Mining Tunneling Machine Cutting Tooth Lifecycle Tracking and Intelligent Replacement Decision System and Method
By setting up structured visual coding and explosion-proof visual acquisition systems on the cutting tooth mechanism, and combining them with the tunneling machine's operating parameters, the wear status of the cutting teeth can be accurately tracked and their lifespan predicted. This solves the problem of blindly replacing cutting teeth and improves the continuity and efficiency of coal mine production.
Patent Information
- Application Number
- CN202611282036.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-08-24
- Publication Date
- 2026-11-17
AI Technical Summary
Existing technologies cannot accurately track and assess the wear condition of individual cutting teeth, leading to blind replacement of cutting teeth, unplanned downtime of excavating machines, and impacting the production efficiency of coal mines.
By employing identification components and mining explosion-proof visual acquisition mechanisms, the identity of the cutting tooth mechanism is identified through structured visual coding. Combined with the operating parameters of the tunneling machine's onboard system, the health and load data of the cutting tooth are calculated to generate intelligent replacement decisions.
It enables precise tracking of the wear status and life prediction of individual cutting teeth, reduces unplanned downtime, optimizes cutting tooth replacement strategies, and ensures the continuity and production efficiency of coal mine tunneling operations.
Smart Images

Figure CN123109652A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of intelligent maintenance technology for mining machinery, specifically to a life cycle tracking and intelligent replacement decision-making system and method for mining tunneling machine cutting teeth. Background Technology
[0002] Cantilever roadheaders are widely used in coal mine roadway excavation operations. The cutting teeth on the cutting head are high-wear components, serving as working tools for directly crushing rock and coal. The complex geological conditions underground and the frequent changes in rock hardness cause severe wear or chipping of the cutting teeth during the cutting process. The health of the cutting teeth directly affects the cutting efficiency and overall lifespan of the roadheader.
[0003] Traditional monitoring of cutting teeth mainly relies on manual inspections by workers during equipment downtime. However, due to the high dust concentration and dim lighting conditions underground, manual observation makes it difficult to objectively and quantitatively assess the true wear of individual cutting teeth. Furthermore, it is susceptible to misjudgments due to subjective experience. Existing equipment monitoring technologies often focus on collecting the overall current of the cutting motor or the vibration signal of the cutting arm. Macroscopic monitoring methods can only reflect the overall load status of the cutting head and cannot accurately locate and track the operation of individual cutting teeth. Since the installation position, cutting angle, and movement trajectory of each cutting tooth on the same cutting head are different, their actual load and wear rate during operation are not consistent. If calculations are made solely based on the overall machine operating time or macroscopic force data, it is impossible to truly reflect the degradation process and remaining service life of each individual cutting tooth.
[0004] Lacking accurate individual tracking and lifespan prediction, mining companies adopt a strategy of fixed-cycle preventative replacement or reactive replacement after tooth breakage. The former, an extensive maintenance model, leads to the premature disposal of some cutting teeth that still have cutting capabilities, increasing ineffective tool procurement costs. In the latter case, the tooth holder is severely damaged due to direct friction with the rock wall caused by the missing cutting tooth. The existing replacement decisions are out of touch with the overall production organization plan of the mine. When maintenance personnel discover hidden dangers and carry out temporary tooth replacement operations, it will interrupt the original roadway excavation plan, causing too many unplanned downtimes, thereby reducing the actual mining progress and production efficiency of the coal mine. Summary of the Invention
[0005] To address the shortcomings of existing technologies, this invention provides a lifecycle tracking and intelligent replacement decision-making system and method for mining tunneling machine cutting teeth. This solves the problem that existing technologies cannot accurately track and evaluate the wear and load status of individual cutting teeth, leading to blind cutting tooth replacement and unplanned downtime of the tunneling machine.
[0006] To achieve the above objectives, the present invention provides the following technical solution: The first aspect of this invention provides a lifecycle tracking and intelligent replacement decision-making system for cutting teeth of mining tunneling machines, the system comprising: An identification component is disposed on the cutting tooth mechanism around the cutting head, and includes an identification information carrier with a structured visual code on its surface, wherein the structured visual code is used to characterize the unique identification information of the target cutting tooth mechanism; a mining explosion-proof visual acquisition mechanism is distributed toward the cutting head and is used to acquire multi-view image sequences of the cutting head. The mining explosion-proof industrial control computer is communicatively connected to the mining explosion-proof visual acquisition mechanism and the tunneling machine onboard system. It is used to register and stitch the multi-view image sequence to generate a correction image, identify the structured visual code in the correction image to obtain a unique identification number, and extract the two-dimensional contour mask of the target cutting tooth mechanism to calculate a comprehensive health score. It is also used to calculate the cumulative cutting load data of the target cutting tooth mechanism based on the operating parameters output by the tunneling machine onboard system, obtain the historical comprehensive health score sequence and cumulative cutting load sequence corresponding to the target cutting tooth mechanism based on the unique identification number, output the remaining service life prediction value based on the historical comprehensive health score sequence and cumulative cutting load sequence, and generate a replacement decision instruction in combination with the obtained planned shutdown window information. A mining human-machine interface terminal, connected to the mining explosion-proof industrial control computer, is used to locate the target cutting tooth mechanism in the display interface and output maintenance prompt information containing the replacement decision command.
[0007] Preferably, the cutting tooth mechanism includes a cutting tooth body and a tooth tip disposed at the working end of the cutting tooth body, wherein an alloy head is embedded in the tooth tip; the identification component is disposed in the non-direct cutting area of the cutting tooth body and further includes a protective cover, a set screw, and a sealing ring; the identification information carrier is embedded in a receiving cavity on the surface of the cutting tooth body; the protective cover covers the outside of the identification information carrier and is fixed to the cutting tooth body by the set screw; The sealing ring is pressed between the protective cover and the cutting tooth body to seal; the protective cover forms a light-transmitting recognition area at least in the area corresponding to the structured visual code, and the light-transmitting recognition area is formed of a transparent wear-resistant material so that the mining explosion-proof visual acquisition mechanism can obtain the structured visual code image through the light-transmitting recognition area.
[0008] By adopting a structure that combines a protective cover with a accommodating cavity, direct frictional damage to the marking information carrier is avoided during cutting operations. In conjunction with a sealing ring, dust intrusion is prevented, ensuring the long-term readability of visual coding under harsh working conditions.
[0009] Preferably, it also includes a vibration damping mounting mechanism fixed to the side of the cutting arm and an explosion-proof dust removal protection mechanism installed on the vibration damping mounting mechanism; The vibration damping installation mechanism includes an upper mounting plate, a lower mounting plate, and a support body connecting the upper mounting plate and the lower mounting plate. The vibration damping installation mechanism is fixed by anti-loosening fasteners, and a high-damping vibration damping pad is provided between the upper mounting plate and the lower mounting plate. The mining explosion-proof visual acquisition mechanism is housed inside the explosion-proof protective cover. The mining explosion-proof visual acquisition mechanism specifically includes an explosion-proof protective cover, a tempered glass window, a camera mounting base, and a main lighting source, an auxiliary lighting source, and an explosion-proof industrial camera mounted on the camera mounting base; the explosion-proof dust removal and protection mechanism includes the explosion-proof protective cover housing, an explosion-proof window, a narrow-slit high-pressure air knife dust removal nozzle, compressed air distribution pipeline, a micro-positive pressure air intake interface, a wiper motor, a micro electric wiper, a wiper blade, a wiper mounting base, and a sealing assembly; The explosion-proof window is embedded in the front side of the explosion-proof protective cover and is externally sealed by the sealing assembly; the micro-positive pressure air inlet is connected to the inner cavity of the explosion-proof protective cover; the compressed air distribution pipeline is laid on the outside of the explosion-proof protective cover, with one end connected to the air source and the other end connected to the narrow-slit high-pressure air knife dust removal nozzle to form a continuous air curtain on the outer surface of the explosion-proof window; the miniature electric wiper is mounted through the wiper mounting base and, driven by the wiper motor, drives the wiper blade to adhere to the outer surface of the explosion-proof window for wiping.
