Intelligent wear monitoring system and method for mill lining plate based on big data

Abstract

The invention relates to the technical field of mill lining plate wear monitoring, and discloses a mill lining plate intelligent wear monitoring system and method based on big data, the system comprises a mill lining plate wear monitoring parameter acquisition module, a mill lining plate wear failure state evaluation module, and a mill lining plate wear monitoring information feedback module; accurately matching the wear failure state spatial three-dimensional model library required by the monitored mill lining plate according to the type of the mill lining plate product by combining the wear failure state spatial three-dimensional model library of different types of mill lining plate products stored based on big data with an intelligent search algorithm and the type identification parameters of the mill lining plate product; intelligent evaluation of the wear failure state of the mill lining plate is carried out based on mill lining plate space three-dimensional model parameters in combination with an intelligent recognition algorithm and a mill lining plate product wear failure state space three-dimensional model library which is customized and matched according to the type of the mill lining plate product, and the wear failure state of the mill lining plate is efficiently and accurately analyzed.

Description

Technical Field

[0001] The present invention relates to the technical field of mill liner wear monitoring, and in particular to a mill liner intelligent wear monitoring system and method based on big data. Background Art

[0002] The mill liner is used to protect the cylinder from direct impact and friction between the grinding body and the material. Different forms of liners can also be used to adjust the movement state of the grinding body to enhance the grinding effect of the grinding body on the material, which helps to improve the grinding efficiency of the mill, increase production and reduce metal consumption. The high manganese steel of the mill liner is gradually replaced by the medium alloy steel liner. However, with the continuous application of the medium alloy steel liner in the mill liner, it has gradually replaced manganese steel and other liners to become the mainstream of market development. In addition to protecting the cylinder, the cylinder liner also affects the movement law of the grinding body. In order to meet the requirements of various working conditions, the shape and material of the liner are also different. When the crushing is the main purpose, the liner is required to have a strong ability to push the grinding body, and the liner should have good impact resistance; when the fine grinding is the main purpose, the protrusion of the liner is relatively small, the pushing effect on the grinding body is weak, the impact is small, and the grinding effect is strong, so the liner is required to have good wear resistance. During use, the mill lining is impacted by the grinding body and the material, causing the mill lining to wear. Therefore, it is necessary to regularly detect the wear status of the mill lining and timely maintain and replace the mill lining in the failed state. Currently, the wear amount of the mill lining surface is measured by the wear sensor point, and the wear characteristics of the mill lining and the wear status are simulated and analyzed by the discrete element simulation algorithm. The existing mill lining wear monitoring cannot truly establish the spatial three-dimensional model characteristics of the mill lining, nor can it efficiently and accurately analyze the wear failure status of the mill lining, and cannot realize visual and efficient feedback of the wear monitoring information of the mill lining, which reduces the accuracy and applicability of the mill lining wear monitoring.

[0003] A Chinese invention patent with announcement number CN115077628B discloses a wear monitoring system and method for a conductive slider of an overhead crane. The system determines the dust impact index of each working period, calculates the correlation degree between two adjacent working periods, and uses the dust impact index of the two adjacent working periods to correct the correlation degree of the two adjacent working periods and obtains the corrected correlation; marks different working periods for classification, analyzes independent working conditions and multiple grouped working conditions including at least two working periods; measures the similarity between the independent working condition and each grouped working condition and analyzes and alarms the independent working condition; the above technical scheme cannot accurately measure the real wear space three-dimensional model of the overhead crane conductive slider, nor can it accurately evaluate the real wear state of the overhead crane conductive slider. Summary of the invention

[0004] 1. Technical issues to be resolved

[0005] In order to solve the problems that the above-mentioned existing mill liner wear monitoring cannot truly establish the spatial three-dimensional model characteristics of the mill liner, cannot efficiently and accurately analyze the wear failure status of the mill liner, cannot realize visual and efficient feedback of the wear monitoring information of the mill liner, and reduces the accuracy and applicability of the mill liner wear monitoring, the above purposes are achieved: efficiently collect the spatial position coordinate parameters of the mill liner, accurately perform the mill liner wear monitoring operation, efficiently establish the spatial three-dimensional model parameters of the mill liner and the appearance image parameters of the mill liner, intelligently identify the type of mill liner, accurately retrieve the target type of mill liner wear failure spatial three-dimensional model library, intelligently analyze the wear failure status of the mill liner, and visualize the feedback of the wear monitoring information of the mill liner.

[0006] (II) Technical solution

[0007] The present invention is implemented by the following technical scheme: a method for intelligent wear monitoring of mill liner, the method comprising the following steps:

[0008] S1, collecting the spatial position coordinate data of the mill liner;

[0009] S2, performing a mill liner wear monitoring operation according to the mill liner spatial position coordinate data;

[0010] S3, synchronously collecting the mill liner space three-dimensional model data and the mill liner appearance image data during the mill liner wear monitoring operation;

[0011] S4, performing product type recognition processing of the mill liner wear monitoring object based on the mill liner appearance image data and the appearance image data of different types of mill liners to generate mill liner product type recognition data;

[0012] S5, performing a search process on the three-dimensional model library of the wear failure state of the mill liner required for mill liner wear monitoring according to the mill liner product type identification data and the three-dimensional model library of the wear failure state of different types of mill liner products, and generating a three-dimensional model library of the wear failure state of the target mill liner product;

[0013] S6, performing wear failure state analysis processing of the mill liner according to the mill liner spatial three-dimensional model data and the wear failure state spatial three-dimensional model library of the target mill liner product, and generating mill liner wear failure state analysis data;

[0014] S7. Construct mill liner wear monitoring result data and execute mill liner wear monitoring information feedback operation.

[0015] Preferably, the operation steps of collecting the spatial position coordinate data of the mill liner are as follows:

[0016] S11, collect the spatial position coordinate parameters of the ball mill liner in the maintenance state online through the position sensor, and generate the mill liner spatial position coordinate data set A = (a 1 ,…,a p ,…,a σ ), p=1,2,3,…,σ; where a p It represents the spatial position coordinate data of the mill liner corresponding to the p-th mill liner collected, σ represents the maximum number of mill liner, the spatial position coordinate data of the mill liner includes the longitude, latitude and altitude of the installation position of the mill liner in the ball mill cylinder, and the inspection and maintenance status represents that the ball mill is in a shutdown inspection and maintenance status.

