipment to obtain the motion mechanism. The operation rule parameterization unit establishes the equipment motion constraint relationship according to the collaborative operation rules between equipment, and realizes the parameterization of motion rules.

[0109] The physical relationship simulation unit can simulate virtual physical relationships based on a physics engine after completing the force analysis of the equipment; the pose data modeling unit can classify and predict data based on the digital twin industrial system, using real-time sensor data and environmental monitoring information generated in actual production through virtual-real fusion technology and deep learning technology.

[0110] The virtual detection unit establishes a virtual sensor based on the detection principle of physical sensors and the set error. Through virtual-real fusion, virtual detection is achieved, and reliable real-time posture data of the equipment is obtained through virtual simulation.

[0111] The data processing unit performs data-driven pose calculations through multi-software collaboration based on a data interaction active communication mechanism, and transmits real-time pose data in the form of data element interfaces.

[0112] The hybrid drive system comprises four hybrid drive modules. Hybrid drive modules one, two, and three primarily ensure the reliable acquisition of the pose of the hydraulic support and scraper conveyor, control the straightness detection of the t-th cut on the working face, and provide data support for hybrid drive module four. Hybrid drive module four, under the constraints of mining technology and spatiotemporal characteristics, obtains the straightness information of the t+1 cut working face through depth prediction and fusion. Specifically, the static straightness detection unit includes hybrid drive modules one, two, and three; the dynamic straightness detection unit includes hybrid drive module four.

[0113] Taking the coal mining machine as an example, the hybrid drive module one utilizes the contact relationship between the coal mining machine and the scraper conveyor, and uses digital twin technology to reliably detect the straightness of the scraper conveyor. Hybrid drive module two, based on the changing characteristics of the relative positional relationship between the hydraulic support and the coal mining machine during the advancement process, obtains the initial straightness information of the working face by mounting a three-dimensional LiDAR on the coal mining machine body, thus initially solving the problem of missing positioning and attitude information for the hydraulic support in working face straightness detection. Hybrid drive module three, based on the connection relationship between the hydraulic support and the scraper conveyor, uses the scraper conveyor's pose fusion connection relationship and sensor data to deduce the pose of the hydraulic support.

[0114] Hybrid drive module four, based on the similarity of equipment pose changes during the advance of the fully mechanized mining face, uses two levels: straightness prediction based on historical data and straightness prediction based on advance similarity. As the real-time pose data is updated, Kalman filtering is used to fuse the straightness information.

[0115] When performing static straightness measurements, the static straightness detection unit needs to deduce the equipment's posture by comprehensively considering the relationships between equipment and the available sensor information.

[0116] Specifically, with the support of the knowledge-driven system and the data-driven system, the straightness information of the scraper conveyor is obtained by inverting the reliable coal mining machine pose data through the hybrid drive module one. With the support of the hybrid drive module two, a non-contact measurement method is introduced to correct the straightness information of the scraper conveyor through the obtained point cloud information. Furthermore, the straightness information of the hydraulic support group based on the three-dimensional point cloud information can be obtained. The straightness information of the equipment group obtained by the hybrid drive modules one and two is transmitted to the hybrid drive module three. Through real-time analysis of the equipment connection relationship, the reliable straightness information of the hydraulic support group, that is, the straightness information of the working face, is finally obtained.

[0117] When performing dynamic straightness measurement, the dynamic straightness detection unit needs to predict the overall advancement posture trajectory by comprehensively considering the patterns in the historical advancement trajectory of the fully mechanized mining equipment and the similarity of posture changes during the overall advancement of the working face. First, based on the collaborative advancement mechanism among fully mechanized mining equipment and the constraints of the comprehensive mining process, the similarity of equipment posture in several adjacent cutting cycles is analyzed. A virtual simulation environment based on coal seam geological exploration information is created to analyze the similarity of equipment posture during overall advancement. Integrating the action execution limit constraints and posture similarity during advancement, a dynamic straightness prediction model is established. Second, based on the historical posture data of the working face, data classification and prediction of the posture of the t+1 cutter are achieved through modeling. Combined with virtual detection technology, dynamic straightness prediction is realized. Finally, through data processing technology, the prediction model and prediction results are deeply integrated to complete the dynamic straightness detection and obtain the straightness of the fully mechanized mining face.

