Real-time state mapping method of anchor drilling machine based on digital twin

Abstract

The present invention belongs to the technical field of digital twin of integrated digging and anchoring machine, and discloses a real-time state mapping method of integrated digging and anchoring machine based on digital twin. The method realizes the real-time state mapping process of integrated digging and anchoring machine integrating the service environment by receiving the relevant data of the actual integrated digging and anchoring machine and combining the digital twin technology. First, the working state information of the integrated digging and anchoring machine is transmitted in real time through the data virtual-real mapping technology; then, based on the construction method of geometric and kinematic model library, the real-time motion simulation is performed; and the lifting, feeding, excavation and other behavior simulations of the integrated digging and anchoring machine are constructed through the finite state machine; the service environment mapping process is realized by combining fast Boolean operation; finally, the data is analyzed and diagnosed based on the artificial intelligence method; and the real-time state mapping of the integrated digging and anchoring machine is completed. The method can realize the remote real-time state mapping transmission and real-time online monitoring of the operation process of the integrated digging and anchoring machine integrating the service environment based on the synchronous mapping and real-time interaction of physical devices and virtual devices.

Description

Technical Field

[0001] The present invention belongs to the technical field of digital twin of an all-in-one drilling and anchoring machine, and in particular relates to a real-time state mapping method of an all-in-one drilling and anchoring machine based on digital twin. Background Art

[0002] The digging and anchoring machine is a kind of equipment used for rapid excavation of coal mine tunnels. Compared with traditional coal mining equipment, the digging and anchoring machine can realize excavation and anchoring on the same device at the same time. In recent years, with the increasing requirements for tunnel excavation equipment, the research and development and application of the digging and anchoring machine have gradually deepened. However, the current research and development of the digging and anchoring machine mainly focuses on equipment design and manufacturing optimization and process optimization. The problems such as accurate monitoring of the construction status and geological environment during the equipment operation and maintenance stage and real-time maintenance have not been effectively solved. The main problems are: (1) It is difficult to collect information from the mobile unit in real time during the production operation; (2) It is difficult to quickly analyze the large amount of collected data, resulting in data waste; (3) There is no effective information communication between the equipment and people. These problems have become the key constraints to overcome the technical difficulties such as jamming and damage of the digging and anchoring machine during its service.

[0003] Digital twin is a dynamic model that integrates virtual and real. It is completely consistent with the corresponding physical entity in the real world. It can accurately and timely reflect the characteristics, behavior, performance, etc. of the physical entity, and plays an important role in promoting the collaborative intelligence and information automation of engineering equipment. The relevant theories of digital twin are still in their infancy. In order to promote the implementation of digital twin, systematic and in-depth research is needed in digital twin modeling, information-physical integration, exchange and collaboration. Summary of the invention

[0004] The purpose of the present invention is to overcome the shortcomings of the prior art and propose a real-time status mapping method for an all-in-one digging and anchoring machine based on digital twins. Starting from the actual workflow of the all-in-one digging and anchoring machine, aiming at the problem that it is difficult to collect information of the mobile unit in real time during the production operation, it is proposed to use virtual-to-real mapping technology to transmit the working status information of the all-in-one digging and anchoring machine in real time; in view of the difficulty of rapid data analysis, it is proposed to combine the geometric and kinematic model library construction method, finite state machine and fast Boolean operation to realize the mapping process of the service environment; in view of the problem that it is difficult to form effective information communication between information and people, it is proposed to realize real-time diagnosis of the all-in-one digging and anchoring machine data based on artificial intelligence methods; and based on the above technologies, the state mapping of the remote real-time status of the all-in-one digging and anchoring machine and the real-time online monitoring of the operation process are realized.