[0017] The present invention accurately collects the spatial position coordinate information of the mill liner online through the position sensor, so as to provide reliable position information for efficiently executing the mill liner wear operation.

[0018] Preferably, the operation steps of performing the mill liner wear monitoring operation according to the mill liner spatial position coordinate data are as follows:

[0019] S21, using a manipulator to carry a wear monitoring device based on the mill liner spatial position coordinate data set A p The mill liner wear monitoring operation is performed on each mill liner in order according to the number of the mill liner, and the wear monitoring equipment includes a three-dimensional laser scanner and a shooting lens.

[0020] The present invention achieves the effect of improving the efficiency of mill liner wear monitoring by efficiently and accurately performing wear monitoring operations on mill liner under inspection and maintenance status based on the spatial position coordinate parameters of the mill liner in combination with a wear monitoring device carried by a manipulator.

[0021] Preferably, the operation steps of synchronously collecting the mill liner space three-dimensional model data and the mill liner appearance image data in the mill liner wear monitoring operation are as follows:

[0022] S31. During the mill liner wear monitoring operation, a 3D laser scanner is used to scan online to establish the spatial 3D solid model parameters of the contact end between the mill liner and the material and the grinding body, and a mill liner spatial 3D model data set B is generated. 1 ,…,b p ,…,b σ ), where b p Represents the mill liner space three-dimensional model data corresponding to the p-th mill liner;

[0023] Similarly, in the process of performing mill liner wear monitoring, the camera lens is used to collect the appearance feature image parameters of the contact end of the mill liner with the material and the grinding body online, and the mill liner appearance image data set C is generated. 1 ,…,c p ,…,c σ ), where c p Represents the mill lining appearance image data corresponding to the p-th mill lining.

[0024] The present invention collects the real space three-dimensional model and appearance image parameters of the mill liner online through a three-dimensional laser scanner and a shooting lens, providing reliable data support for subsequent scientific analysis of the product type and wear failure analysis of the mill liner, thereby achieving the effect of providing reliable data support for subsequent scientific analysis of the product type and wear failure analysis of the mill liner.

[0025] Preferably, the product type identification processing of the mill liner wear monitoring object is performed based on the mill liner appearance image data and the appearance image data of different types of mill liners, and the operation steps of generating the mill liner product type identification data are as follows:

[0026] S41, when the mill liner space three-dimensional model data set B and the mill liner appearance image data set C are collected, establish different types of mill liner appearance image data sets where d o Represents the appearance image data of different types of mill linings corresponding to the oth mill lining product type, Indicates the maximum value of the number of mill lining product types, the mill lining product types include flat lining type, strip lining type, corrugated lining type, ridge lining type, step lining type, groove lining type, drum lining type, small corrugated lining type and graded lining type; the appearance image data of different types of mill linings represent the appearance feature image information of the contact end of different types of mill linings with materials and grinding bodies;

[0027] S42, using a depth-first search algorithm to search the mill lining plate appearance image data set C for the mill lining plate appearance image data set C. p The different types of mill liner appearance image data d in the different types of mill liner appearance image data set D are ordered according to the number of the mill liner. o Perform appearance image feature matching to search for the mill liner appearance image data c p The matching appearance image data d of the different types of mill linings o The corresponding mill lining product type, and generate the mill lining product type identification data set C' = (c' 1 ,…,c' p ,…,c'σ ), where c' p Represents the mill lining product type identification data corresponding to the p-th mill lining.

[0028] The present invention intelligently identifies the product type of the monitored mill liner by combining the appearance image parameters of different types of mill liner plates stored based on big data with a depth-first search algorithm and the appearance image parameters of the mill liner plates, thereby achieving the effect of accurately and dynamically identifying the product type during the mill liner wear monitoring process.

[0029] Preferably, the operation steps of searching and processing the three-dimensional model library of the wear failure state of the mill liner required for mill liner wear monitoring based on the mill liner product type identification data and the three-dimensional model library of the wear failure state of different types of mill liner products are as follows to generate the three-dimensional model library of the wear failure state of the target mill liner product:

[0030] S51. Establish a collection of spatial three-dimensional model libraries of wear failure states of different types of mill lining products Where E o represents the wear failure state space three-dimensional model library of different types of mill lining products corresponding to the oth mill lining product type, where E o =(e o,1 ,…,e o,k ,…,e o,τ ), k = 1, 2, 3, ..., τ, where e o,k The three-dimensional model library E representing the wear failure state space of different types of mill lining products o The wear failure state spatial three-dimensional model data of different types of mill liner products corresponding to the k-th mill liner wear failure type, τ represents the maximum number of mill liner wear failure types, and the wear failure state spatial three-dimensional model data of different types of mill liner products represent the spatial three-dimensional solid model data of the contact end of different types of mill liner products with materials and grinding bodies in the wear failure state, and the mill liner wear failure types include furrows, pits, cracks and fractures on the surface of the contact end of the mill liner with materials and grinding bodies;

[0031] S52, using a bidirectional search algorithm to search the mill lining product type identification data set C for the mill lining product type identification data set C. p The three-dimensional model library E of the wear failure state space of different types of mill lining products is arranged in order according to the number of mill lining products. o Perform mill lining product type keyword matching to search for the mill lining product type identification data c' p The corresponding three-dimensional model library E of the wear failure state space of different types of mill liner products o, and construct the target mill liner product wear failure state space three-dimensional model library set F = (F 1 ,…,F p ,…,F σ ), where F p represents the target mill liner product wear failure state space three-dimensional model library required for the pth mill liner, where F p =(f p,1 ,…,f p,u ,…,f p,υ ), u=1,2,3,…,υ, where f p,u The three-dimensional model library F representing the wear failure state space of the target mill liner product p The u-th target mill lining product wear failure state space three-dimensional model data, υ represents the maximum number of target mill lining product wear failure state space three-dimensional models.

[0032] The present invention realizes accurate matching of the wear failure state spatial three-dimensional model library required for the monitored mill liner according to the mill liner product type through a wear failure state spatial three-dimensional model library of different types of mill liner products stored based on big data, combined with a bidirectional search algorithm and mill liner product type identification parameters, thereby achieving the effect of improving the efficiency and reliability of mill liner wear monitoring.