[0118] The digital twin detection system for the straightness of the fully mechanized mining face provided in this embodiment can be applied to the scenario of determining the health status of equipment. The scenario of determining the health status includes the straightness detection and status determination of the fully mechanized mining face; the content provided in this embodiment pertains to the straightness detection stage of the fully mechanized mining face. Specifically, in the process of determining the health status, during the operation and maintenance of the fully mechanized mining face production system, the health status of the equipment can be determined based on the results of the straightness detection of the fully mechanized mining face—that is, based on the comparison results of the straightness of the fully mechanized mining face with the set straightness—and then it can be determined whether to prompt the staff to perform maintenance on the fully mechanized mining equipment group.

[0119] In an exemplary embodiment, a digital twin detection method for the spatial straightness of a fully mechanized mining face is provided. This method is implemented using a digital twin detection system for the spatial straightness of a fully mechanized mining face. The various components of the system can be configured via software, firmware, hardware, or a combination thereof. Specific means or methods of configuration are well known to those skilled in the art and will not be elaborated upon here. When implemented via software or firmware, programs constituting the software are installed from a storage medium or network onto a computer with a dedicated hardware architecture. When various programs are installed, the computer can perform various functions. The method mentioned in this application can be integrated onto a computer server or placed on the cloud or other servers for execution via a computer.

[0120] The digital twin detection method for the spatial straightness of fully mechanized mining faces includes:

[0121] Step 100: Acquire information data. Information data includes: environmental monitoring information data, equipment performance data, and pose sensing data during the production process.

[0122] Step 200: Using virtual simulation, determine real-time pose data based on mapping relationships and information data. The mapping relationship is a one-to-one correspondence between the actual working face and the virtual physical relationship established through virtual physical relationship simulation based on parameterized operation rule data; the parameterized operation rule data is obtained through analytical processing of equipment coordinated motion based on spatial kinematics and motion mechanisms; the motion mechanism is determined through analytical analysis of the individual motion of the fully mechanized mining equipment and the pose correlation between fully mechanized mining equipment; the parameterized operation rule data consists of equipment motion constraint parameters determined according to the coordinated operation rules between fully mechanized mining equipment.

[0123] Step 300: Perform pose calculation processing based on real-time pose data to obtain the calculated pose data.

[0124] Step 400: Perform inversion and correction processing based on the calculated pose data to obtain the straightness information of the equipment group.

[0125] Step 500: Determine the real-time straightness information of the working face based on the straightness information of the equipment group.

[0126] Step 600: Based on the mapping relationship and pose similarity, determine the predicted straightness information of the working face according to the straightness information of the working face. The pose similarity is determined based on historical pose data, the historical advance trajectory of the fully mechanized mining equipment, and the coupling relationship between the coal seam and the equipment pose during the advance process;

[0127] Step 700: Based on the predicted straightness information and the real-time straightness information of the working face, perform straightness fusion processing to obtain the straightness of the fully mechanized mining face. The straightness of the fully mechanized mining face is used to characterize the undulation of the laying trajectory of the fully mechanized mining equipment group during the advancement process.

[0128] In one embodiment, a virtual simulation is used to determine real-time pose data based on mapping relationships and information data, specifically including:

[0129] Based on the detection principle of physical sensors and the set error, a virtual sensor is determined.

[0130] Based on virtual sensors, virtual detection of fully mechanized mining equipment is performed according to the mapping relationship to obtain virtual sensing data.

[0131] The real-time pose data is determined by using virtual simulation based on virtual sensing data and information data.

[0132] As an optional implementation method, the solved pose data is inverted and corrected to obtain the straightness information of the equipment group, specifically including:

[0133] The straightness information of the scraper conveyor in the fully mechanized mining equipment is obtained by inversion based on the calculated coal mining machine posture data.

[0134] The straightness information of the hydraulic support group in the fully mechanized mining equipment is determined based on the point cloud information. The point cloud data is obtained by collecting the positioning and attitude information of the hydraulic support group using a three-dimensional lidar through a non-contact measurement method.

[0135] The straightness information of the scraper conveyor is corrected using point cloud information to obtain corrected straightness information.

[0136] Based on the straightness information of the equipment group and the parameterized operation rule data, the straightness information of the working face is determined; the straightness information of the equipment group includes: the straightness information and the corrected straightness information of the hydraulic support group.

[0137] In one embodiment, the straightness of the fully mechanized mining face is obtained by performing straightness fusion processing using a Kalman filter method based on the predicted straightness information and the real-time straightness information of the working face.