[0005] The objective of the present invention is achieved through the following technical solutions:

[0006] A real-time state mapping method for an integrated anchoring and drilling machine based on digital twins comprises the following steps:

[0007] Step 1: Use data virtual-real mapping technology to transmit the working status information of the anchor drilling machine in real time;

[0008] Step 2: Based on the construction method of geometric and kinematic model library, real-time motion simulation is performed;

[0009] Step 3: Use the finite state machine to build the lifting, feeding, and excavation behavior simulation of the anchoring and digging machine;

[0010] Step 4: Combine fast Boolean operations to implement the service environment mapping process;

[0011] Step 5: Analyze and diagnose the data based on artificial intelligence methods; complete the real-time status mapping of the drilling and anchoring machine.

[0012] Furthermore, the data virtual-real mapping technology specifically includes the following steps:

[0013] Open the network communication port in the digital twin client of the integrated mining and anchoring machine, set the actual data acquisition and monitoring system of the integrated mining and anchoring machine as the server, listen and wait for the client's connection; during the connection process, the sensor acquisition program is blocked and not looped. When it is detected that the digital twin client connection is completed, the sensor acquisition data is transmitted to the server; at this time, two synchronization tasks will be created, including the actual integrated mining and anchoring machine continuing to execute the task and the twin data being completely transmitted to the digital twin client, and real-time status mapping will be performed in the digital twin client according to the sensor data; at the same time, data integrity judgment will be performed, and the next wave of data stream transmission will be carried out after data integrity is achieved.

[0014] Furthermore, the sensor data includes displacement sensor data, pressure sensor data, photoelectric sensor data and smoke sensor data.

[0015] Furthermore, when the client is too large, a distributed network configuration is performed to form a server cluster; three-way handshake and four-way wave are used for data communication: a timer is set to execute the data sending function at fixed intervals, and each time it needs to be received, it is stored in the asynchronous buffer area and waits for acceptance. When the previous message is received, the next message is sent; at this time, the asynchronous function is entered, and when the sending fails, a warning message is sent.

[0016] Furthermore, the method for constructing the geometric and kinematic model library in step 2 specifically includes the following steps:

[0017] The geometric model of the mining and anchoring machine is reconstructed and rendered. Before the actual mapping begins, the geometric model is reconstructed and rendered in the model rendering software 3dsMax to obtain a virtual entity of the mining and anchoring machine with geometric dimensions and rendering relationships. For the motion mechanism of the mining and anchoring machine, the kinematic characteristics of each mechanism are analyzed separately. According to the actual working conditions, kinematic pairs and drives are established for the mechanism in the kinematic simulation software. The changes of each kinematic pair of the mechanism and the change of the main drive are obtained through post-processing, and the relevant data are stored in the historical database. Combined with the actual sensor data, fast data driving is achieved through interpolation calculation.

[0018] Furthermore, the specific steps of geometric model reconstruction and rendering are as follows:

[0019] Step a: Create a geometric model of the anchoring machine in 3D modeling software, including geometric dimensions and parameter relationships;

[0020] Step b: Import 3dsmax, redefine the assembly relationship, reconstruct the model tree, and establish the corresponding assembly relationship;

[0021] Step c: Add, delete, and adjust the mesh, re-divide the redundant mesh nodes, and reduce the weight of the model;

[0022] Step d: Delete redundant parts, collapse or delete the redundant parts, and reduce unnecessary parts or features in the model;

[0023] Step e: Make surface materials, maps, textures, etc.

[0024] Furthermore, the kinematics simulation software is Admas, Creo, Ug or SolidWorks.

[0025] Furthermore, the finite state machine in step 3 constructs the lifting, feeding and digging behavior simulation of the anchoring and digging machine, including the following specific steps: after a given sensor data input, the sensor data is matched with a certain state, and the anchoring and digging machine is converted to the indicated state.

[0026] Furthermore, the process of implementing the service environment mapping by combining fast Boolean operation in step 4 specifically includes the following steps:

[0027] Step 4.1: Obtain the surface position and color information of the target coal seam;

[0028] Step 4.2: Based on the position information of the integrated miner and anchor machine, the bounding box algorithm is used to obtain the intersection of the integrated miner and anchor machine and the coal seam, as well as the area to be removed by Boolean operation in the coal seam, and the surface rendering shape, position and color of the area are saved;

[0029] Step 4.3: Remove the shape and position of the cross section saved in step 4.2 to make it the opposite of the shape of the anchor and digging machine;

[0030] Step 4.4: Color the removed part and assign the color information saved in step 4.2 to the intersection surface.