[0033] Preferably, the wear failure state analysis of the mill liner is performed based on the mill liner spatial three-dimensional model data and the target mill liner product wear failure state spatial three-dimensional model library, and the operation steps for generating the mill liner wear failure state analysis data are as follows:

[0034] S61, obtaining the mill liner spatial three-dimensional model data set B and the target mill liner product wear failure state spatial three-dimensional model library set F;

[0035] S62, the mill liner space three-dimensional model data set B is p The target mill liner product wear failure state space three-dimensional model library F in the target mill liner product wear failure state space three-dimensional model library set F is ordered according to the number of the mill liner. p The internal three-dimensional model data of the wear failure state of the target mill liner product f p,u Perform spatial three-dimensional entity model feature matching, and generate a mill liner wear failure state analysis data set G according to the spatial three-dimensional entity model feature matching result. The specific operation steps for generating the mill liner wear failure state analysis data set G are as follows:

[0036] S621, initialization, update the maximum number of iterations T, update and generate the individual positions of the white shark population in the failed state identification, and the calculation formula for the individual positions of the white shark population in the failed state identification is as follows: W i,j =ι+Φ×(η-ι), where Wi,j represents the position of failure state identification white shark individual i in j-dimensional space, that is, the position of failure state identification white shark individual i in the search space of the target mill lining product wear failure state space three-dimensional model library set F with spatial dimension σ, η and ι represent the upper limit and lower limit of the target mill lining product wear failure state space three-dimensional model library set F respectively, and Φ represents a random number between (0,1);

[0037] S622, speed update stage, failure state identification white shark individuals search for the three-dimensional model data b of the wear failure state space of the target mill liner product in the search space of the three-dimensional model library set F according to the spatial three-dimensional solid model feature matching. p The target mill liner product wear failure state space three-dimensional model data f p,u The prey moves to sense the wear failure state spatial three-dimensional model data of the target mill liner product f p,u Position, and update its own speed. The calculation formula for updating its own speed is as follows: Where V i,t+1 represents the speed of failure state identification white shark individual i in the search space of the three-dimensional model library set F of the target mill liner wear failure state space after t+1 iterations, V i,t W represents the speed of failure state identification white shark individual i in the search space of the three-dimensional model library set F of the target mill liner wear failure state space after t iterations, best,t represents the optimal position of the failure state identification white shark individual in the search space of the three-dimensional model library set F of the wear failure state space of the target mill liner product after t+1 iterations, W i,t It represents the position of the failure state identification white shark individual i in the search space of the three-dimensional model library set F of the wear failure state space of the target mill liner product after t iterations; It represents the failure state identification white shark individual i after t iterations in the three-dimensional model library set F search space of the target mill liner wear failure state space and the speed V i,t The corresponding position; μ represents the algorithm shrinkage coefficient, κ 1 and κ 2 Respectively represent W best,t and The control coefficient, λ 1 and λ 2 Both represent random numbers between (0,1);

[0038] S623, position update stage, failure state identification white shark individuals in the target mill liner product wear failure state space three-dimensional model library set F search space by moving towards the mill liner space three-dimensional model data b p The optimally matched or suboptimally matched wear failure state spatial three-dimensional model data f of the target mill lining product p,u The prey moves to update its position in the search space of the three-dimensional model library set F of the target mill liner product wear failure state space. The calculation formula for the failure state identification white shark individual position update is as follows: Where W' i,t+1 It represents the position of the failure state identification white shark individual i in the search space of the target mill liner product wear failure state space three-dimensional model library set F after t+1 iterations in the position update phase. Represents a bitwise operator, is a logical vector, ▽ and are all binary vectors, Λ represents the attraction coefficient of the failure state identification white shark individual approaching prey, Indicates failure state identification of white shark individual motion product wave frequency;

[0039] The failure state identification white shark individual moves toward the optimal failure state identification white shark individual position in the target mill liner product wear failure state space three-dimensional model library set F search space to get close to the mill liner space three-dimensional model data b p The most matching three-dimensional model data of the wear failure state of the target mill lining product f p,u The position of the prey, the failure state identification white shark individual position update calculation formula is as follows: W" i,t+1 =W best,t +r 1 ×Ξ×sgn(r 2 -0.5),r 3 <Γ, where W" i,t+1 represents the position of the failure state identification white shark individual i in the search space of the three-dimensional model library set F of the wear failure state space of the target mill liner product after t+1 iterations in the position update phase. 1 、r 2 、r 3 All represent random numbers between (0,1), Ξ represents the distance between the failure state identification white shark individual and the prey, that is, the three-dimensional model library set F of the wear failure state space of the target mill liner product is searched for the three-dimensional model data b of the mill liner space. p The target mill liner product wear failure state space three-dimensional model data f p,u; sgn represents the sign return function, Γ represents the failure state to identify the olfactory and visual parameters of white shark individuals close to the best prey;

[0040] S624, in the fish school behavior stage, the failure state identification white shark population retains the optimal failure state identification white shark individual position in the target mill liner product wear failure state space three-dimensional model library set F search space in the update stage, and updates other failure state identification white shark individual positions according to the optimal failure state identification white shark individual position to obtain the search space of the target mill liner product wear failure state space three-dimensional model library set F to search for the mill liner space three-dimensional model data b p The target mill liner product wear failure state space three-dimensional model data f p,u , the calculation formula for updating the position of the individual white shark in the failure state identification is as follows: P"' i,t+1 =(P' i,t+1 +P” i,t+1 ) / 2Φ, where P'' i,t+1 It indicates that the fish school behavior stage updates other failure states to identify the position of the white shark individual i in the search space of the wear failure state space three-dimensional model library set F of the target mill lining product after t+1 iterations, that is, the three-dimensional model data b of the mill lining space is searched out in the search space of the three-dimensional model library set F of the wear failure state space of the target mill lining product. p The target mill liner product wear failure state space three-dimensional model data f p,u ;

[0041] S625, when the maximum number of iterations is met, output the three-dimensional model data b of the mill liner space p The wear failure state spatial three-dimensional model data of the target mill liner product f p,u The spatial three-dimensional solid model feature matching results are performed, and the mill liner wear failure state analysis data set G = (g 1 ,…,g p ,…,g σ ), where g p represents the mill liner wear failure state analysis data corresponding to the p-th mill liner;

[0042] When b p With f p,u The spatial three-dimensional solid model feature matching is successful, indicating that the pth mill liner in the ball mill is in a wear failure state, and the mill liner wear failure state analysis data g is output. p for failure;

[0043] When b p With f p,uIf the spatial 3D solid model feature is not matched successfully, it means that the pth mill liner in the ball mill is not in the wear failure state, then the mill liner wear failure state analysis data g is output. p is normal.