[0138] The digital twin detection method for the spatial straightness of fully mechanized mining faces mentioned in this application also includes: using a data interaction active communication mechanism for data transmission. This data interaction active communication mechanism transmits data in the form of a data element interface.

[0139] Based on the characteristics of the mining process of the fully mechanized mining equipment face and the collaborative operation characteristics between equipment, the entire driving process also adopts the method of "distributed pose extrapolation and nested hybrid driving modules".

[0140] The hybrid drive process for straightness detection is shown in Figure 2. The architecture diagram of hybrid drive module one is shown in Figure 3. The working process of the data-driven system is as follows: Through the virtual detection unit, based on the establishment of digital twins of physical sensors such as infrared sensors, shaft encoders, and SINS and the dynamic compensation of errors, a virtual-real mapping between the virtual sensors and their corresponding installation positions is established to realize virtual detection of the cutting operation process; through the data processing unit, multi-software collaborative calculations are performed to obtain reliable cutting information P of the t-cutter coal mining machine. s And it is represented by the data element interface OM1 / M_ / 01-00 / 01 / x,y,z,θ,ψ. The values ​​are transmitted to other modules in the form of x, y, z, θ, ψ. These are the x, y, and z coordinates of the coal mining machine, as well as the pitch angle, yaw angle, and roll angle.

[0141] The virtual pose detection method for the coal mining machine corresponding to Hybrid Drive Module 1 is as follows:

[0142] Input: Real-time detection value P s a

[0143] Output: Virtual detection value P s v

[0144] 1. Sensor measurement method: Mp{infrared sensor, shaft encoder, SINS} → Mp{Mpr, Mpo, Mps};

[0145] 2. Actual sensor measurements: Ra{Rar, Rao, Ras}; Virtual sensor measurements: Rv{Rvr, Rvo, Rvs};

[0146] 3. Virtual detection method: Mvp{Mvr, Mvo, Mvs} → Raycast class, Transform component;

[0147] 4. Initialization: Mvp{Mvr, Mvo, Mvs} ← Virtual detection unit;

[0148] 5 / / Virtual Detection Process:

[0149] 6 if Ra update

[0150] 7 Mvi←Rvi; Rv←Mvi; / / i=r, o, s

[0151] 8 P s v ←Data processing technology (Rv); / / Multi-software collaborative computing for multi-sensor data fusion

[0152] 9 OM1 / M_ / 01-00 / 01 / x, y, z, θ, ψ, / / Output in metadata interface format

[0153] 10 end if

[0154] 11 return P s v

[0155] At this point, the driving process corresponding to the knowledge-driven system is: establishing the scraper conveyor pose inversion model. The coal mining machine uses the scraper conveyor as the track for cutting operations. Based on the direct contact and structural relationship between the two, using Rodriguez parameters and spatial relationship transformation methods, the pose change law of the scraper conveyor based on the coal mining machine's pose is established. Furthermore, based on the data characteristics when the coal mining machine passes through two adjacent intermediate troughs and the structural characteristics of the intermediate troughs, the boundary points of the intermediate troughs are calculated, and the pose parameter P can be obtained. a The theoretical model is established; based on the constraint relationship between two adjacent central slots during propulsion, a hierarchical combination of central slots based on the cutting direction is established in Unity3D, and the constraints between equipment actions are simulated by setting the angle and displacement limit values ​​of CharacterJoint.

[0156] The straightness detection process supported by the hybrid drive system is as follows: Based on the actual motion characteristics of the coal mining machine and the data requirements for trajectory inversion, auxiliary calculation points are marked in Unity3D, including the origin of the coordinate system where the sensors are installed, the slipper detection points, and the coal mining machine pose positioning points; the obtained pose calculation model is then applied virtually, and through the data processing unit, Unity3D and Matlab interact across platforms to perform real-time collaborative calculations, realizing the pose P of the T-blade scraper conveyor. a Real-time inversion is used to calculate the straightness of the scraper conveyor, and the pose of each central trough is updated using the Transform component. As the cutting direction changes, virtual constraints are added to drive the parent-child relationship direction update, achieving real-time virtual reconstruction of the scraper conveyor pose, which is then expressed as data element interfaces OM1 / M_ / 03-0i / 01 / x,y,z,θ,ψ. Transmitted in the form of .