[0010] The second aspect of this invention provides a method for lifecycle tracking and intelligent replacement decision-making of cutting teeth for mining tunneling machines, the method comprising the following steps: A multi-view image sequence of the cutting head is acquired and registered and stitched to generate a corrected image; the structured visual code of the surface of the target cutting tooth mechanism is identified in the corrected image to obtain a unique identification number, the two-dimensional contour mask of the target cutting tooth mechanism is extracted and the spatial coordinates are determined; the corrected two-dimensional contour mask is compared with the standard contour mask of the same model cutting tooth mechanism in the undamaged state, feature parameters are extracted and a comprehensive health score is calculated. The operating parameters output by the tunneling machine's onboard system are obtained to determine the current equivalent output load of the cutting head. The load is then allocated to the target cutting tooth mechanism according to the load allocation weight to obtain the instantaneous equivalent cutting load, and cumulative cutting load data is generated by cumulative calculation. Based on the unique identification number, the historical comprehensive health score sequence and cumulative cutting load sequence corresponding to the target cutting tooth mechanism are obtained. A multi-dimensional time series feature matrix is constructed based on the obtained sequence, and the trained time series data analysis model is input to output the remaining service life prediction value. The remaining service life prediction value is compared with the preset service life threshold, and a replacement decision instruction is generated by combining the obtained planned shutdown window information; the spatial coordinates and replacement decision instruction are mapped to the display interface of the mining human-machine interaction terminal, and the positioning guidance screen and maintenance prompt information are output.
[0011] Preferably, the steps of identifying the structured visual code of the target cutting mechanism surface in the corrected image, obtaining the unique identification number, extracting the two-dimensional contour mask of the target cutting mechanism, and determining its spatial coordinates include: Search for candidate regions that match the external geometric contour of the protective cover, extract the pixel blocks within the candidate regions corresponding to the light-transmitting recognition area, convert them into a binary matrix using a dynamic grayscale threshold and parse them to obtain the unique identification number; using the geometric center coordinates of the external geometric contour of the protective cover as the reference point, extract the set of connected edge pixels and perform hole filling and boundary smoothing processing to generate the two-dimensional contour mask; query the corresponding preset three-dimensional installation coordinates according to the unique identification number, and perform position correction on the preset three-dimensional installation coordinates in combination with camera calibration parameters, the current pose of the cutting head and the detection position to obtain the spatial coordinates of the target cutting tooth mechanism.
[0012] Background interference is reduced by candidate region search and dynamic grayscale thresholding to improve mask extraction accuracy; position correction is performed by combining camera parameters and cutting head pose to improve the calculation accuracy of spatial coordinates of the cutting tooth mechanism.
[0013] Preferably, the step of comparing the corrected two-dimensional contour mask with the standard contour mask of the same type of cutting tooth mechanism in a non-destructive state, extracting feature parameters, and calculating a comprehensive health score includes: correcting the two-dimensional contour mask to the same preset reference viewpoint and scale as the standard contour mask; comparing the current contour with the standard contour, and extracting the wear area ratio, chipping index, and symmetry deviation; mapping the wear area ratio, chipping index, and symmetry deviation to the corresponding intervals using the maximum-minimum normalization method to obtain normalized parameters; reading the pre-calibrated weight coefficients, and calculating the comprehensive health score based on the weighted linear combination model.
[0014] Preferably, the steps of obtaining the operating parameters output by the tunneling machine's onboard system to determine the current equivalent output load of the cutting head, allocating the load to the target cutting tooth mechanism according to the load allocation weight to obtain the instantaneous equivalent cutting load, and accumulating and calculating to generate cumulative cutting load data include: The effective cutting area at the current sampling time is determined based on the current pose of the cutting head (6), the relative position of the cutting head (6) and the current coal and rock working face, and the structural parameters of the cutting head (6). Based on the spatial coordinates of each cutting tooth mechanism (1) in the coordinate system of the cutting head (6) and the positional relationship between the effective cutting area, the cutting tooth mechanism located in the effective cutting area is determined as the effective cutting cutting tooth mechanism at the current sampling time, forming a set of effective cutting tooth mechanisms. The initial load weight is determined based on the installation radius of each effective cutting tooth mechanism, the preset tooth position load coefficient, and the angular position correction coefficient, and the load distribution weight is obtained by normalization. The current equivalent output load is allocated to each effective cutting tooth mechanism according to the load allocation weight, and the instantaneous equivalent cutting load of the target cutting tooth mechanism at the current sampling time is obtained. The instantaneous equivalent cutting load is accumulated with the sampling time interval and superimposed with the previously saved cumulative cutting load to generate cumulative cutting load data.
[0015] By using the cutting head position and the installation parameters of the cutting tooth mechanism, the output load of the entire tunneling machine is decomposed to the specific cutting tooth mechanism in the effective cutting area, and the historical load accumulation of a single cutting tooth mechanism is quantified.
[0016] Preferably, the steps of obtaining the historical comprehensive health score sequence and cumulative cutting load sequence corresponding to the target cutting tooth mechanism according to the unique identification number, constructing a multi-dimensional time series feature matrix based on the obtained sequence, and inputting the trained time series data analysis model to output the remaining service life prediction value include: reading the historical comprehensive health score sequence and cumulative cutting load sequence from the life cycle database according to the unique identification number; The historical comprehensive health score sequence and the cumulative cut-off load sequence are mapped according to the timestamp, and a multi-dimensional time series feature matrix is constructed using a sliding time window. The multi-dimensional time series feature matrix is input into the trained time series data analysis model to perform time series analysis on the state degradation trend and output the predicted value of the remaining service life.
[0017] By utilizing unique identification numbers to extract historical data, and performing joint time-series analysis of morphological wear and load accumulation data, the reliability of predicting remaining service life is improved.
[0018] Preferably, the step of comparing the remaining useful life prediction value with a preset useful life threshold and generating a replacement decision instruction in combination with the obtained planned downtime window information includes: comparing the remaining useful life prediction value with a pre-configured safe useful life threshold and an economic useful life threshold, wherein the safe useful life threshold is less than the economic useful life threshold; and generating an immediate replacement instruction when the remaining useful life prediction value is less than or equal to the safe useful life threshold. When the predicted remaining service life is between the safe service life threshold and the economic service life threshold, the mine equipment maintenance schedule and planned shutdown window information are obtained; if there is a planned shutdown window whose start time is within the allowable waiting time range and whose duration meets the requirements for gear replacement operation, a planned window replacement instruction is generated; otherwise, a continue observation instruction is generated; when the predicted remaining service life is greater than the economic service life threshold, a continue observation instruction is generated.
[0019] By combining the remaining service life prediction results with the mine shutdown schedule window for tiered decision-making, forced shutdowns for gear replacement can be avoided, reducing the impact of maintenance operations on normal production.
[0020] Preferably, the step of mapping the spatial coordinates and replacement decision instructions to the display interface of the mining human-machine interaction terminal and outputting the positioning guidance screen and maintenance prompt information includes: parsing the maintenance decision dataset containing the unique identification number of the target cutting tooth mechanism, spatial coordinates, comprehensive health score, remaining service life prediction value and replacement decision instructions; When the mining human-machine interface terminal displays the three-dimensional model of the cutting head, it establishes a positional correspondence between the corresponding tooth nodes in the three-dimensional model of the cutting head based on the tooth installation parameters; it generates a highlighted label at the corresponding display position and overlays the unique identification number, comprehensive health score, remaining service life prediction value, and replacement decision instruction.
[0021] By mapping the status data and replacement instructions of the cutting tooth mechanism to a 3D model, operators are provided with positioning guidance for tooth replacement, shortening the time spent locating the cutting tooth mechanism to be maintained.
[0022] This invention provides a system and method for lifecycle tracking and intelligent replacement decision-making of cutting teeth for mining tunneling machines. It has the following beneficial effects: 1. This invention sets up an identification component with structured visual coding in the non-direct cutting area of the cutting tooth body of the cutting tooth mechanism, and combines it with a mining explosion-proof visual acquisition mechanism to obtain multi-view images, so as to realize the accurate positioning and tracking of the identity of a single cutting tooth mechanism in underground working conditions. The system extracts the two-dimensional contour mask of the target cutting tooth mechanism and compares the feature parameters of the two-dimensional contour mask with the standard contour mask to calculate the comprehensive health score. This overcomes the problems of inconsistent detection standards and low efficiency in traditional manual underground inspection based on experience, and improves the objectivity of the wear condition assessment of the cutting tooth mechanism.
[0023] 2. Based on the health assessment of the cutting tooth mechanism, this invention introduces real-time operating parameters of the tunneling machine's onboard system. By determining the current position of the cutting head, the effective cutting area is identified, and the equivalent output load is weighted and distributed to the target cutting tooth mechanism to generate cumulative cutting load data. By integrating the external geometric wear state of the cutting tooth mechanism with the dynamic stress conditions, multi-dimensional field feature inputs are provided for the time-series data analysis model, reducing the calculation deviation caused by estimating the life based solely on working time or single wear amount, and improving the reliability of remaining service life prediction.