[0031] Furthermore, the method for analyzing and diagnosing data based on artificial intelligence methods described in step 5 specifically includes the following steps:

[0032] The given sensor data is saved in the operating condition database, and the data is trained in combination with artificial intelligence methods. The trained artificial intelligence model is used for fault prediction. The digital twin of the anchor drilling machine processes the three-phase current of the oil pump motor and the three-phase current of the left loader after simple data processing, and then inputs the data into the trained BP neural network to calculate the status score of the equipment in real time and give a warning when the score is low. The three-phase current of the cutting motor and construction parameters such as the feed amount, feed speed, cutting amount and cutting speed during operation are input into the broken tooth rate prediction model trained in advance with a support vector machine to obtain the broken tooth risk level and broken tooth rate expectation in real time, and give a real-time warning.

[0033] The beneficial effects of the present invention are as follows:

[0034] (1) Based on the concept of digital twin, the present invention proposes a data virtual-real mapping technology for the anchor drilling machine, collects and transmits twin data online, and realizes the information transmission from real to virtual based on asynchronous transmission;

[0035] (2) The present invention adopts a kinematic library construction method, based on the historical physical simulation database and real-time drive data, to achieve rapid simulation analysis of the anchoring machine and its service environment;

[0036] (3) The real-time state mapping method of the anchor and digger based on digital twin in the method of the present invention combines finite state machines, fast Boolean operations and other methods to achieve periodic mapping of the working states of the anchor and digger during lifting, feeding, excavation, and coal slag recovery, and real-time simulation of the service environment during the movement process. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 A real-time state mapping flow chart of the method of the present invention;

[0038] Figure 2 This is the technical flow chart of virtual-real mapping of data of anchor drilling machine;

[0039] Figure 3 It is a network communication flow chart;

[0040] Figure 4 (a) is a schematic diagram of the model tree reconstruction of the virtual entity of the anchor drilling machine;

[0041] Figure 4 (b) is a schematic diagram of the grid editing of the virtual entity construction of the anchor drilling machine;

[0042] Figure 4 (c) is a schematic diagram of the model collapse constructed by the virtual entity of the anchor-digger machine;

[0043] FIG4 (d) is a schematic diagram of material mapping texture production for virtual entity construction of the anchor-digger machine;

[0044] Figure 4 (e) is a schematic diagram of the digital twin virtual entity constructed by the virtual entity of the anchor drilling machine;

[0045] Figure 5 This is a schematic diagram of the dynamic model of the integrated anchor and digging machine after processing;

[0046] Figure 6 It is the state transfer flow chart of the finite state machine;

[0047] Figure 7 It is a schematic diagram of the fast Boolean operation steps;

[0048] Figure 8 This is a schematic diagram of the coal and rock cutting process of the integrated digging and anchoring machine;

[0049] Fig. 9 Schematic diagram of the convolutional neural network training process. DETAILED DESCRIPTION

[0050] The present invention will be described in detail below based on the accompanying drawings and preferred embodiments, and the purpose and effects of the present invention will become more clear. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0051] The real-time state mapping method of the anchoring machine based on digital twin of the present invention first transmits the working state information of the anchoring machine in real time through data virtual-real mapping technology; then performs real-time motion simulation based on the geometric and kinematic model library construction method; and constructs the lifting, feeding, excavation and other behavior simulations of the anchoring machine through the finite state machine; realizes the service environment mapping process by combining fast Boolean operations; finally, analyzes and diagnoses the data based on artificial intelligence methods; and completes the real-time state mapping of the anchoring machine; its flow chart is as follows Figure 1 shown.