[0044] The present invention performs intelligent evaluation of the wear failure state of the mill liner based on the three-dimensional model parameters of the mill liner space combined with the Beluga optimization algorithm and the three-dimensional model library of the wear failure state of the mill liner product customized according to the mill liner product type, so as to achieve the effect of efficient and accurate analysis of the wear failure state of the mill liner.

[0045] Preferably, the operation steps of constructing the mill liner wear monitoring result data and performing the mill liner wear monitoring information feedback operation are as follows:

[0046] S71, combining the mill liner spatial position coordinate data set A, the mill liner spatial three-dimensional model data set B and the mill liner wear failure state analysis data set G to construct mill liner wear monitoring result data H = (A, B, G);

[0047] S72, pushing the mill liner wear monitoring result data H to the mill liner wear monitoring platform through the Internet of Things communication network and displaying and outputting it through a display screen.

[0048] The present invention scientifically constructs the mill liner wear monitoring result information, pushes it to the mill liner wear monitoring platform online in combination with the Internet of Things communication network, and cooperates with the display screen to intuitively output and display it, so as to achieve the effect of autonomous and efficient collection and processing of mill liner wear monitoring information and visual rapid feedback.

[0049] A mill liner intelligent wear monitoring system based on big data, used to implement the mill liner intelligent wear monitoring method, the system includes a mill liner wear monitoring parameter acquisition module, a mill liner wear failure state assessment module, and a mill liner wear monitoring information feedback module;

[0050] The mill liner wear monitoring parameter acquisition module includes a mill liner space position coordinate acquisition unit, a mill liner wear monitoring operation execution unit, a mill liner space three-dimensional model establishment unit, and a mill liner appearance image acquisition unit;

[0051] The mill liner spatial position coordinate acquisition unit acquires the mill liner spatial position coordinate data through a position sensor; the mill liner wear monitoring operation execution unit executes the mill liner wear monitoring operation according to the mill liner spatial position coordinate data in combination with the wear monitoring equipment carried by the manipulator; the mill liner spatial three-dimensional model establishment unit acquires the mill liner spatial three-dimensional model data through a three-dimensional laser scanner; the mill liner appearance image acquisition unit acquires the mill liner appearance image data through a shooting lens;

[0052] The mill liner wear failure state assessment module includes a different type of mill liner appearance image storage unit, a mill liner type identification unit, a different type of mill liner product wear failure state space three-dimensional model library storage unit, a target mill liner product wear failure state space three-dimensional model library search unit, and a mill liner wear failure state analysis unit;

[0053] The different types of mill lining appearance image storage unit is based on the different types of mill lining appearance image data stored in big data; the mill lining type identification unit performs product type identification processing on the mill lining wear monitoring object based on the mill lining appearance image data and the different types of mill lining appearance image data to generate mill lining product type identification data; the different types of mill lining product wear failure state space three-dimensional model library storage unit is based on the different types of mill lining product wear failure state space three-dimensional model library stored in big data; the target mill lining product wear failure state space three-dimensional model library The three-dimensional model library search unit performs a search process on the three-dimensional model library of the mill liner failure state required for mill liner wear monitoring according to the mill liner product type identification data and the three-dimensional model library of the wear failure state of different types of mill liner products, and generates a three-dimensional model library of the wear failure state of the target mill liner product; the mill liner wear failure state analysis unit performs a wear failure state analysis process on the mill liner according to the three-dimensional model data of the mill liner space and the three-dimensional model library of the wear failure state of the target mill liner product, and generates mill liner wear failure state analysis data;

[0054] The mill liner wear monitoring information feedback module includes a mill liner wear monitoring result information building unit and a mill liner wear monitoring information feedback operation execution unit;

[0055] The mill liner wear monitoring result information construction unit is used to construct the mill liner wear monitoring result data; the mill liner wear monitoring information feedback operation execution unit executes the mill liner wear monitoring information feedback operation according to the mill liner wear monitoring result data.

[0056] (III) Beneficial effects

[0057] The present invention provides a system and method for intelligent wear monitoring of mill liner based on big data, which has the following beneficial effects:

[0058] 1. Accurately collect the spatial position coordinate information of the mill liner online through the position sensor, providing reliable position information for the efficient execution of mill liner wear operations; based on the spatial position coordinate parameters of the mill liner combined with the wear monitoring equipment carried by the manipulator, the wear monitoring operation of the mill liner under maintenance is efficiently and accurately performed, thereby improving the efficiency of mill liner wear monitoring; the real space 3D model and appearance image parameters of the mill liner are collected online through the 3D laser scanner and shooting lens, providing reliable data support for the subsequent scientific analysis of the product type and wear failure analysis of the mill liner, thereby improving the accuracy of mill liner wear monitoring.

[0059] 2. Intelligent identification of the product type of the monitored mill lining is performed by combining the appearance image parameters of different types of mill linings based on big data storage with intelligent search algorithms and mill lining appearance image parameters, so as to realize accurate and dynamic identification of product types in the mill lining wear monitoring process and improve the accuracy of mill lining wear monitoring; the wear failure state spatial three-dimensional model library of different types of mill lining products based on big data storage combined with intelligent search algorithms and mill lining product type identification parameters is used to accurately match the wear failure state spatial three-dimensional model library required for the monitored mill lining according to the mill lining product type, thereby improving the efficiency and reliability of mill lining wear monitoring; intelligent evaluation of the wear failure state of the mill lining is performed based on the mill lining spatial three-dimensional model parameters combined with intelligent identification algorithms and the wear failure state spatial three-dimensional model library of the mill lining product customized according to the mill lining product type, thereby realizing efficient and accurate analysis of the wear failure state of the mill lining and improving the intelligence and quality of mill lining wear monitoring.