[0024] 3. This invention utilizes an intelligent replacement decision-making mechanism that combines planned downtime window information. After comparing the predicted remaining service life of the cutting tooth mechanism with safety and economic life thresholds, it generates decision instructions in conjunction with the mine equipment maintenance schedule. Through a maintenance strategy with multiple constraints, it can prioritize tooth replacement operations within the regular downtime window while ensuring that the equipment does not suffer serious damage. This reduces the number of unplanned downtimes caused by sudden tooth breakage or blind preventative replacement, thus ensuring the continuity of coal mine tunneling operations. Attached Figure Description
[0025] Figure 1 This is a schematic diagram of the system interaction timing of the present invention; Figure 2 This is a diagram of the overall system architecture of the present invention; Figure 3 This is a schematic diagram of the method flow of the present invention; Figure 4 This is a flowchart of the automatic image acquisition and multi-view registration and correction process for the cutting head according to the present invention; Figure 5 This is a timing diagram of the interaction between the target cutting tooth mechanism identity resolution and spatial pose extraction in this invention; Figure 6 This is a flowchart of the two-dimensional contour feature extraction and comprehensive health score calculation of the target cutting tooth mechanism of the present invention; Figure 7 This is a flowchart of the dynamic allocation and time-series accumulation calculation of the equivalent cutting load of the cutting teeth according to the present invention. Figure 8 This is a flowchart of the remaining service life prediction of cutting teeth based on a time-series data analysis model according to the present invention. Figure 9 This is a schematic diagram of the remaining service life prediction process for the cutting teeth according to the present invention; Figure 10 This is a logic diagram showing the linkage between the intelligent replacement decision engine and the planned shutdown window of the present invention. Figure 11 This is a schematic diagram of the overall structure of the present invention; Figure 12 This is a schematic diagram of the sealing ring of the present invention; Figure 13 This is a schematic diagram of the structure of the cutting tooth body of the present invention; Figure 14 This is a schematic diagram of the wiper blade of the present invention; Figure 15 This is a schematic diagram of the structure of the tempered glass window of the present invention; Figure 16 This is a schematic diagram of the anti-loosening fastener of the present invention; Figure 17 This is a comparison chart of the overall health score and the time-series evolution of the cumulative cutting load of the target cutting tooth mechanism of the present invention. Figure 18 This is a diagram showing the prediction and intelligent replacement decision window for the remaining service life of the cutting teeth in this invention. Figure 19 This is a visual guidance diagram of the mining human-computer interaction terminal of the present invention.
[0026] The components include: 1. Cutting tooth mechanism; 11. Tooth tip; 12. Alloy head; 13. Cutting tooth body; 14. Set screw; 15. Protective cover; 16. Identification information carrier; 17. Sealing ring; 2. Explosion-proof dust removal protection mechanism; 21. Narrow slit high-pressure air knife dust removal nozzle; 22. Compressed air distribution pipeline; 23. Explosion-proof protective cover; 24. Micro positive pressure air inlet; 25. Wiper motor; 26. Wiper blade; 27. Miniature electric wiper; 28. Wiper mounting base. 29. Sealing assembly; 210. Explosion-proof viewing window; 3. Explosion-proof visual acquisition mechanism for mining; 31. Explosion-proof protective cover; 32. Main lighting source; 33. Auxiliary supplementary lighting source; 34. Explosion-proof industrial camera; 35. Tempered glass viewing window; 36. Camera mounting base; 4. Vibration damping mounting mechanism; 41. Upper mounting plate; 42. High damping vibration damping pad; 43. Support body; 44. Anti-loosening fasteners; 45. Lower mounting plate; 5. Cutting arm; 6. Cutting head. Detailed Implementation
[0027] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0028] See attached document Figure 1 Appendix Figure 2 and attached Figure 19 This invention provides a lifecycle tracking and intelligent replacement decision system for mining tunneling machine cutting teeth. The system includes: an identification component, a mining explosion-proof visual acquisition mechanism 3, an explosion-proof dust removal and protection mechanism 2, a vibration reduction installation mechanism 4, a mining explosion-proof industrial control computer, and a mining human-machine interaction terminal.
[0029] The mine explosion-proof industrial control computer establishes a data communication connection with the mine explosion-proof vision acquisition mechanism 3 and receives the operating parameters output by the tunneling machine's onboard system; the mine human-machine interaction terminal is connected to the data output end of the mine explosion-proof industrial control computer. Image acquisition, identification of the cutting tooth mechanism 1, status assessment, calculation of cumulative cutting load, prediction of remaining service life and replacement decision are all completed by the mine explosion-proof industrial control computer at the tunneling machine body end, forming an edge computing processing architecture oriented to the underground site.
[0030] In the front-end execution area, the cutting tooth mechanism 1 is installed around the cutting head 6. The cutting tooth mechanism 1 includes a front tooth tip 11, an alloy head 12, and a rear cutting tooth body 13. The identification component is set in the non-direct cutting area of the cutting tooth body 13. The identification component includes an identification information carrier 16, a protective cover 15, a set screw 14, and a sealing ring 17.
[0031] The identification information carrier 16 is embedded in the receiving cavity provided on the surface of the cutting tooth body 13. The identification information carrier 16 is provided with a structured visual code on the side facing the protective cover 15 to represent the unique identity information of the corresponding cutting tooth mechanism 1. The protective cover 15 is completely covered on the outside of the identification information carrier 16 and fixed to the cutting tooth body 13 by the set screw 14. The protective cover 15 forms a light-transmitting recognition area at least in the area corresponding to the structured visual code, so that the mine explosion-proof visual acquisition mechanism 3 can obtain the structured visual code image on the surface of the identification information carrier 16 through the light-transmitting recognition area. The mine explosion-proof visual acquisition mechanism 3 can obtain the structured visual code image from the corresponding perspective when the cutting tooth mechanism 1 is in different circumferential postures by acquiring multi-view images of the cutting head 6.
[0032] The light-transmitting recognition area is made of transparent and wear-resistant material to provide dustproof, waterproof and impact-resistant protection for the identification information carrier 16 while allowing the structured visual code to be optically recognized. The sealing ring 17 is pressed between the mating surfaces of the protective cover 15 and the cutting tooth body 13 to seal the cavity where the identification information carrier 16 is located. The set screw 14 passes through the protective cover 15 and is screwed and fixed to the cutting tooth body 13 to limit the loosening of the protective cover 15 under vibration and impact conditions of the cutting tooth mechanism 1.
[0033] The vibration damping mounting mechanism 4 is fixed to the side of the cutting arm 5. The vibration damping mounting mechanism 4 includes an upper mounting plate 41, a high-damping vibration damping pad 42, a bracket body 43, anti-loosening fasteners 44, and a lower mounting plate 45. The lower mounting plate 45 is attached to and fixed to the outer surface of the cutting arm 5. The bracket body 43 is supported and connected between the upper mounting plate 41 and the lower mounting plate 45. The high-damping vibration damping pad 42 is clamped between the upper mounting plate 41 and the connecting end face of the bracket body 43. The anti-loosening fasteners 44 pass through the upper mounting plate 41, the high-damping vibration damping pad 42, and the bracket body 43 to achieve overall locking.
[0034] The explosion-proof visual acquisition mechanism 3 and the explosion-proof dust removal and protection mechanism 2 are installed on the top surface of the upper mounting plate 41 and are distributed towards the cutting head 6. The explosion-proof dust removal and protection mechanism 2 includes an explosion-proof protective cover 23, an explosion-proof viewing window 210, a narrow slit high-pressure air knife dust removal nozzle 21, a compressed air distribution pipeline 22, a micro positive pressure air inlet 24, a wiper motor 25, a mini electric wiper 27, a wiper rubber strip 26, a sealing assembly 29, and a wiper mounting base 28.
[0035] The explosion-proof window 210 is embedded in the front side of the explosion-proof protective cover 23 through the sealing component 29. The compressed air distribution pipe 22 is arranged on the upper outside of the explosion-proof protective cover 23, with one end connected to the air source and the other end connected to the narrow-slit high-pressure air knife dust removal nozzle 21. The nozzle of the narrow-slit high-pressure air knife dust removal nozzle 21 faces the outer surface of the explosion-proof window 210. The micro-positive pressure air inlet 24 penetrates the rear side wall of the explosion-proof protective cover 23 and connects to the inner cavity of the explosion-proof protective cover 23.
[0036] The wiper mounting base 28 is fixed to the lower front end of the explosion-proof cover housing 23. The wiper motor 25 is installed inside the wiper mounting base 28. The power input end of the mini electric wiper 27 is connected to the output shaft of the wiper motor 25. The wiper rubber strip 26 is fixed to the wiping end of the mini electric wiper 27 and the wiper rubber strip 26 is attached to the outer surface of the explosion-proof window 210.