[0157] The architecture diagram of Hybrid Drive Module 2 is shown in Figure 4. The working process of the data-driven system is as follows: By mounting a 3D LiDAR on the coal mining machine body, the sensing information of the coal mining machine is enhanced, obtaining the positioning information of the coal mining machine and the attitude information of the hydraulic support column; based on the digital twin concept, a virtual 3D LiDAR is added. According to the radar positioning information, the A-LOAM algorithm is used to calculate the position information of the coal mining machine, and the meta-interface data in Hybrid Drive Module 1 is fused to optimize the pose information of the t-th cutter coal mining machine; through a virtual detection unit, the real-time obtained support point cloud information is matched with the virtual point cloud information, and the hydraulic support attitude sensing data is fused to achieve complete hydraulic support attitude A-LOAM. h The detection outputs the attitude values ​​in the form of meta-interface data OM2 / M_ / 02-0i / 02 / α,β,γ,δ,h / #, where α,β,γ,δ,h are the rear link tilt angle, front link tilt angle, shield beam tilt angle, top beam tilt angle, and support height of the hydraulic support, respectively.

[0158] The virtual attitude detection method for the hydraulic support corresponding to the hybrid drive module is as follows:

[0159] Input: 3D point cloud, hydraulic support attitude sensing information (Ash), coal mining machine pose data (PS1)

[0160] Output: Hydraulic support attitude data A h

[0161] 1. Sensor virtual measurement value Avh;

[0162] 2Avh ← Virtual detection unit (Ash);

[0163] 3 / / Virtual Detection Process:

[0164] 4 if Ah update

[0165] 5. Coal mining machine positioning information Ps(x, y, z) ← A-LOAM algorithm solves 3D point cloud information

[0166] 6. Ps ← Data processing techniques (Ps(x, y, z) + Ps1); / / Using data processing techniques to perform data fusion.

[0167] 7. Column attitude information A ← 3D point cloud information; / / i = r, o, s

[0168] 8 A h ←A-based virtual hydraulic support attitude matching;

[0169] 9 OM2 / M_ / 02-0i / 02 / α,β,γ,δ,h / #←A h / / Output in the form of metadata interface

[0170] 10 end if

[0171] 11 return Ps, A h

[0172] As shown in Figure 4, the working process of the knowledge-driven system is as follows: Based on the scraper conveyor pose inversion model, the straightness of the scraper conveyor is corrected; the floating connection relationship between the hydraulic support and the scraper conveyor is analyzed, and the motion law of the connected mechanism (floating connection mechanism) is determined using the spatial kinematics of the industrial robot. Analysis was performed, and a pose derivation model of the hydraulic support based on the scraper conveyor's pose was established using forward kinematics; By integrating it into the Unity3D underlying layer and supporting real-time resolution of floating connections, a simulation model of the pose deduction of the hydraulic support based on the pose of the scraper conveyor was established.

[0173] As shown in Figure 4, the straightness of the t-cutter working face is obtained under the hybrid drive module: Under the drive of the reliable coal mining machine pose Ps, the pose of the scraper conveyor is inverted in real time to obtain the corrected straightness information Pa of the scraper conveyor. The poses of each middle trough are then expressed as meta-interface data OM2 / _M / 03-0i / 00 / x,y,z,θ,ψ. The output is in the form of point cloud data. Based on the obtained point cloud information, the relative pose relationship between the hydraulic support and the scraper conveyor is extracted. Based on the inversion, the pose information of the scraper conveyor is obtained. The motion law of the floating connection mechanism is used to drive the calculation of the hydraulic support pose. The theoretical calculation results of the hydraulic support pose are fused with the pose results obtained by matching the point cloud to obtain the pose information of the hydraulic support. As the coal mining machine advances, the straightness information of the hydraulic support group is obtained. The pose of each support is output as meta-interface data OM2 / M_ / 02-0i / 00 / x,y,z,θ,ψ. Output in the form of .

[0174] At this point, the corresponding scraper conveyor pose reconstruction method is as follows:

[0175] Input: Coal mining machine position parameter value P s Hydraulic support posture information A h Point cloud information

[0176] Output: Hydraulic support position P h

[0177] 1 if P s renew

[0178] 2 P a ←Based on P s Value multi-software collaborative computing inversion; %P a Correction

[0179] 3 ΔP← Point cloud information; % Extract the relative pose relationship between the hydraulic support and the scraper conveyor.

[0180] 4 Theoretical derivation of the pose of the hydraulic support

[0181] 5 P h ←P' h +A h ;% Fusion of inference data and matching data

[0182] 6 end if

[0183] 7 OM2 / M_ / 02-0i / 00 / x,y,z,θ,ψ, The pose of the support is transmitted via the meta interface.