[0060] 3. Through scientifically constructing the mill lining wear monitoring result information based on the mill lining position information, three-dimensional solid model and wear failure analysis result information combined with data processing, the mill lining wear monitoring result information can be collected and processed independently and efficiently; based on the mill lining wear monitoring result information combined with the Internet of Things communication network, it is pushed online to the mill lining wear monitoring platform and displayed intuitively on the display screen, so as to realize the visual rapid feedback of mill lining wear monitoring and improve the efficiency and applicability of mill lining wear monitoring operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 A schematic diagram of a module of a mill liner intelligent wear monitoring system based on big data provided by the present invention;

[0062] Figure 2 The present invention provides a flow chart of a method for intelligent wear monitoring of a mill liner. DETAILED DESCRIPTION

[0063] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0064] The embodiment of the intelligent wear monitoring system and method for mill liner based on big data is as follows:

[0065] Embodiment 1:

[0066] See also Figure 1 - Figure 2 , a method for intelligent wear monitoring of mill liner, the method comprising the following steps:

[0067] S1, collecting the spatial position coordinate data of the mill liner;

[0068] S2. Perform mill liner wear monitoring operations based on the mill liner spatial position coordinate data;

[0069] S3, synchronously collecting the mill liner space three-dimensional model data and the mill liner appearance image data during the mill liner wear monitoring operation;

[0070] S4, performing product type recognition processing of the mill liner wear monitoring object based on the mill liner appearance image data and the appearance image data of different types of mill liners, and generating mill liner product type recognition data;

[0071] S5, searching and processing the three-dimensional model library of the wear failure state of the mill liner required for mill liner wear monitoring according to the mill liner product type identification data and the three-dimensional model library of the wear failure state of different types of mill liner products, and generating the three-dimensional model library of the wear failure state of the target mill liner product;

[0072] S6, performing wear failure state analysis and processing of the mill liner according to the mill liner space three-dimensional model data and the target mill liner product wear failure state space three-dimensional model library, and generating mill liner wear failure state analysis data;

[0073] S7. Construct mill liner wear monitoring result data and execute mill liner wear monitoring information feedback operation.

[0074] For further information, see Figure 1 - Figure 2 The steps for collecting the spatial position coordinate data of the mill liner are as follows:

[0075] S11, collect the spatial position coordinate parameters of the ball mill liner in the maintenance state online through the position sensor, and generate the mill liner spatial position coordinate data set A = (a 1 ,…,a p ,…,a σ ), p=1,2,3,…,σ; where a p It represents the spatial position coordinate data of the mill liner corresponding to the p-th mill liner collected, σ represents the maximum number of mill liner, the spatial position coordinate data of the mill liner includes the longitude, latitude and altitude of the installation position of the mill liner in the ball mill cylinder, and the inspection and maintenance status represents that the ball mill is in the state of shutdown inspection and maintenance.

[0076] The operating steps for performing mill liner wear monitoring operations based on the mill liner spatial position coordinate data are as follows:

[0077] S21, using a manipulator equipped with a wear monitoring device based on the mill liner spatial position coordinate data set A p The mill lining wear monitoring operation is performed on each mill lining in order according to the number of the mill linings. The wear monitoring equipment includes a three-dimensional laser scanner and a shooting lens.

[0078] The steps for simultaneously collecting the mill liner space 3D model data and the mill liner appearance image data in the mill liner wear monitoring operation are as follows:

[0079] S31. During the mill liner wear monitoring operation, a 3D laser scanner is used to scan online to establish the spatial 3D solid model parameters of the contact end between the mill liner and the material and the grinding body, and a mill liner spatial 3D model data set B is generated. 1 ,…,b p ,…,b σ ), where b p Represents the mill liner space three-dimensional model data corresponding to the p-th mill liner;

[0080] Similarly, in the process of performing mill liner wear monitoring, the camera lens is used to collect the appearance feature image parameters of the contact end of the mill liner with the material and the grinding body online, and the mill liner appearance image data set C is generated. 1 ,…,c p ,…,c σ ), where c p Represents the mill lining appearance image data corresponding to the p-th mill lining.

[0081] Through the mill liner spatial position coordinate acquisition unit, the position sensor is used to accurately collect the mill liner spatial position coordinate information online, providing reliable position information for the efficient execution of mill liner wear operations; the mill liner wear monitoring operation execution unit, based on the mill liner spatial position coordinate parameters combined with the manipulator equipped with wear monitoring equipment, efficiently and accurately performs wear monitoring operations on the mill liner under maintenance status, thereby improving the efficiency of mill liner wear monitoring; the mill liner spatial three-dimensional model establishment unit and the mill liner appearance image acquisition unit cooperate with each other, and the real space three-dimensional model and appearance image parameters of the mill liner are collected online through a three-dimensional laser scanner and a shooting lens, providing reliable data support for the subsequent scientific analysis of the product type and wear failure analysis of the mill liner, thereby improving the accuracy of mill liner wear monitoring.

[0082] For further information, see Figure 1 - Figure 2 Based on the mill liner appearance image data and the appearance image data of different types of mill liners, the product type recognition processing of the mill liner wear monitoring object is performed, and the operation steps for generating the mill liner product type recognition data are as follows:

[0083] S41. When the mill liner space 3D model data set B and the mill liner appearance image data set C are collected, different types of mill liner appearance image data sets are established. where d o represents the appearance image data of different types of mill linings corresponding to the oth mill lining product type, ζ represents the maximum number of mill lining product types, and the mill lining product types include flat lining type, strip lining type, corrugated lining type, ridge lining type, step lining type, groove lining type, drum lining type, small corrugated lining type and graded lining type; the appearance image data of different types of mill linings represent the appearance feature image information of the contact end of different types of mill linings with materials and grinding bodies;

[0084] S42, using a depth-first search algorithm to search the mill lining appearance image data set C for the mill lining appearance image data c p The appearance image data d of different types of mill liners in the appearance image data set D are ordered according to the number of mill liners. o Perform appearance image feature matching to search for the mill lining appearance image data c p Matching different types of mill liner appearance image data o The corresponding mill lining product type, and generate the mill lining product type identification data set C' = (c' 1 ,…,c' p ,…,c' σ ), where c' pRepresents the mill lining product type identification data corresponding to the p-th mill lining.