[0037] The mining explosion-proof visual acquisition mechanism 3 is housed within the explosion-proof protective housing 23. The mining explosion-proof visual acquisition mechanism 3 includes an explosion-proof protective housing 31, a main lighting source 32, an auxiliary supplementary lighting source 33, an explosion-proof industrial camera 34, a tempered glass viewing window 35, and a camera mounting base 36. The explosion-proof industrial camera 34 is fixed to the camera mounting base 36. The camera mounting base 36 is connected to the upper mounting plate 41 through an elastic vibration damping connector to form a secondary vibration damping unit for the camera. The explosion-proof protective housing 23 covers the outer periphery of the mining explosion-proof visual acquisition mechanism 3 and is fixed to the upper mounting plate 41. The explosion-proof protective housing 31 covers the outside of the explosion-proof industrial camera 34.
[0038] The tempered glass window 35 is embedded in the front opening of the explosion-proof protective cover 31. The main lighting source 32 is distributed in a ring structure around the lens of the explosion-proof industrial camera 34, and the auxiliary lighting source 33 is installed on both sides of the explosion-proof protective cover 31.
[0039] The mining explosion-proof industrial control computer integrates a cutting tooth mechanism 1 identification unit, a status assessment unit, a life cycle management unit, a life prediction unit, and an intelligent decision-making unit. The mining explosion-proof industrial control computer receives image data collected by the explosion-proof industrial camera 34 via cable and communicates with the tunneling machine's onboard system to receive the operating parameters output by the tunneling machine's onboard system. The mining human-machine interaction terminal receives the decision instructions generated by the mining explosion-proof industrial control computer and outputs the positioning coordinates and maintenance information of the cutting tooth mechanism 1.
[0040] See attached document Figure 3 This invention provides a method for lifecycle tracking and intelligent replacement decision-making of cutting teeth for mining tunneling machines, the method comprising the following steps: S1. When the tunneling machine is stopped or the cutting head 6 is rotating at low speed, the mine explosion-proof vision acquisition mechanism 3 is started to acquire a multi-view image sequence of the cutting head 6; noise reduction, contrast enhancement and lens distortion correction are performed on the multi-view image sequence, and multi-frame image registration, pose correction and stitching are performed using the rotation speed information of the cutting head 6 as the inter-frame pose change constraint to generate a corrected image covering the target detection area of the cutting head 6. S2, in the corrected image, the structured visual code on the surface of the identification information carrier 16 is identified through the light-transmitting identification area of the protective cover 15. The structured visual code is decoded to obtain the unique identification number of the target cutting tooth mechanism 1, and a mapping relationship between the physical entity of the target cutting tooth mechanism 1 and the life cycle data file is established. Simultaneously, the pixel-level two-dimensional contour mask of the target cutting tooth mechanism 1 is extracted, and the spatial coordinates of the target cutting tooth mechanism 1 in the coordinate system of the cutting head 6 are determined according to the camera calibration parameters of the mining explosion-proof visual acquisition mechanism 3, the current pose information of the cutting head 6, and the installation position of the cutting tooth mechanism 1 on the cutting head 6. S3, correct the two-dimensional contour mask of the target cutting tooth mechanism 1 to the preset reference viewpoint and reference scale, and compare it with the standard contour mask of the same model cutting tooth mechanism 1 in the undamaged state, extract the wear area ratio, chipping index and symmetry deviation feature parameters of the target cutting tooth mechanism 1; normalize the feature parameters, and use the weighted linear combination model to calculate the comprehensive health score of the target cutting tooth mechanism 1. S4. Obtain the operating parameters of the cutting motor, the instantaneous speed of the cutting head 6, and the spatial coordinates of each cutting tooth mechanism 1 in the coordinate system of the cutting head 6. Determine the current equivalent output load of the cutting head 6 based on the operating parameters of the cutting motor. Determine the effective cutting area based on the relative position of the cutting head 6 and the current coal and rock working face and the structural parameters of the cutting head 6. Based on the positional relationship between the spatial coordinates of each cutting tooth mechanism 1 in the coordinate system of the cutting head 6 and the effective cutting area, determine the cutting tooth mechanism in the effective cutting area as the cutting tooth mechanism currently participating in the cutting, form an effective cutting tooth mechanism set, and determine the corresponding load distribution weight based on the installation radius, current angular position, and preset tooth position load coefficient of each effective cutting tooth mechanism 1. The equivalent output load of the cutting head 6 is distributed to each effective cutting tooth mechanism 1 according to the load distribution weight, so as to obtain the instantaneous equivalent cutting load of the corresponding cutting tooth mechanism 1. The instantaneous equivalent cutting load is accumulated with the cutting time to generate the cumulative cutting load data corresponding to the unique identification number of each cutting tooth mechanism 1. S5. Based on the unique identification number of the target cutting tooth mechanism, extract the historical comprehensive health score sequence and cumulative cutting load sequence from the life cycle database of cutting tooth mechanism 1. Match the two types of time series data according to the timestamp, and construct a multi-dimensional time series feature matrix using a sliding time window. Input the multi-dimensional time series feature matrix into the trained time series data analysis model to perform time series analysis on the state degradation trend of the target cutting tooth mechanism and output the predicted value of the remaining service life of the target cutting tooth mechanism. S6, compare the predicted remaining service life of the target cutting tooth mechanism with the safe service life threshold and the economic service life threshold respectively; when the predicted remaining service life is between the safe service life threshold and the economic service life threshold, obtain the mine equipment maintenance schedule and planned shutdown window information, and determine whether there is a planned shutdown window that meets the conditions for tooth replacement operation before the target cutting tooth mechanism 1 enters the safe service life boundary, and generate a replacement decision among immediate replacement, planned window replacement or continued observation. S7, the spatial coordinate data of the cutting tooth mechanism 1 carrying the replacement command is parsed and mapped to the display interface of the mining human-machine interaction terminal. The location of the target cutting tooth mechanism 1 is highlighted in the display interface, and the identity information, status information, remaining service life and maintenance prompt information of the corresponding cutting tooth mechanism 1 are superimposed and displayed, so as to output the positioning guidance screen and on-site maintenance prompt information of the target cutting tooth mechanism 1.
[0041] See attached document Figure 4 In specific implementations, step S1 provided by the present invention may include the following steps: S101, the mine explosion-proof industrial control computer receives the detection trigger signal of the cutting tooth mechanism 1 issued by the airborne main control system and reads the current operating status of the tunneling machine. When the tunneling machine is in a stopped state or the cutting head 6 is in a low-speed rotation state that meets the image acquisition requirements, the mine explosion-proof industrial control computer starts the explosion-proof dust removal protection mechanism 2. Compressed air enters the explosion-proof protective housing 23 through the micro positive pressure air inlet interface 24, so that the inside of the explosion-proof protective housing 23 maintains a positive pressure state higher than that of the external environment. Meanwhile, compressed air from underground is delivered to the narrow-slit high-pressure air knife dust removal nozzle 21 via compressed air distribution pipeline 22, forming a continuous air curtain on the outer surface of the explosion-proof window 210. For coal dust, water mist, or dirt adhering to the surface of the explosion-proof window 210, the wiper motor 25 drives the miniature electric wiper 27 to wipe it clean. After the window is cleaned, the mine explosion-proof industrial control computer starts the main lighting source 32 and the auxiliary supplementary lighting source 33, and triggers the explosion-proof industrial camera 34 to acquire a multi-view image sequence of the cutting head 6.
[0042] S102, the mining explosion-proof industrial control computer receives the multi-view image sequence of the cutting head 6 transmitted by the explosion-proof industrial camera 34. The mining explosion-proof industrial control computer inputs the multi-view image sequence of the cutting head 6 frame by frame into the image processing module. The mining explosion-proof industrial control computer uses a spatial domain median filtering algorithm to smooth the pixel matrix of each frame in the multi-view image sequence of the cutting head 6, reducing the number of isolated noise points contained in the pixel matrix of the frame image. The mining explosion-proof industrial control computer uses an adaptive histogram equalization algorithm to perform contrast mapping on the smoothed pixel matrix, adjust the pixel value range in the dark area, and output the preprocessed image sequence.
[0043] S103, the mining explosion-proof industrial control computer reads the real-time rotation speed of the cutting head 6 and determines the rotation angle increment corresponding to the adjacent frame according to the acquisition time interval between adjacent image frames; according to the pre-calibrated camera intrinsic parameters and lens distortion parameters of the explosion-proof industrial camera 34, it performs distortion correction on the pre-processed image.
[0044] The mining explosion-proof industrial control computer uses the rotation angle increment as the initial constraint for the pose change of adjacent frames. It extracts and matches image feature points in the overlapping area of adjacent frames, determines the registration transformation parameters corresponding to different local image regions based on the matching results, and maps images from different perspectives to a unified reference coordinate system in combination with the pre-stored geometric structure information of the cutting head 6. For the overlapping area of adjacent perspectives, pixel fusion processing is performed to generate a corrected image covering the target detection area of the cutting head 6.