[0052] The data virtual-real mapping technology specifically includes the following steps:

[0053] like Figure 2As shown, the network communication port is opened in the digital twin client of the anchor-mining machine, and the actual anchor-mining machine data acquisition and monitoring system is set as the server, listening and waiting for the client connection. During the connection process, the sensor acquisition program is blocked and does not loop. When the digital twin client connection is detected, the sensor acquisition data is transmitted to the server. At this time, two synchronization tasks will be created, including the actual anchor-mining machine continuing to execute the task and the twin data being fully transmitted to the digital twin client, and real-time status mapping is performed in the digital twin client according to the sensor data. At the same time, data integrity judgment is performed, and the next wave of data stream transmission is performed after integrity is achieved. The sensor data includes displacement sensor data, pressure sensor data, photoelectric sensor data, and smoke sensor data.

[0054] When the number of clients is too large, consider using a distributed network configuration to form a server cluster. Figure 3 , using three-way handshake and four-way handshake for data communication: set the timer, execute the data sending function every fixed time (50ms), store it in the asynchronous buffer every time it needs to be received, wait for acceptance, and send the next message when the previous message is received. At this time, the asynchronous function will be entered, and a warning message will be sent when the sending fails.

[0055] The geometric and kinematic model library construction method specifically includes the following steps: reconstructing and rendering the geometric model of the anchor-digger machine; before starting the actual mapping, reconstructing and rendering the model in the model rendering software 3dsMax to obtain a virtual entity of the anchor-digger machine with geometric dimensions and rendering relationships; analyzing the kinematic characteristics of each mechanism of the motion mechanism of the anchor-digger machine separately; establishing kinematic pairs and drives for the mechanism in the kinematic simulation software according to the actual working conditions; obtaining the changes of each kinematic pair of the mechanism and the changes of the main drive through post-processing; storing the relevant data in a historical database; and realizing fast data driving through interpolation calculation in combination with the actual sensor data; the kinematic simulation software includes but is not limited to: Admas, Creo, Ug, SolidWorks, etc.

[0056] Before starting the actual mapping, it is necessary to construct a virtual entity of the anchor-digger machine as a model carrier for real-time state mapping. The present invention will reconstruct and render the model in the model rendering software 3dsMax. As shown in FIG4 , in order to obtain a virtual entity of the anchor-digger machine with geometric dimensions and rendering relationship, the following operations need to be performed:

[0057] (1) First, a geometric model of the anchoring machine is constructed in 3D modeling software, including geometric dimensions and parameter relationships;

[0058] (2) Import 3dsmax and redefine the assembly relationship. As shown in Figure 4(a), reconstruct the model tree and establish the corresponding assembly relationship;

[0059] (3) Add, delete, and adjust the mesh. As shown in Figure 4(b), redivide the redundant mesh nodes to reduce the model weight;

[0060] (4) Delete redundant parts. As shown in Figure 4(c), the redundant parts are meshed or deleted to reduce unnecessary parts or features in the model.

[0061] (5) Create surface materials, maps, textures, etc. See Figure 4(d).

[0062] Combining the above five steps, we can obtain a digital twin virtual entity, and then use the twin data to drive the twin service. For dynamics, it is necessary to establish the corresponding dynamic model of the anchor and digger in the dynamic simulation software. For the motion mechanism of the anchor and digger, analyze the kinematic characteristics of each mechanism separately, and establish kinematic pairs and drives for the mechanism in Adams according to the actual working conditions. The changes of each kinematic pair of the mechanism and the changes of the main drive are obtained through post-processing, and the relevant data are stored in the historical database. Combined with the actual sensor data, fast data driving is achieved through interpolation calculation. The dynamic model processed by Adams is as follows: Figure 5 shown.

[0063] The finite state machine simulates the lifting, feeding, and excavation behaviors of the anchor and digger. After a given sensor data input, the sensor data is matched with a certain state, and the anchor and digger is converted to the indicated state. The sensor and state matching method mainly performs one-to-one corresponding operations based on different sensor values ​​and different state thresholds. For the entire working process of the anchor and digger, taking the trenching arm as an example, its working state is as follows: Figure 6 shown.