[0184] 8 return P h

[0185] The architecture diagram of Hybrid Drive Module 3 is shown in Figure 5. The working process in the data-driven system is as follows: During propulsion, the propulsion stroke information (PI) can be obtained through the electro-hydraulic controller of the hydraulic support. However, due to the execution error of the propulsion mechanism and the spatial motion characteristics, this information deviates from the actual value. Two infrared ranging sensors (IRS) are symmetrically installed on the upper surface of the hydraulic support base. The relative position and relative yaw angle between the hydraulic support and the scraper conveyor are indirectly measured through the two ranging values. Through data processing technology, Unity3D reads the detection data in real time and achieves dynamic compensation of sensor errors through virtual-real fusion. Accurate stroke data and ranging information D(d,d1,d2) are obtained through the virtual detection unit.

[0186] As shown in Figure 5, the working process of the knowledge-driven system is as follows: based on the structural characteristics and motion constraints between the hydraulic support and the scraper conveyor, a motion adjustment strategy is set for the simulation process. and the rules for updating pitch, roll, and yaw angles. To ensure that there is no motion interference between equipment during the simulation, and that the movement distance and attitude angle changes meet safety production requirements; to explain the operating mechanism. With rule T h By performing fusion and pose deduction, and based on obtaining the motion law of the floating connection mechanism, in A h During execution, the motion of the floating connection mechanism is analyzed in real time, and a pose deduction model of the hydraulic support is established through forward kinematics.

[0187] As shown in Figure 5, the two-way virtual simulation of straightness establishes iterative conditions for comparing virtual and real travel information and ranging information based on the measurement accuracy of the sensors. On the basis of obtaining reliable detection data, the relative yaw angle between the two pieces of equipment is obtained based on the two ranging information. The simulation is performed under the constraints of the pitch and roll angle limits in the mid-slot. Simultaneously, the motion of the floating connection mechanism is analyzed in real time until the virtual and real information values ​​are consistent, thus obtaining the pose parameters P of the hydraulic support. h′ The simulation results are fused with the support pose information obtained from the hybrid drive module 2 to obtain reliable parameters P. h As the cutting operation progresses, the straightness of the hydraulic support group of the t-blade is obtained, and it is expressed as data element interface OM3 / M_ / 02-0i / 00 / x,y,z,θ,ψ. Transmitted in the form of .

[0188] The specific deduction method is as follows:

[0189] Input: D(d,d1,d2),OM2 / M_ / 03-0i / 00 / x,y,z,θ,ψ, OM2 / M_ / 02-0i / 00 / x,y,z,θ,ψ,

[0190] Output: Hydraulic support position P h

[0191] 1. Deducing sensor information D v (d v ,d 1v ,d 2v )

[0192] 2. Initialization: P a (x a ,y a ,z a )←D(d,d1,d2)+P a % Location Update

[0193] 3 if D! = D v

[0194] 4 % Base posture update range

[0195] 5 % Base posture iteration

[0196] 6 % Hydraulic support position and posture real-time calculation

[0197] 7 end if

[0198] 8 else if

[0199] 9 Ph1 ←OM2 / M_ / 02-0i / 00 / x,y,z,θ,ψ , % Active receiving module 2 transmits hydraulic support pose data

[0200] 10 P h ←P h1 +P h2 % Hydraulic support pose data fusion and correction

[0201] 11 end

[0202] 12 return P h

[0203] The architecture diagram of Hybrid Drive Module 4 is shown in Figure 6. The working process of the data-driven system is as follows: 1) Active acquisition of pose data: The system actively acquires the equipment pose dataset (d, d1, d2, P) from the meta-interface data output by Hybrid Drive Modules 1 to 3 through a data interaction active communication mechanism. h ,P a ,P s A s ); 2) T+1 tool straightness modeling: Historical straightness information (P) is obtained through equipment pose data classification. h ,P a ) t≤t Historical truncation information (P) s A s ) t≤t Real-time travel distance measurement information (d, d1, d2) t Based on historical information on the straightness of the longwall face, an LSTM neural network is used to predict the straightness information of the face at the (t+1)th cut; in a virtual environment, based on the predicted information and historical cutting information P s Establish the pose adjustment range of the two equipments, using (d,d1,d2) as prior information. Drive the pose adjustment of the scraper conveyor and the hydraulic support group through the difference between virtual and real information until the virtual and real information are consistent, and complete the correction of the predicted straightness of the t+1 cutter.