[0045] See attached document Figure 5 In specific implementations, step S2 provided by the present invention may include the following steps: S201, the mine explosion-proof industrial control computer inputs the corrected image into the identity resolution module. The mine explosion-proof industrial control computer uses a contour detection algorithm to search for candidate regions that match the external geometric contour of the protective cover 15 within the corrected image, and extracts the pixel blocks within the candidate regions that correspond to the light-transmitting recognition area. The mine explosion-proof industrial control computer uses the Otsu method to calculate the maximum inter-class variance of the pixel blocks and determine the dynamic grayscale threshold. Using the dynamic grayscale threshold, the pixel blocks containing the surface structured visual encoding of the identification information carrier 16 are converted into a binary matrix. The matrix decoding logic is called to perform information array parsing on the binary matrix and output the unique identification number of the target cutting tooth mechanism 1. The mine explosion-proof industrial control computer establishes the association relationship between the physical entity of the target cutting tooth mechanism 1 and the life cycle data file corresponding to the unique identification number in the life cycle database.
[0046] S202, the mine explosion-proof industrial control computer uses the geometric center coordinates of the extracted outer geometric contour of the protective cover 15 as the search reference point, reads the geometric prior model parameters of the target cutting tooth mechanism 1 pre-stored inside the mine explosion-proof industrial control computer, and combines the search reference point and the geometric prior model parameters of the target cutting tooth mechanism 1 to delineate the region of interest containing the physical boundary of the target cutting tooth mechanism 1 in the corrected image; the mine explosion-proof industrial control computer uses the edge detection operator to perform derivative operation on the pixel gray-level gradient of the region of interest and extracts the set of connected edge pixels, and performs hole filling and boundary smoothing processing on the set of connected edge pixels in combination with morphological closing operation, assigns the pixels inside the closed boundary as the foreground feature value and the pixels outside the closed boundary as the background feature value, and generates a pixel-level two-dimensional contour mask corresponding to the size of the target cutting tooth mechanism 1 in the current camera view.
[0047] S203, when the system first establishes the life cycle file of the cutting tooth mechanism 1 or when a new cutting tooth mechanism 1 is installed, the life cycle management unit establishes an initial mapping relationship between the unique identification number of the cutting tooth mechanism 1 and the corresponding tooth position installation parameters based on the unique identification number of the cutting tooth mechanism 1 and the corresponding tooth position installation parameters. In the subsequent operation and detection process, the mining explosion-proof industrial control computer queries the mapping relationship based on the unique identification number of the target cutting tooth mechanism 1 to obtain the preset three-dimensional installation coordinates of the corresponding tooth position relative to the cutting head 6. Combining the camera calibration parameters of the mining explosion-proof vision acquisition mechanism 3, the current pose of the cutting head 6 and the detection position of the target cutting tooth mechanism 1 in the calibration image, the preset three-dimensional installation coordinates are corrected to obtain the three-dimensional spatial coordinates of the target cutting tooth mechanism 1 in the coordinate system of the cutting head 6.
[0048] See attached document Figure 6 In specific implementations, step S3 provided by the present invention may include the following steps: S301, the mine explosion-proof industrial control computer corrects the two-dimensional contour mask of the target cutting tooth mechanism 1 to the same reference viewpoint and reference scale as the standard contour mask based on the spatial attitude of the target cutting tooth mechanism 1 and the current image imaging scale. The state assessment unit retrieves the standard contour mask of the same model cutting tooth mechanism 1 in the undamaged state, and calculates the area loss ratio of the current contour based on the area of the standard contour to obtain the wear area ratio of the target cutting tooth mechanism 1; the mine explosion-proof industrial control computer compares the local curvature changes of the corresponding edges of the current contour and the standard contour to obtain the chipping index characterizing the degree of local defects; the central symmetry axis is determined based on the geometric principal axis of the current contour, and the difference in contour area distribution on both sides of the central symmetry axis is compared to obtain the symmetry deviation.
[0049] S302, the status assessment unit predetermines the effective upper and lower boundaries corresponding to the wear area ratio, chipping index and symmetry deviation based on historical valid samples, and maps the three feature parameters to the interval of 0 to 1 using the maximum and minimum value normalization method to obtain the normalized wear area ratio, normalized chipping index and normalized symmetry deviation. When the feature value to be processed exceeds the corresponding effective upper and lower boundaries, it is truncated with the corresponding boundary value to reduce the impact of outliers on the comprehensive health score.
[0050] S303, the condition assessment unit reads the weighting coefficients corresponding to the normalized wear area ratio, normalized chipping index, and normalized symmetry deviation, and the sum of each weighting coefficient is 1. It then calculates the comprehensive health score of the target cutting tooth mechanism 1 according to the following expression: ; In the formula, The overall health score of the target cutting mechanism 1; The weight coefficients for the corresponding normalized wear area ratio in the feature weight assignment matrix; Assign weight coefficients to the corresponding normalized collapse index in the feature weight assignment matrix; Assign weight coefficients to the corresponding normalized symmetry deviation feature parameters in the feature weight assignment matrix; The normalized wear area ratio of the target cutting tooth mechanism 1; The normalized chipping index is the target cutting tooth mechanism 1; The normalized symmetry deviation characteristic parameter is the target cutting tooth mechanism 1.
[0051] The weighting coefficients are pre-calibrated based on the correlation between each characteristic parameter in the historical failure samples and the failure state of the cutting tooth mechanism 1. After the comprehensive health score is generated, the life cycle management unit writes the comprehensive health score into the corresponding life cycle data file according to the unique identification number and detection time of the target cutting tooth mechanism 1.
[0052] The overall health score can be calculated using a two-dimensional contour mask of the target cutting tooth mechanism. The health score can be achieved through the following methods: As the first optional method, the system extracts a single shape feature of the two-dimensional contour mask, such as the absolute area value of the two-dimensional mask, the aspect ratio of the minimum bounding rectangle, or the total perimeter of the edge contour, and linearly maps the single feature value to a preset threshold range to obtain a comprehensive health score.
[0053] As a second option, the mining explosion-proof industrial control computer can deploy an image classification or scoring model based on a convolutional neural network. The system takes the extracted two-dimensional contour mask of the target cutting tooth mechanism 1 as a direct input matrix and inputs it into a neural network model that has been trained in advance using a large amount of mask data with different wear levels. The model directly outputs a score value representing the health status of the cutting tooth mechanism 1 in an end-to-end manner.
[0054] See attached document Figure 7 In specific implementations, step S4 provided by the present invention may include the following steps: S401, the mining explosion-proof industrial control computer synchronously acquires the operating parameters of the cutting motor, the instantaneous speed of the cutting head 6, and the three-dimensional spatial coordinates of each cutting tooth mechanism 1 in the coordinate system of the cutting head 6 according to a unified time reference. The operating parameters of the cutting motor include at least one of real-time current, output torque, and output power. When the output torque can be directly acquired, the output torque is used as the basis for calculating the equivalent output torque of the cutting head 6. When the output power and speed are acquired, the equivalent output torque is calculated according to the correspondence between the output power and the angular velocity. When the motor current parameter is used, the equivalent output torque is obtained by converting it according to the pre-calibrated current-torque correspondence, thereby obtaining the equivalent output load corresponding to the overall load of the cutting head 6 at the current sampling time.
[0055] S402, the mine explosion-proof industrial control computer determines the effective cutting area at the current sampling time in the coordinate system of the cutting head 6 based on the current pose of the cutting head 6, the relative position of the cutting head 6 and the current coal and rock working face, and the structural parameters of the cutting head 6; compares the three-dimensional spatial coordinates of each cutting tooth mechanism 1 in the coordinate system of the cutting head 6 with the effective cutting area, determines whether each cutting tooth mechanism 1 is located in the current space area participating in coal and rock cutting, and forms the set of effective cutting tooth mechanisms at the current sampling time by the cutting tooth mechanisms located in the effective cutting area.