[0064] like Figure 7 As shown, the process of realizing service environment mapping by combining fast Boolean operation specifically includes the following steps:

[0065] (1) First, obtain the surface position and color information of the target coal seam;

[0066] (2) Based on the position information of the integrated miner and anchor machine, the bounding box algorithm is used to obtain the intersection of the integrated miner and anchor machine and the coal seam, as well as the area to be removed by Boolean operation in the coal seam, and the surface rendering shape, position and color of the area are saved;

[0067] (3) removing the shape and position of the cross section saved in (2) so that it is opposite to the shape of the anchoring and mining machine;

[0068] (4) Coloring the removed part, assigning the color information saved in (2) to the intersection surface;

[0069] By integrating the geometric model of the anchor drill into the service environment, the service environment can be dynamically simulated in real time based on the twin data transmitted in real time. The twin data received by the twin service is called for real-time driving, and the dynamic changes of the integrated service environment are realized based on the results of the kinematic model call combined with fast Boolean operations. Figure 8 As shown, the process of the integrated miner and anchor machine cutting coal and rock is shown. The rock strata in the service environment are cut in real time, and the cutting area and position response are the same as the corresponding cutting drum motion trajectory, thus achieving consistency in the changes in the service environment of the integrated miner and anchor machine.

[0070] The data analysis and diagnosis method based on artificial intelligence methods includes: saving the given sensor data into the operating condition database, training the data with artificial intelligence methods, and using the trained artificial intelligence model for fault prediction; the digital twin of the anchor drilling machine inputs the three-phase current of the oil pump motor, the three-phase current of the left loader and other data into the trained BP neural network after simple data processing, calculates the status score of the equipment in real time, and gives a warning when the score is low; the three-phase current of the cutting motor and the construction parameters such as the feed amount, feed speed, cutting amount and cutting speed during operation are input into the broken tooth rate prediction model trained in advance with a support vector machine, and the broken tooth risk level and broken tooth rate expectation are obtained in real time, and a real-time warning is given; artificial intelligence methods include but are not limited to: support vector machines, convolutional neural networks, long short-term memory neural networks, etc. The operating condition database includes write, store, and read functions; and the function of recording the data of displacement sensors, pressure sensors, photoelectric sensors, and smoke sensors. Among them, Fig. 9 This is the training process of a convolutional neural network, which can be used for fault diagnosis of operating conditions.

Claims

1. A real-time state mapping method for an integrated anchoring and digging machine based on digital twins, characterized in that: The steps include: Step 1: Use data virtual-real mapping technology to transmit the working status information of the anchor drilling machine in real time; Step 2: Based on the construction method of geometric and kinematic model library, real-time motion simulation is performed; Step 3: Use the finite state machine to build the lifting, feeding, and excavation behavior simulation of the anchoring and digging machine; Step 4: Combine fast Boolean operations to implement the service environment mapping process; Step 4.1: Obtain the surface position and color information of the target coal seam; Step 4.2: Based on the position information of the integrated miner and anchor machine, the bounding box algorithm is used to obtain the intersection of the integrated miner and anchor machine and the coal seam, as well as the area to be removed by Boolean operation in the coal seam, and the surface rendering shape, position and color of the area are saved; Step 4.3: Remove the shape and position of the cross section saved in step 4.2 to make it the opposite of the shape of the anchor and digging machine; Step 4.4: Color the removed part and assign the color information saved in step 4.2 to the intersection surface; Step 5: Analyze and diagnose the data based on artificial intelligence methods; complete the real-time status mapping of the drilling and anchoring machine.