[0204] The method for determining the straightness information of the predicted working face is as follows:

[0205] Input: Equipment pose dataset DM(d,d1,d2,P) h ,P a ,P s A s )

[0206] Output: Straightness parameters of the fully mechanized mining face

[0207] 1. Initialization: D(d,d1,d2)+(P)h ,P a)t≤t +(P s A s ) t≤t ←DM(d,d1,d2,P h ,P a ,P s A s );

[0208] 2 Virtual travel distance measurement information D v (d v ,d 1v ,d 2v )

[0209] 3 % working face straightness prediction

[0210] 4 % pose adjustment range determined

[0211] 5 if D! = D v

[0212] 6. Virtual Equipment Position Adjustment ←

[0213] 7 end if

[0214] 8

[0215] 9 returns

[0216] As shown in Figure 6, the working process of the knowledge-driven system is as follows: 1) Obtaining the similarity law of equipment posture changes: Based on the cutting characteristics and spatiotemporal characteristics of the coal mining machine during the advancement of the fully mechanized mining face, combined with the influence of production process, the similarity analysis of equipment posture within several adjacent cutting cycles can be performed; by integrating coal seam geological exploration information and establishing a virtual simulation scene of the fully mechanized mining face in Unity3D, the similarity analysis of posture changes during the overall advancement is extended to obtain the similarity law of posture changes of hydraulic support and scraper conveyor during the overall advancement process; 2) Straightness theory prediction model based on mining operation rules: After obtaining Based on the constraints of the limit value of the pushing stroke when the integrated hydraulic support is pushed and the characteristic that it can only move forward and not backward when the support is moved, a straightness prediction model of the equipment is established based on the straightness information of the working surface of the t-cutter, which can realize the theoretical deduction of the straightness of the working surface of the t+1-cutter.

[0217] As shown in Figure 6, considering the influence of the coal mining machine's cutting trajectory on the equipment's attitude and the influence of its shifting characteristics on its position, the Kalman filter method is used to fuse the theoretical deduction results and data prediction results of the straightness of the t+1 cutter face, ultimately obtaining the straightness information of the t+1 cutter face.

[0218] This study focuses on a high-depth fully mechanized mining face. The face has a burial depth of 364m–417m, averaging 390.5m, a strike length of 462m, and an dip length of 220m. It exhibits small fold structures, causing variations in the coal seam dip angle. This results in some undulations in the height of the two roadways and localized overburden and underburden mining, making equipment management difficult. The average angle of the face is 5.4°, with a maximum angle of 10°. The face utilizes an MG750 / 1900-GWD coal mining machine, ZY8000 / 26 / 56D hydraulic supports, and SGZ-1000 / 1400 scraper conveyors for coal transport. The center-to-center distance between the supports is 1.75m. During the initial mining phase of the 9711 face, bottom coal was retained, with a thicker layer at the top and a thinner layer at the bottom. The height difference between the upper and lower roadways was small, with a difference of 20.7m between the return and intake airways, as shown in Figure 7.

[0219] Experiments were conducted on a large-scale working face layout. Based on the verification of each hybrid drive module, the straightness detection method using a knowledge-and-data hybrid drive was tested on the equipment group corresponding to hydraulic supports 95-38. The fully mechanized mining face was continuously advanced for 8 cutting cycles, and the straightness of the working face was monitored in real time during the 9th cutting cycle. The straightness analysis results of the fully mechanized mining face after applying this system are shown in Figures 8-11.

[0220] The straightness of the working surface of the t+1 cutter is detected by the combined action of hybrid drive modules one through four, and the straightness information of the scraper conveyor is shown in Figures 8 and 9. As can be seen from Figures 8 and 9, the positioning error of the scraper conveyor is within 0.4 dm, and the attitude error is within 0.7°, indicating high positioning and attitude accuracy. When detecting the straightness of the hydraulic support group, Figures 10 and 11 show that the positioning error of the hydraulic support is within 0.3 dm, and the attitude error is within 0.9°, indicating high positioning and attitude accuracy.