[0056] For any cutting tooth mechanism 1 in the set of effective cutting tooth mechanisms 1, the mining explosion-proof industrial control computer reads the installation radius and the corresponding preset tooth position load coefficient, and determines the angular position correction coefficient according to the relative angular position of the cutting tooth mechanism 1 in the effective cutting area. The tooth position load coefficient is used to characterize the relative load difference of different fixed tooth positions under the same cutting working condition, and is pre-calibrated according to the structural parameters of the cutting head 6, tooth position distribution and historical operating data. The angular position correction coefficient is used to characterize the change in the load participation degree of the cutting tooth mechanism 1 during the process of entering, passing through and leaving the effective cutting area, and is obtained according to the pre-calibrated correspondence between angular position and correction coefficient. The mining explosion-proof industrial control computer determines the initial load weight of each effective cutting tooth mechanism 1 according to the installation radius, tooth position load coefficient and angular position correction coefficient, and normalizes the initial load weight of all effective cutting tooth mechanisms 1 at the same sampling time so that the sum of the load allocation weights is 1. Then, according to the load allocation weight, the equivalent output load of the cutting head 6 is allocated to each effective cutting tooth mechanism 1 to obtain the instantaneous equivalent cutting load of the target cutting tooth mechanism 1 at the current sampling time.
[0057] S403, the mining explosion-proof industrial control computer continuously acquires the instantaneous equivalent cutting load of the target cutting tooth mechanism 1 according to the sampling time sequence. When the target cutting tooth mechanism 1 is in an effective cutting state, the instantaneous equivalent cutting load at the current sampling time is accumulated and calculated with the corresponding sampling time interval to obtain the load increment. When the target cutting tooth mechanism 1 is not in an effective cutting state, the sampling time period is not included in the effective cutting load. The mining explosion-proof industrial control computer superimposes the load increment generated in each calculation cycle with the cumulative cutting load previously saved by the target cutting tooth mechanism 1 to form the continuously updated cumulative cutting load data since the installation of the cutting tooth mechanism 1. The life cycle management unit saves the cumulative cutting load data according to the unique identification number and timestamp of the target cutting tooth mechanism 1, and will use it as the time sequence input data for subsequent prediction of the remaining service life.
[0058] See attached document Figure 8 With appendix Figure 9 In specific implementations, step S5 provided by the present invention may include the following steps: S501, the mining explosion-proof industrial control computer reads the historical comprehensive health score sequence and cumulative cutting load sequence stored in the life cycle database according to the detection time sequence based on the unique identification number of the target cutting tooth mechanism 1, and matches the two types of time series data according to the timestamp. The data of a preset number of continuous time steps is used as a sliding time window. Each time step includes at least the comprehensive health score and cumulative cutting load at the corresponding time, thereby constructing a multi-dimensional time series feature matrix corresponding to the target cutting tooth mechanism 1. The preset number is determined in advance based on the data acquisition cycle, the typical service cycle of the cutting tooth mechanism 1, and the model verification results.
[0059] S502, the life prediction unit inputs the multi-dimensional time-series feature matrix into the trained time-series data analysis model, extracts time-series features of the health status change and cumulative cutting load change of the target cutting tooth mechanism 1, and outputs the corresponding remaining life prediction results. The time-series data analysis model adopts at least one of recurrent neural network, long short-term memory network or time-series regression model. The training samples of the time-series data analysis model come from historical cutting tooth mechanisms 1 with complete life-cycle records. For each historical cutting tooth mechanism 1, the comprehensive health score and cumulative cutting load corresponding to different detection times are used as model input data, and the remaining service time between the corresponding detection time and the time when the cutting tooth mechanism 1 reaches the preset failure state or the actual retirement time is used as the supervision label; a training sample set is formed based on multiple historical cutting tooth mechanisms 1, and the parameters of the time series data analysis model are trained. After the training is completed, the model is deployed in the life prediction unit.
[0060] S503, for the current target cutting tooth mechanism 1, the life prediction unit inputs the comprehensive health score and cumulative cutting load corresponding to the latest sliding time window into the trained time series data analysis model to obtain the remaining life prediction value. When the adopted time series data analysis model outputs the future health change sequence, the health failure threshold pre-calibrated according to the historical failure samples is used as the failure criterion, and the difference between the predicted time when the health first reaches the health failure threshold and the current time is used as the remaining life prediction value. When the time-series data analysis model directly outputs the remaining service life, the model output value is used as the predicted value of the remaining service life. The life cycle management unit writes the predicted value of the remaining service life into the corresponding life cycle data file according to the unique identification number of the target cutting tooth mechanism 1 and the prediction time.
[0061] In specific implementations, step S6 provided by the present invention may include the following steps: S601, the intelligent decision unit reads the predicted value of the remaining service life of the target cutting gear mechanism 1, as well as the safe service life threshold and economic service life threshold pre-configured for the current cutting gear mechanism 1 model and operating conditions. The safe service life threshold is less than the economic service life threshold. The safe service life threshold is used to characterize the cutting gear mechanism 1 entering the remaining service life boundary that requires priority maintenance, and the economic service life threshold is used to determine whether the cutting gear mechanism 1 enters the planned replacement decision interval.
[0062] S602, when the predicted remaining service life is less than or equal to the safe service life threshold, the intelligent decision-making unit generates an immediate replacement instruction. When the safe service life threshold is less than the predicted remaining service life and less than or equal to the economic service life threshold, the intelligent decision-making unit reads the mine equipment maintenance schedule and planned shutdown window information, and determines the current allowable waiting time range of the target cutting tooth mechanism 1 based on the difference between the predicted remaining service life and the safe service life threshold. If there is a planned shutdown window whose start time is within the allowable waiting time range and whose duration meets the requirements for tooth replacement operation, a planned window replacement instruction is generated. If there is no planned shutdown window whose start time and duration both meet the requirements, a continue observation instruction is generated, and the replacement decision is re-executed based on the predicted remaining service life obtained from subsequent detection. When the predicted remaining service life is greater than the economic service life threshold, the intelligent decision-making unit generates a continue observation instruction.
[0063] S603, the mine explosion-proof industrial control computer associates the unique identification number, three-dimensional spatial coordinates, comprehensive health score, remaining service life prediction value and corresponding replacement decision instructions of the target cutting tooth mechanism 1, generates the maintenance decision dataset of the target cutting tooth mechanism 1, and sends the maintenance decision dataset to the mine human-machine interaction terminal.
[0064] See attached document Figure 10 In specific implementations, step S7 provided by the present invention may include the following steps: S701, the mining explosion-proof industrial control computer reads the maintenance decision dataset and parses it to obtain the unique identification number, three-dimensional spatial coordinates, comprehensive health score, remaining service life prediction value and replacement decision instructions of the target cutting tooth mechanism 1.
[0065] S702, when the mining human-machine interaction terminal displays a two-dimensional on-site image, the mining explosion-proof industrial control computer projects the target cutting tooth mechanism 1 onto the corresponding pixel position of the currently displayed image based on the camera calibration parameters of the explosion-proof industrial camera 34, the current pose of the cutting head 6, and the three-dimensional spatial coordinates of the target cutting tooth mechanism 1; when the mining human-machine interaction terminal displays a three-dimensional model of the cutting head 6, the corresponding tooth position node in the three-dimensional model is determined based on the tooth position installation parameters corresponding to the target cutting tooth mechanism 1, thereby establishing the correspondence between the actual spatial position of the target cutting tooth mechanism 1 and the display position of the mining human-machine interaction terminal.
[0066] S703, the mining human-machine interaction terminal generates at least one of the following at the display position corresponding to the target cutting tooth mechanism 1: a highlighted border, a color mark, and an identity label. It also overlays and displays the unique identity number, comprehensive health score, remaining service life prediction value, and corresponding replacement decision information of the target cutting tooth mechanism 1 to generate the on-site positioning and maintenance guidance screen of the target cutting tooth mechanism 1.
[0067] See attached document Figure 11 To be continued Figure 18 To aid in understanding the technical solution of this invention, the following is an application embodiment of a mining tunneling machine cutter lifecycle tracking and intelligent replacement decision system and method.
[0068] The cutting head 6 of the tunneling machine is equipped with a cutting tooth mechanism 1. During the system operation and testing, the mine explosion-proof industrial control computer starts the explosion-proof dust removal and protection mechanism 2. Compressed air enters the explosion-proof protective housing 23 through the micro positive pressure air inlet 24, so that the inside of the explosion-proof protective housing 23 is kept at a positive pressure higher than the external environment. The underground compressed air is delivered to the narrow-slit high-pressure air knife dust removal nozzle 21 through the compressed air distribution pipeline 22, forming a continuous air curtain on the outer surface of the explosion-proof window 210. The wiper motor 25 drives the micro electric wiper 27 to move, so that the wiper blade 26 sticks to the outer surface of the explosion-proof window 210 to wipe it, thereby completing the cleaning of the explosion-proof window 210.