[0085] According to the mill liner product type identification data and the wear failure state space 3D model library of different types of mill liner products, the mill liner failure state space 3D model library required for mill liner wear monitoring is searched and processed, and the operation steps for generating the target mill liner product wear failure state space 3D model library are as follows:

[0086] S51. Establish a three-dimensional model library set of wear failure states of different types of mill lining products E = (E 1 ,…,E o ,…,E ζ ), where E o represents the wear failure state space three-dimensional model library of different types of mill lining products corresponding to the oth mill lining product type, where E o =(e o,1 ,…,e o,k ,…,e o,τ ), k = 1, 2, 3, ..., τ, where e o,k Represents the wear failure state space three-dimensional model library E of different types of mill liner products o The wear failure state spatial three-dimensional model data of different types of mill liner products corresponding to the k-th mill liner wear failure type, τ represents the maximum number of mill liner wear failure types, and the wear failure state spatial three-dimensional model data of different types of mill liner products represent the spatial three-dimensional solid model data of the contact end of different types of mill liner products with materials and grinding bodies in the wear failure state, and the mill liner wear failure types include furrows, pits, cracks and fractures on the surface of the contact end of the mill liner with materials and grinding bodies;

[0087] S52, using a bidirectional search algorithm to identify the mill lining product type in the mill lining product type identification data set C

[0088] Datac p The wear failure state space 3D model library E of different types of mill lining products is ordered according to the number of mill lining products. o Mill lining product type

[0089] Keyword matching, search out mill lining product type identification data c p Corresponding three-dimensional model library E of wear failure state space of different types of mill lining products o , and construct the target mill liner product wear failure state space three-dimensional model library set F = (F 1 ,…,F p ,…,F σ), where F p represents the target mill liner product wear failure state space three-dimensional model library required for the pth mill liner, where F p =(f p,1 ,…,f p,u ,…,f p,υ ), u=1,2,3,…,υ, where f p,u The three-dimensional model library F representing the wear failure state space of the target mill liner product p The u-th target mill lining product wear failure state space three-dimensional model data, υ represents the maximum number of target mill lining product wear failure state space three-dimensional models.

[0090] According to the mill liner space three-dimensional model data and the target mill liner product wear failure state space three-dimensional model library, the mill liner wear failure state analysis is carried out. The operation steps for generating mill liner wear failure state analysis data are as follows:

[0091] S61, obtaining a mill liner spatial three-dimensional model data set B and a target mill liner product wear failure state spatial three-dimensional model library set F;

[0092] S62, the mill liner space three-dimensional model data set B is used to p The target mill liner product wear failure state space 3D model library F in the target mill liner product wear failure state space 3D model library F is sorted according to the number of mill liner products. p The internal target mill liner product wear failure state space 3D model data f p,u Perform spatial three-dimensional entity model feature matching, and generate a mill liner wear failure state analysis data set G according to the spatial three-dimensional entity model feature matching result. The specific operation steps for generating the mill liner wear failure state analysis data set G are as follows:

[0093] S621, initialization, update the maximum number of iterations T, update and generate the individual positions of the white shark population in the failed state identification, and the calculation formula for the individual positions of the white shark population in the failed state identification is as follows: W i,j =ι+Φ×(η-ι), where W i,j represents the position of failure state identification white shark individual i in the j-dimensional space, that is, the position of failure state identification white shark individual i in the search space of the target mill lining product wear failure state space three-dimensional model library set F with a spatial dimension of σ, η and ι represent the upper and lower limits of the target mill lining product wear failure state space three-dimensional model library set F respectively, and Φ represents a random number between (0,1);

[0094] S622, speed update stage, failure state identification white shark individuals search for the mill liner space three-dimensional model data b in the target mill liner wear failure state space three-dimensional model library set F search space based on the spatial three-dimensional solid model feature matching. p The matching target mill liner product wear failure state space three-dimensional model data f p,u Prey moves to sense the wear failure state spatial 3D model data of the target mill liner product f p,u Position, and update its own speed. The calculation formula for updating its own speed is as follows: Where V i,t+1 represents the speed of failure state identification white shark individual i in the search space of the three-dimensional model library set F of the target mill liner wear failure state space after t+1 iterations, V i,t W represents the speed of failure state identification white shark individual i in the search space of the three-dimensional model library set F of the target mill liner wear failure state space after t iterations, best,t It represents the optimal position of the failure state identification white shark individual in the search space of the three-dimensional model library set F of the target mill liner wear failure state space after t+1 iterations, W i,t It represents the position of failure state identification white shark individual i in the search space of the three-dimensional model library set F of the target mill liner product wear failure state space after t iterations; It represents the failure state identification white shark individual i after t iterations in the target mill liner wear failure state space three-dimensional model library set F search space and speed V i,t The corresponding position; μ represents the algorithm shrinkage coefficient, κ 1 and κ 2 Respectively represent W best,t and The control coefficient, λ 1 and λ 2 Both represent random numbers between (0,1);

[0095] S623, position update stage, failure state identification white shark individuals in the target mill liner product wear failure state space three-dimensional model library set F search space by moving towards the mill liner space three-dimensional model data b p The optimal or suboptimal matching target mill liner product wear failure state space three-dimensional model data f p,u The prey moves to update its position in the search space of the three-dimensional model library set F of the target mill liner product wear failure state space. The calculation formula for the failure state recognition white shark individual position update is as follows: Where W' i,t+1It represents the position of the failure state identification white shark individual i in the target mill liner wear failure state space three-dimensional model library set F search space after t+1 iterations in the position update phase. Represents a bitwise operator, is a logical vector, ▽ and are all binary vectors, Λ represents the attraction coefficient of the failure state identification white shark individual approaching prey, Indicates failure state identification of white shark individual motion product wave frequency;

[0096] The failure state identification white shark individual moves toward the optimal failure state identification white shark individual position in the target mill liner product wear failure state space three-dimensional model library set F search space to get close to the mill liner space three-dimensional model data b p The most matching target mill liner product wear failure state space 3D model data f p,u The position of the prey, the failure state identification white shark individual position update calculation formula is as follows: W" i,t+1 =W best,t +r 1 ×Ξ×sgn(r 2 -0.5),r 3 <Γ, where W" i,t+1 represents the position update phase of searching for the optimal failure state identification white shark individual position. The failure state identification white shark individual i is located in the search space of the three-dimensional model library set F of the target mill liner product wear failure state space after t+1 iterations. r 1 、r 2 、r 3 Both represent random numbers between (0,1), Ξ represents the distance between the failure state identification white shark individual and the prey, that is, the three-dimensional model library set F of the wear failure state space of the target mill lining product is searched for the mill lining space three-dimensional model data b p The matching target mill liner product wear failure state space three-dimensional model data f p,u ; sgn represents the sign return function, Γ represents the failure state to identify the olfactory and visual parameters of white shark individuals close to the best prey;