Claims

1. A digital twin detection system for the spatial straightness of a fully mechanized mining face, characterized in that, The digital twin detection system for the spatial straightness of the fully mechanized mining face includes: a knowledge-driven system, a data-driven system, and a hybrid-driven system. Both the knowledge-driven system and the data-driven system are connected to the hybrid driving system; The knowledge-driven system is used for: Based on spatial kinematics, the motion mechanism of the equipment is obtained, and the coordinated motion of the equipment is analyzed and processed to obtain parameterized operation rule data. The motion mechanism is determined by analyzing the single-machine motion of the fully mechanized mining equipment and the positional relationship between the fully mechanized mining equipment. The parameterized operation rule data is the equipment motion constraint parameters determined according to the coordinated operation rules between the fully mechanized mining equipment. Based on the parameterized operation rule data, a virtual physical relationship simulation is performed to determine the mapping relationship between the operation of virtual and real equipment; The data-driven system is used for: The real-time pose data is determined by using a virtual simulation method based on the mapping relationship and the collected information data; the information data includes: environmental monitoring information data, equipment performance data, and pose sensing data during the production process; The pose data is calculated based on the real-time pose data to obtain the calculated pose data. The hybrid drive system is used for: Based on the calculated pose data, inversion and correction processes are performed to obtain the straightness information of the equipment group. The real-time straightness information of the working face is determined based on the straightness information of the equipment group. Based on the mapping relationship and pose similarity, the predicted straightness information of the working face is determined according to the straightness information of the working face; the pose similarity is determined based on historical pose data, the historical advance trajectory of the fully mechanized mining equipment, and the coupling relationship between the coal seam and the equipment pose during the advance process; Based on the predicted straightness information of the working face and the real-time straightness information of the working face, straightness is determined. The straightness of the fully mechanized mining face is obtained through degree fusion processing; the straightness of the fully mechanized mining face is used to characterize the undulation of the laying trajectory of the fully mechanized mining equipment group during the advancement process.

2. The digital twin detection system for the spatial straightness of a fully mechanized mining face according to claim 1, characterized in that, The knowledge-driven system includes: a runtime mechanism analysis unit, a runtime rule parameterization unit, and a physical relationship simulation unit; The operation mechanism analysis unit is connected to the operation rule parameterization unit; the operation rule parameterization unit is also connected to the physical relationship simulation unit; The operation mechanism analysis unit is used to determine the operation mechanism of the fully mechanized mining equipment based on the individual machine motion of the fully mechanized mining equipment and the positional relationship between the fully mechanized mining equipment. The operation rule parameterization unit is used to analyze and process the equipment's coordinated motion based on spatial kinematics to obtain parameterized operation rule data. The physical relationship simulation unit is used to perform virtual physical relationship simulation based on the physical engine and the stress conditions of the fully mechanized mining equipment, and to determine the mapping relationship between the virtual and real equipment operation according to the parameterized operation rule data.

3. The digital twin detection system for the spatial straightness of a fully mechanized mining face according to claim 1, characterized in that, The data-driven system includes: a pose data modeling unit, a virtual detection unit, and a data processing unit; The pose data modeling unit is coupled to the virtual detection unit; the data processing unit is connected to both the pose data modeling unit and the virtual detection unit. The pose data modeling unit is used to collect initial information data and classify the initial information data to obtain information data; The virtual detection unit is used for: Based on the detection principle of physical sensors and the set error, a virtual sensor is determined; Based on the virtual sensor, the fully mechanized mining equipment is virtually inspected according to the mapping relationship to obtain... Virtual sensor data; Real-time pose data is determined based on the virtual sensing data and the information data using a virtual simulation method. The data processing unit is used to perform collaborative pose calculation on the real-time pose data using multidisciplinary software to obtain the calculated pose data. The data processing unit is also used to transmit the calculated pose data to each unit and the hybrid driving system based on the active communication mechanism of data interaction.

4. The digital twin detection system for the spatial straightness of a fully mechanized mining face according to claim 1, characterized in that, The hybrid drive system includes: a static straightness detection unit and a dynamic straightness detection unit; The static straightness detection unit is connected to the knowledge-driven system, the data-driven system, and the dynamic straightness detection unit, respectively; the dynamic straightness detection unit is also connected to the knowledge-driven system and the data-driven system, respectively. The static straightness detection unit is used for: The straightness information of the scraper conveyor in the fully mechanized mining equipment is obtained by inversion based on the calculated coal mining machine posture data. The straightness information of the hydraulic support group in the fully mechanized mining equipment is determined based on the point cloud information; the point cloud data is obtained by non-contact measurement method, which is to collect the positioning and attitude information of the hydraulic support group through three-dimensional lidar. The straightness information of the scraper conveyor is corrected using point cloud information to obtain the corrected straightness information of the scraper conveyor. Based on the straightness information of the equipment group and the parameterized operation rule data, the straightness information of the working face is determined; the straightness information of the equipment group includes: the straightness information of the hydraulic support group and the corrected straightness information of the scraper conveyor; The dynamic straightness detection unit is used for: Based on the mapping relationship and the pose similarity, the predicted straightness information of the working surface is determined according to the straightness information of the working surface; Based on the predicted straightness information of the working face and the real-time straightness information of the working face, straightness fusion processing is performed to obtain the straightness of the fully mechanized mining working face.