[0069] When the tunneling machine is stopped or the cutting head 6 is rotating at low speed, the mine explosion-proof industrial control computer starts the main lighting source 32 and the auxiliary supplementary lighting source 33, and triggers the explosion-proof industrial camera 34 to acquire a multi-view image sequence of the cutting head 6. The mine explosion-proof industrial control computer inputs the multi-view image sequence of the cutting head 6 frame by frame into the image processing module. The spatial domain median filtering algorithm and the adaptive histogram equalization algorithm are used to perform smoothing calculation and contrast mapping on the multi-view image sequence. The rotation angle increment of the cutting head 6 is used as the initial constraint for the pose change of adjacent frames. Image feature points are extracted and multi-frame image registration, pose correction and stitching are performed to generate a corrected image covering the target detection area of the cutting head 6.
[0070] The mine explosion-proof industrial control computer searches for candidate regions that match the external geometric contour of the protective cover 15 in the calibration image, extracts the pixel blocks corresponding to the light-transmitting recognition area, converts the extracted pixel blocks into a binary matrix, calls the matrix decoding logic to perform information array parsing on the binary matrix, decodes and obtains the unique identification number of the target cutting tooth mechanism 1 corresponding to the surface structured visual encoding of the identification information carrier 16, and extracts the pixel-level two-dimensional contour mask of the target cutting tooth mechanism 1 with the geometric center coordinates of the external geometric contour of the protective cover 15 as the search reference point.
[0071] The mining explosion-proof industrial control computer corrects the two-dimensional contour mask of the target cutting tooth mechanism 1 to a preset reference viewpoint and reference scale. The state assessment unit compares the corrected two-dimensional contour mask with the standard contour mask of the same model cutting tooth mechanism 1 under undamaged state, and calculates the wear area ratio, chipping index and symmetry deviation of the target cutting tooth mechanism 1. The state assessment unit normalizes and maps the wear area ratio, chipping index and symmetry deviation to the corresponding numerical range to obtain the normalized wear area ratio, normalized chipping index and normalized symmetry deviation. The state assessment unit reads the pre-calibrated weight coefficients and calculates the comprehensive health score of the target cutting tooth mechanism 1 according to the weighted linear combination model.
[0072] The mining explosion-proof industrial control computer synchronously acquires the operating parameters of the cutting motor and the instantaneous speed of the cutting head 6, and calculates the current equivalent output load of the cutting head 6. The mining explosion-proof industrial control computer determines whether the target cutting tooth mechanism 1 is within the effective cutting area based on the current position of the cutting head 6, and includes the cutting tooth mechanism 1 within the effective cutting area into the effective cutting tooth mechanism 1 set. The mining explosion-proof industrial control computer reads the installation radius of the target cutting tooth mechanism 1 and the corresponding preset tooth position load coefficient, and determines the initial load weight by combining the angular position correction coefficient. The mining explosion-proof industrial control computer normalizes the initial load weight to obtain the load allocation weight, and allocates the equivalent output load of the cutting head 6 to the target cutting tooth mechanism 1 according to the load allocation weight, thereby obtaining the instantaneous equivalent cutting load of the target cutting tooth mechanism 1. The mining explosion-proof industrial control computer performs time accumulation calculation on the instantaneous equivalent cutting load according to the sampling time sequence to generate cumulative cutting load data. The life cycle management unit writes the cumulative cutting load data into the corresponding life cycle data file.
[0073] The life prediction unit reads the historical comprehensive health score sequence and cumulative cutting load sequence of the target cutting tooth mechanism 1 stored in the life cycle database according to the detection time sequence based on the unique identification number, combines them according to the timestamp, and constructs a multi-dimensional time series feature matrix using a sliding time window. The life prediction unit inputs the multi-dimensional time series feature matrix into the trained time series data analysis model, extracts time series features of the health status change and cumulative cutting load change of the target cutting tooth mechanism 1, and outputs the remaining service life prediction value of the target cutting tooth mechanism 1. The life cycle management unit writes the remaining service life prediction value into the corresponding life cycle data file.
[0074] The intelligent decision-making unit reads the pre-configured safe life threshold and economic life threshold for the current cutting tooth mechanism 1 model. When the predicted remaining life of the target cutting tooth mechanism 1 is between the safe life threshold and the economic life threshold, the intelligent decision-making unit obtains the mine equipment maintenance schedule and planned shutdown window information. When the intelligent decision-making unit determines that there is a planned shutdown window that meets the conditions for tooth replacement operation, it generates a planned window replacement instruction. The mine explosion-proof industrial control computer sends the maintenance decision dataset containing a unique identification number, three-dimensional spatial coordinates, comprehensive health score, predicted remaining life, and replacement decision instruction to the mine human-machine interaction terminal. The mine human-machine interaction terminal generates a highlighted label on the tooth position node corresponding to the cutting head 6 model, and overlays the unique identification number, comprehensive health score, predicted remaining life, and planned window replacement instruction of the target cutting tooth mechanism 1. The on-site maintenance personnel complete the replacement operation of the target cutting tooth mechanism 1 during the planned shutdown window based on the output information of the mine human-machine interaction terminal.
[0075] While specific embodiments of the present invention have been described above, those skilled in the art should understand that these specific embodiments are merely illustrative. Those skilled in the art can omit, substitute, and modify the details of the above methods and systems in various ways without departing from the principles and essence of the present invention. For example, combining the above method steps to perform substantially the same function and achieve substantially the same result using substantially the same method falls within the scope of the present invention.
Claims
1. A lifecycle tracking and intelligent replacement decision-making system for mining tunneling machine cutting teeth, characterized in that the system... include: The identification component is set on the cutting tooth mechanism (1) around the cutting head (6), and includes an identification information carrier (16) with a structured visual code on its surface. The structured visual code is used to characterize the unique identity information of the target cutting tooth mechanism (1). The mining explosion-proof visual acquisition mechanism (3) is distributed toward the cutting head (6) and is used to acquire multi-view image sequences of the cutting head (6); The mine explosion-proof industrial control computer is connected to the mine explosion-proof visual acquisition mechanism (3) and the tunneling machine onboard system. It is used to register and stitch the multi-view image sequence to generate a correction image, identify the structured visual code in the correction image to obtain a unique identification number, and extract the two-dimensional contour mask of the target cutting tooth mechanism (1) to calculate the comprehensive health score. It is also used to calculate the cumulative cutting load data of the target cutting tooth mechanism (1) based on the operating parameters output by the tunneling machine onboard system, obtain the historical comprehensive health score sequence and cumulative cutting load sequence corresponding to the target cutting tooth mechanism (1) based on the unique identification number, output the remaining service life prediction value based on the historical comprehensive health score sequence and cumulative cutting load sequence, and generate a replacement decision instruction in combination with the obtained planned shutdown window information; The mining human-machine interaction terminal is connected to the mining explosion-proof industrial control computer and is used to locate the target cutting tooth mechanism (1) in the display interface and output maintenance prompt information containing the replacement decision command.
2. The mining tunneling machine cutting tooth lifecycle tracking and intelligent replacement decision system according to claim 1, characterized in that, The cutting tooth mechanism (1) includes a cutting tooth body (13) and a tooth tip (11) disposed at the working end of the cutting tooth body (13), wherein an alloy head (12) is embedded on the tooth tip (11); the identification component is disposed in the non-direct cutting area of the cutting tooth body (13) and also includes a protective cover (15), a set screw (14) and a sealing ring (17). The identification information carrier (16) is embedded in the cavity on the surface of the cutting tooth body (13); the protective cover (15) covers the outside of the identification information carrier (16) and is fixed to the cutting tooth body (13) by the set screw (14); the sealing ring (17) is pressed between the protective cover (15) and the cutting tooth body (13) for sealing; the protective cover (15) forms a light-transmitting recognition area at least in the area corresponding to the structured visual code, and the light-transmitting recognition area is formed of transparent wear-resistant material so that the mining explosion-proof visual acquisition mechanism (3) can obtain the structured visual code image through the light-transmitting recognition area.