[0097] S624, in the fish school behavior stage, the failure state identification white shark population retains the optimal failure state identification white shark individual position in the target mill lining product wear failure state space three-dimensional model library set F search space in the update stage, and updates other failure state identification white shark individual positions according to the optimal failure state identification white shark individual position to obtain the mill lining product wear failure state space three-dimensional model library set F search space search space search space and the mill lining space three-dimensional model data b. pThe matching target mill liner product wear failure state space three-dimensional model data f p,u , the calculation formula for updating the position of the individual white shark in the failure state identification is as follows: P"' i,t+1 =(P' i,t+1 +P” i,t+1 ) / 2Φ, where P'' i,t+1 It indicates that the fish school behavior stage updates other failure states to identify the position of the white shark individual i in the search space of the target mill lining product wear failure state space three-dimensional model library set F after t+1 iterations, that is, the target mill lining product wear failure state space three-dimensional model library set F search space searches for the mill lining space three-dimensional model data b p The matching target mill liner product wear failure state space three-dimensional model data f p,u ;

[0098] S625. When the maximum number of iterations is met, output the mill liner space three-dimensional model data b p The wear failure state spatial 3D model data of the target mill liner product f p,u The spatial three-dimensional solid model feature matching results are performed, and the mill liner wear failure state analysis data set G = (g 1 ,…,g p ,…,g σ ), where g p represents the mill liner wear failure state analysis data corresponding to the p-th mill liner;

[0099] When b p With f p,u The spatial three-dimensional solid model feature matching is successful, indicating that the pth mill liner in the ball mill is in a wear failure state, and the mill liner wear failure state analysis data g is output. p for failure;

[0100] When b p With f p,u If the spatial 3D solid model feature is not matched successfully, it means that the pth mill liner in the ball mill is not in the wear failure state, then the mill liner wear failure state analysis data g is output. p is normal.

[0101] Through the cooperation of different types of mill lining appearance image storage units and mill lining type identification units, the product type of the monitored mill lining is intelligently identified based on the different types of mill lining appearance image parameters stored in big data combined with the intelligent search algorithm and the mill lining appearance image parameters, so that the mill lining wear monitoring process can accurately and dynamically identify the product type, thereby improving the accuracy of mill lining wear monitoring; the three-dimensional model library storage unit of the wear failure state space of different types of mill lining products and the three-dimensional model library search unit of the wear failure state space of the target mill lining product cooperate with each other, and the wear failure state space of different types of mill lining products stored in big data can be accurately and dynamically identified based on the intelligent search algorithm and the mill lining appearance image parameters. The state space three-dimensional model library is combined with the intelligent search algorithm and the mill lining product type identification parameters to realize the accurate matching of the wear and failure state space three-dimensional model library required for the monitored mill lining according to the mill lining product type, thereby improving the efficiency and reliability of mill lining wear monitoring; the mill lining wear failure state analysis unit performs intelligent evaluation of the mill lining wear failure state based on the mill lining space three-dimensional model parameters combined with the intelligent identification algorithm and the mill lining product wear failure state space three-dimensional model library customized according to the mill lining product type, thereby realizing efficient and accurate analysis of the mill lining wear failure state and improving the intelligence and quality of mill lining wear monitoring.

[0102] For further information, see Figure 1 - Figure 2 The steps for constructing the mill liner wear monitoring result data and performing the mill liner wear monitoring information feedback operation are as follows:

[0103] S71, combining the mill liner spatial position coordinate data set A, the mill liner spatial three-dimensional model data set B and the mill liner wear failure state analysis data set G to construct the mill liner wear monitoring result data H = (A, B, G);

[0104] S72, pushing the mill lining wear monitoring result data H to the mill lining wear monitoring platform through the Internet of Things communication network and displaying and outputting it through a display screen.

[0105] Through the mill lining wear monitoring result information construction unit, the mill lining wear monitoring result information is scientifically constructed based on the mill lining position information, three-dimensional solid model and wear failure analysis result information combined with data processing, so as to realize the independent and efficient collection and processing of mill lining wear monitoring information; the mill lining wear monitoring information feedback operation execution unit is based on the mill lining wear monitoring result information combined with the Internet of Things communication network and pushed online to the mill lining wear monitoring platform and cooperated with the display screen for intuitive output display, so as to realize the visual and rapid feedback of mill lining wear monitoring and improve the efficiency and applicability of mill lining wear monitoring operation.

[0106] Embodiment 2:

[0107] See also Figure 1 - Figure 2 , a mill liner intelligent wear monitoring system based on big data, used to implement a mill liner intelligent wear monitoring method, the system includes a mill liner wear monitoring parameter acquisition module, a mill liner wear failure state assessment module, and a mill liner wear monitoring information feedback module;

[0108] The mill liner wear monitoring parameter acquisition module includes a mill liner space position coordinate acquisition unit, a mill liner wear monitoring operation execution unit, a mill liner space three-dimensional model establishment unit, and a mill liner appearance image acquisition unit;

[0109] The mill liner spatial position coordinate acquisition unit acquires the mill liner spatial position coordinate data through the position sensor; the mill liner wear monitoring operation execution unit performs the mill liner wear monitoring operation based on the mill liner spatial position coordinate data combined with the wear monitoring equipment carried by the manipulator; the mill liner spatial three-dimensional model establishment unit acquires the mill liner spatial three-dimensional model data through the three-dimensional laser scanner; the mill liner appearance image acquisition unit acquires the mill liner appearance image data through the shooting lens;

[0110] The mill liner wear failure state assessment module includes a mill liner appearance image storage unit of different types, a mill liner type identification unit, a mill liner product wear failure state space three-dimensional model library storage unit of different types, a target mill liner product wear failure state space three-dimensional model library search unit, and a mill liner wear failure state analysis unit;

[0111] A unit for storing appearance images of different types of mill liners, based on the appearance image data of different types of mill liners stored in big data; a unit for identifying the type of mill liners, based on the appearance image data of mill liners and the appearance image data of different types of mill liners, to generate mill liners product type identification data; a unit for storing the spatial three-dimensional model library of wear failure states of different types of mill liners, based on the spatial three-dimensional model library of wear failure states of different types of mill liners stored in big data; a unit for searching the spatial three-dimensional model library of wear failure states of target mill liners, to search the spatial three-dimensional model library of mill liners failure states required for mill liners wear monitoring according to the product type identification data of mill liners and the spatial three-dimensional model library of wear failure states of different types of mill liners, to generate the spatial three-dimensional model library of wear failure states of target mill liners; a unit for analyzing the wear failure states of mill liners, based on the spatial three-dimensional model data of mill liners and the spatial three-dimensional model library of wear failure states of target mill liners, to generate the wear failure states analysis data of mill liners;

[0112] The mill liner wear monitoring information feedback module includes a mill liner wear monitoring result information building unit and a mill liner wear monitoring information feedback operation execution unit;

[0113] The mill liner wear monitoring result information construction unit is used to construct the mill liner wear monitoring result data; the mill liner wear monitoring information feedback operation execution unit executes the mill liner wear monitoring information feedback operation according to the mill liner wear monitoring result data.