5. A digital twin detection method for the spatial straightness of a fully mechanized mining face, characterized in that, The digital twin detection method for the spatial straightness of the fully mechanized mining face is implemented using the digital twin detection system for the spatial straightness of the fully mechanized mining face as described in any one of claims 1-4; the digital twin detection method for the spatial straightness of the fully mechanized mining face includes: Acquire information data; the information data includes: environmental monitoring information data, equipment performance data, and posture sensing data during the production process; A virtual simulation method is used to determine real-time pose data based on the mapping relationship and the information data. The mapping relationship is a one-to-one correspondence between the actual working face and the virtual physical relationship established by the parameterized operation rule data. The parameterized operation rule data is obtained by analyzing and processing the equipment cooperative motion based on spatial kinematics and motion mechanism. The motion mechanism is determined by analyzing and determining the single-machine motion of the fully mechanized mining equipment and the pose correlation between fully mechanized mining equipment. The parameterized operation rule data consists of equipment motion constraint parameters determined according to the cooperative operation rules between fully mechanized mining equipment. The pose data is calculated based on the real-time pose data to obtain the calculated pose data. Based on the calculated pose data, inversion and correction processes are performed to obtain the straightness information of the equipment group. The real-time straightness information of the working face is determined based on the straightness information of the equipment group. Based on the mapping relationship and pose similarity, the predicted straightness information of the working face is determined according to the straightness information of the working face; the pose similarity is determined based on historical pose data, the historical advance trajectory of the fully mechanized mining equipment, and the coupling relationship between the coal seam and the equipment pose during the advance process; Based on the predicted straightness information of the working face and the real-time straightness information of the working face, straightness fusion processing is performed to obtain the straightness of the fully mechanized mining face; the straightness of the fully mechanized mining face is used to characterize the undulation of the laying trajectory of the fully mechanized mining equipment group during the advancement process.

6. The digital twin detection method for the spatial straightness of a fully mechanized mining face according to claim 5, characterized in that, Using a virtual simulation approach, real-time pose data is determined based on the mapping relationship and the aforementioned information data, specifically including: Based on the detection principle of physical sensors and the set error, a virtual sensor is determined; Based on the virtual sensor, virtual detection of the fully mechanized mining equipment is performed according to the mapping relationship to obtain virtual sensing data; Real-time pose data is determined by using virtual simulation based on the virtual sensing data and the information data.

7. The digital twin detection method for the spatial straightness of a fully mechanized mining face according to claim 5, characterized in that, Based on the calculated pose data, inversion and correction processes are performed to obtain the straightness information of the equipment group, specifically including: The straightness information of the scraper conveyor in the fully mechanized mining equipment is obtained by inversion based on the calculated coal mining machine posture. The straightness information of the hydraulic support group in the fully mechanized mining equipment is determined based on the point cloud information; the point cloud data is obtained by non-contact measurement method, which is to collect the positioning and attitude information of the hydraulic support group through three-dimensional lidar. The straightness information of the scraper conveyor is corrected using point cloud information to obtain corrected straightness information; Based on the straightness information of the equipment group and the parameterized operation rule data, the straightness information of the working face is determined; the straightness information of the equipment group includes: the straightness information and the corrected straightness information of the hydraulic support group.

8. The digital twin detection method for the spatial straightness of a fully mechanized mining face according to claim 5, characterized in that, Based on the predicted straightness information of the working face and the real-time straightness information of the working face, a Kalman filter method is used to perform straightness fusion processing to obtain the straightness of the fully mechanized mining working face.

9. The digital twin detection method for the spatial straightness of a fully mechanized mining face according to claim 5, characterized in that, The digital twin detection method for the spatial straightness of the fully mechanized mining face also includes: Data transmission is performed using a proactive data interaction communication mechanism.

10. The digital twin detection method for the spatial straightness of a fully mechanized mining face according to claim 9, characterized in that, The data interaction active communication mechanism transmits data in the form of a data element interface.

Images