3. The mining tunneling machine cutting tooth lifecycle tracking and intelligent replacement decision system according to claim 1, characterized in that, It also includes a vibration damping mounting mechanism (4) fixed to the side of the cutting arm (5) and an explosion-proof dust removal protection mechanism (2) installed on the vibration damping mounting mechanism (4); The vibration damping installation mechanism (4) includes an upper mounting plate (41), a lower mounting plate (45), and a support body (43) connecting the upper mounting plate (41) and the lower mounting plate (45). The vibration damping installation mechanism (4) is fixed by anti-loosening fasteners (44), and a high-damping vibration damping pad (42) is provided between the upper mounting plate (41) and the lower mounting plate (45). The mining explosion-proof visual acquisition mechanism (3) is housed in the internal space of the explosion-proof protective cover (23); the mining explosion-proof visual acquisition mechanism (3) specifically includes an explosion-proof protective cover (31), a tempered glass window (35), a camera mounting base (36), and a main lighting source (32), an auxiliary supplementary lighting source (33), and an explosion-proof industrial camera (34) mounted on the camera mounting base (36); The explosion-proof dust removal and protection mechanism (2) includes the explosion-proof protective cover housing (23), the explosion-proof window (210), the narrow-slit high-pressure air knife dust removal nozzle (21), the compressed air distribution pipeline (22), the micro-positive pressure air inlet (24), the wiper motor (25), the micro electric wiper (27), the wiper blade (26), the wiper mounting base (28), and the sealing assembly (29); the explosion-proof window (210) is embedded in the front side of the explosion-proof protective cover housing (23) and is externally sealed by the sealing assembly (29); the micro-positive pressure air inlet (24) is connected to the inner cavity of the explosion-proof protective cover housing (23); The compressed air distribution pipeline (22) is laid outside the explosion-proof protective cover housing (23), with one end connected to the air source and the other end connected to the narrow-slit high-pressure air knife dust removal nozzle (21) to form a continuous air curtain on the outer surface of the explosion-proof window (210); the mini electric wiper (27) is installed through the wiper mounting base (28) and driven by the wiper motor (25) to drive the wiper rubber strip (26) to adhere to the outer surface of the explosion-proof window (210) for wiping.
4. A method for tracking and intelligently replacing the lifecycle of cutting picks in mining tunneling machines, using the lifecycle tracking and intelligent replacement decision-making system for cutting picks in mining tunneling machines as described in any one of claims 1 to 3, characterized in that... The method includes the following steps: A multi-view image sequence of the cutting head (6) is acquired and registered and stitched to generate a corrected image; the structured visual code of the surface of the target cutting tooth mechanism (1) is identified in the corrected image, a unique identification number is obtained, the two-dimensional contour mask of the target cutting tooth mechanism (1) is extracted and the spatial coordinates are determined; the two-dimensional contour mask is corrected and compared with the standard contour mask of the same type of cutting tooth mechanism (1) in the undamaged state, feature parameters are extracted and a comprehensive health score is calculated; The operating parameters output by the tunneling machine's onboard system are obtained to determine the current equivalent output load of the cutting head (6). The load is allocated to the target cutting tooth mechanism (1) according to the load allocation weight to obtain the instantaneous equivalent cutting load, and the cumulative cutting load data is generated by cumulative calculation. The historical comprehensive health score sequence and cumulative cutting load sequence corresponding to the target cutting tooth mechanism (1) are obtained according to the unique identification number. A multi-dimensional time series feature matrix is constructed according to the obtained sequence, and the remaining service life prediction value is output by inputting the trained time series data analysis model. The remaining service life prediction value is compared with the preset service life threshold, and a replacement decision instruction is generated by combining the obtained planned shutdown window information; the spatial coordinates and replacement decision instruction are mapped to the display interface of the mining human-machine interaction terminal, and the positioning guidance screen and maintenance prompt information are output.
5. The method for lifecycle tracking and intelligent replacement decision-making of mining tunneling machine cutting teeth according to claim 4, characterized in that, The steps of identifying the structured visual code of the surface of the target cutting tooth mechanism (1) in the corrected image, obtaining the unique identification number, extracting the two-dimensional contour mask of the target cutting tooth mechanism (1), and determining the spatial coordinates include: Search for candidate regions that match the external geometric contour of the protective cover (15), extract the pixel blocks in the candidate regions that correspond to the light-transmitting recognition region, convert them into a binary matrix using a dynamic grayscale threshold and parse them to obtain the unique identification number; take the geometric center coordinates of the external geometric contour of the protective cover (15) as the reference point, extract the connected edge pixel point set and perform hole filling and boundary smoothing processing to generate the two-dimensional contour mask. Based on the unique identification number, query the corresponding preset three-dimensional installation coordinates, and combine the camera calibration parameters, the current pose of the cutting head (6) and the detection position to perform position correction on the preset three-dimensional installation coordinates to obtain the spatial coordinates of the target cutting tooth mechanism (1).
6. The method for lifecycle tracking and intelligent replacement decision-making of mining tunneling machine cutting teeth according to claim 4, characterized in that, The steps of comparing the corrected two-dimensional contour mask with the standard contour mask of the same type of cutting tooth mechanism (1) in a non-destructive state, extracting feature parameters, and calculating the comprehensive health score include: The two-dimensional contour mask is corrected to the same preset reference viewpoint and scale as the standard contour mask; the current contour is compared with the standard contour, and the wear area ratio, chipping index and symmetry deviation are extracted; the wear area ratio, chipping index and symmetry deviation are mapped to the corresponding intervals using the maximum and minimum value normalization method to obtain normalization parameters; the pre-calibrated weight coefficients are read, and the comprehensive health score is calculated according to the weighted linear combination model.
7. The method for lifecycle tracking and intelligent replacement decision-making of mining tunneling machine cutting teeth according to claim 4, characterized in that, The steps of obtaining the operating parameters output by the tunneling machine's onboard system to determine the current equivalent output load of the cutting head (6), allocating it to the target cutting tooth mechanism (1) according to the load allocation weight to obtain the instantaneous equivalent cutting load, and accumulating and calculating to generate cumulative cutting load data include: The effective cutting area at the current sampling time is determined based on the current pose of the cutting head (6), the relative position of the cutting head (6) and the current coal and rock working face, and the structural parameters of the cutting head (6). Based on the spatial coordinates of each cutting tooth mechanism (1) in the coordinate system of the cutting head (6) and the positional relationship between the effective cutting area, the cutting tooth mechanism located in the effective cutting area is determined as the effective cutting cutting tooth mechanism at the current sampling time, forming a set of effective cutting tooth mechanisms. The initial load weight is determined based on the installation radius of each effective cutting tooth mechanism, the preset tooth position load coefficient, and the angular position correction coefficient, and the load distribution weight is obtained by normalization. The current equivalent output load is allocated to each effective cutting tooth mechanism (1) according to the load allocation weight, and the instantaneous equivalent cutting load of the target cutting tooth mechanism (1) at the current sampling time is obtained; the instantaneous equivalent cutting load is accumulated with the sampling time interval and superimposed with the previously saved cumulative cutting load to generate cumulative cutting load data.
8. The method for lifecycle tracking and intelligent replacement decision-making of mining tunneling machine cutting teeth according to claim 4, characterized in that, The steps of obtaining the historical comprehensive health score sequence and cumulative cutting load sequence corresponding to the target cutting tooth mechanism (1) based on the unique identification number, constructing a multi-dimensional time series feature matrix based on the obtained sequence, and inputting the trained time series data analysis model to output the remaining service life prediction value include: Based on the unique identification number, the historical comprehensive health score sequence and the cumulative cut-off load sequence are read from the life cycle database; the historical comprehensive health score sequence and the cumulative cut-off load sequence are matched according to the timestamp, and a multi-dimensional time series feature matrix is constructed using a sliding time window; the multi-dimensional time series feature matrix is input into the trained time series data analysis model to perform time series analysis on the state degradation trend and output the remaining service life prediction value.
9. The method for lifecycle tracking and intelligent replacement decision-making of mining tunneling machine cutting teeth according to claim 4, characterized in that, The step of comparing the remaining service life prediction value with a preset service life threshold and generating a replacement decision instruction in combination with the obtained planned downtime window information includes: comparing the remaining service life prediction value with a pre-configured safe service life threshold and an economic service life threshold, wherein the safe service life threshold is less than the economic service life threshold. When the predicted remaining service life is less than or equal to the safe service life threshold, an immediate replacement instruction is generated; when the predicted remaining service life is between the safe service life threshold and the economic service life threshold, mine equipment maintenance schedule and planned shutdown window information are obtained; if there is a planned shutdown window whose start time is within the allowable waiting time range and whose duration meets the requirements for gear replacement operation, a planned window replacement instruction is generated; otherwise, a continue observation instruction is generated; when the predicted remaining service life is greater than the economic service life threshold, a continue observation instruction is generated.
10. The method for lifecycle tracking and intelligent replacement decision-making of mining tunneling machine cutting teeth according to claim 4, characterized in that, The steps of mapping the spatial coordinates and replacement decision commands to the display interface of the mining human-machine interface terminal, and outputting the positioning guidance screen and maintenance prompts, include: The maintenance decision dataset containing the unique identification number, spatial coordinates, comprehensive health score, remaining service life prediction value and replacement decision instruction of the target cutting tooth mechanism (1) is analyzed; when the mining human-machine interaction terminal displays the three-dimensional model of the cutting head (6), the corresponding tooth node in the three-dimensional model of the cutting head (6) is determined according to the tooth position installation parameters to establish the position correspondence; a highlight mark is generated at the corresponding display position, and the unique identification number, comprehensive health score, remaining service life prediction value and replacement decision instruction are superimposed and displayed.