2. The real-time state mapping method of the anchor drilling machine based on digital twin according to claim 1 is characterized in that: The data virtual-real mapping technology specifically includes the following steps: Open the network communication port in the digital twin client of the anchor drilling machine, set the actual anchor drilling machine data acquisition and monitoring system as the server, listen and wait for the client's connection; During the connection process, the sensor acquisition program is blocked and does not loop. When it is detected that the digital twin client connection is completed, the sensor collected data will be transmitted to the server. At this time, two synchronization tasks will be created, including the actual drilling and anchoring machine continuing to execute the task and the twin data being completely transmitted to the digital twin client. Real-time status mapping will be performed in the digital twin client based on the sensor data. At the same time, data integrity judgment will be performed, and the next wave of data stream transmission will be carried out after data integrity is achieved.

3. The real-time state mapping method of the anchoring and digging machine based on digital twin according to claim 2 is characterized in that: The sensor data includes displacement sensor data, pressure sensor data, photoelectric sensor data and smoke sensor data.

4. The real-time state mapping method of the anchoring and digging machine based on digital twin according to claim 2 is characterized in that: When the client is too large, a distributed network configuration is performed to form a server cluster; three-way handshake and four-way wave are used for data communication: a timer is set to execute the data sending function at fixed intervals, and each time a message needs to be received, it is stored in the asynchronous buffer area and waits for receipt. When the previous message is received, the next message is sent; at this time, the asynchronous function is entered, and a warning message is sent when the sending fails.

5. The real-time state mapping method of the anchor drilling machine based on digital twin according to claim 1 is characterized in that: The method for constructing the geometric and kinematic model library described in step 2 specifically includes the following steps: The geometric model of the mining and anchoring machine is reconstructed and rendered. Before the actual mapping begins, the geometric model is reconstructed and rendered in the model rendering software 3dsMax to obtain a virtual entity of the mining and anchoring machine with geometric dimensions and rendering relationships. For the motion mechanism of the mining and anchoring machine, the kinematic characteristics of each mechanism are analyzed separately. According to the actual working conditions, kinematic pairs and drives are established for the mechanism in the kinematic simulation software. The changes of each kinematic pair of the mechanism and the change of the main drive are obtained through post-processing, and the relevant data are stored in the historical database. Combined with the actual sensor data, fast data driving is achieved through interpolation calculation.

6. The real-time state mapping method of the anchoring and digging machine based on digital twin according to claim 5 is characterized in that: The specific steps of geometric model reconstruction and rendering are as follows: Step a: Create a geometric model of the anchoring machine in 3D modeling software, including geometric dimensions and parameter relationships; Step b: Import 3dsmax, redefine the assembly relationship, reconstruct the model tree, and establish the corresponding assembly relationship; Step c: Add, delete, and adjust the mesh, re-divide the redundant mesh nodes, and reduce the weight of the model; Step d: Delete redundant parts, collapse or delete the redundant parts, and reduce unnecessary parts or features in the model; Step e: Make surface materials, maps and textures.

7. The real-time state mapping method of the anchoring and digging machine based on digital twin according to claim 5 is characterized in that: The kinematics simulation software is Admas, Creo, Ug or SolidWorks.

8. The real-time state mapping method of the anchoring and digging machine based on digital twin according to claim 1 is characterized in that: The finite state machine in step 3 constructs the lifting, feeding and digging behavior simulation of the anchoring and digging machine, including the following specific steps: after a given sensor data input, the sensor data is matched with a certain state, and the anchoring and digging machine is converted to the indicated state.

9. The real-time state mapping method of the anchoring and digging machine based on digital twin according to claim 1 is characterized in that: The method for analyzing and diagnosing data based on artificial intelligence methods described in step 5 specifically includes the following steps: The given sensor data is saved in the operating condition database, and the data is trained in combination with artificial intelligence methods. The trained artificial intelligence model is used for fault prediction. The digital twin of the anchor drilling machine processes the three-phase current of the oil pump motor and the three-phase current of the left loader after simple data processing, and calculates the status score of the equipment in real time, giving a warning when the score is low. The three-phase current of the cutting motor as well as the feed amount, feed speed, lower cutting amount and lower cutting speed during operation are input into the broken tooth rate prediction model trained in advance with a support vector machine, and the broken tooth risk level and broken tooth rate expectation are obtained in real time, and a real-time warning is issued.

Images