[0114] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for intelligent wear monitoring of mill liner, characterized in that: The method comprises the following steps: S1, collecting the spatial position coordinate data of the mill liner; S2, performing a mill liner wear monitoring operation according to the mill liner spatial position coordinate data; S3, synchronously collecting the mill liner space three-dimensional model data and the mill liner appearance image data during the mill liner wear monitoring operation; S4, performing product type recognition processing of the mill liner wear monitoring object based on the mill liner appearance image data and the appearance image data of different types of mill liners to generate mill liner product type recognition data; S5, performing a search process on the three-dimensional model library of the wear failure state of the mill liner required for mill liner wear monitoring according to the mill liner product type identification data and the three-dimensional model library of the wear failure state of different types of mill liner products, and generating a three-dimensional model library of the wear failure state of the target mill liner product; S6, performing wear failure state analysis processing of the mill liner according to the mill liner spatial three-dimensional model data and the wear failure state spatial three-dimensional model library of the target mill liner product, and generating mill liner wear failure state analysis data; S7. Construct mill liner wear monitoring result data and execute mill liner wear monitoring information feedback operation.

2. The method for intelligent wear monitoring of mill liner according to claim 1, characterized in that: The S1 comprises the following steps: S11, collect the spatial position coordinate parameters of the ball mill liner under maintenance and repair through the position sensor online, and generate the spatial position coordinate data set A of the ball mill liner = (a1, ..., a p ,…,a σ ), p=1,2,3,…,σ; where a p It represents the spatial position coordinate data of the mill liner corresponding to the p-th mill liner collected, and σ represents the maximum number of mill liners.

3. The method for intelligent wear monitoring of mill liner according to claim 2, characterized in that: The S2 comprises the following steps: S21, using a manipulator to carry a wear monitoring device based on the mill liner spatial position coordinate data set A p Perform mill liner wear monitoring operations on each mill liner in order according to the number of mill liner.

4. The method for intelligent wear monitoring of mill liner according to claim 3 is characterized in that: The S3 comprises the following steps: S31. During the mill liner wear monitoring operation, a 3D laser scanner is used to scan online to establish the spatial 3D solid model parameters of the contact end between the mill liner and the material and the grinding body, and a mill liner spatial 3D model data set B = (b1, ..., b p ,…,b σ ), where b p Represents the mill liner space three-dimensional model data corresponding to the p-th mill liner; Similarly, in the process of performing mill liner wear monitoring, the camera lens is used to collect the appearance feature image parameters of the contact end of the mill liner with the material and the grinding body online, and the mill liner appearance image data set C = (c1,…,c p ,…,c σ ), where c p Represents the mill lining appearance image data corresponding to the p-th mill lining.

5. The method for intelligent wear monitoring of mill liner according to claim 4 is characterized in that: The S4 comprises the following steps: S41. When the acquisition of B and C is completed, a data set of appearance images of different types of mill liners is established. where d o Represents the appearance image data of different types of mill linings corresponding to the oth mill lining product type, Indicates the maximum number of mill lining product types; S42, using a depth-first search algorithm to find the c in C p According to the number of mill liners, they are numbered in order as described in D o Perform appearance image feature matching to search for the image that matches the c p Match the d o The corresponding mill lining product type is generated, and the mill lining product type identification data set C'=(c'1,…,c' p ,…,c' σ ), where c' p Represents the mill lining product type identification data corresponding to the p-th mill lining.

6. The method for intelligent wear monitoring of mill liner according to claim 5, characterized in that: The S5 comprises the following steps: S51. Establish a collection of spatial three-dimensional model libraries of wear failure states of different types of mill lining products Where E o represents the wear failure state space three-dimensional model library of different types of mill lining products corresponding to the oth mill lining product type, where E o =(e o,1 ,…,e o,k ,…,e o,τ ), k＝1,2,3,…,τ , where e o,k Indicates the E o The wear failure state space three-dimensional model data of different types of mill liner products corresponding to the kth mill liner wear failure type, τ represents the maximum number of mill liner wear failure types; S52, using a bidirectional search algorithm to find the c′ in C′ p The mill liners are numbered in order according to the number of E o Perform keyword matching of mill lining product type and search for the c′ p The corresponding E o , and construct a three-dimensional model library set F = (F1,…,F p ,…,F σ ), where F p represents the target mill liner product wear failure state space three-dimensional model library required for the pth mill liner, where F p =(f p,1 ,…,f p,u ,…,f p,υ ), u=1,2,3,…,υ, where f p,u The three-dimensional model library F representing the wear failure state space of the target mill liner product p The u-th target mill lining product wear failure state space three-dimensional model data, υ represents the maximum number of target mill lining product wear failure state space three-dimensional models.

7. The method for intelligent wear monitoring of mill liner according to claim 6, characterized in that: The S6 comprises the following steps: S61, obtaining the B and the F; S62, the b in B p The mill liners are numbered in order according to the number of mill liners and are described in F p Internal f p,u Perform spatial three-dimensional entity model feature matching, and generate a mill liner wear failure state analysis data set G according to the spatial three-dimensional entity model feature matching result. The specific operation steps for generating the mill liner wear failure state analysis data set G are as follows: S621, initialization, updating the maximum number of iterations T, and updating the generation of failure states to identify individual positions of white shark populations; S622, speed update stage, the failure state identification white shark individual searches for the one that matches the b in the F search space according to the spatial three-dimensional entity model feature matching. p Match the f p,u The prey moves to sense the f p,u Position, and update its own speed; S623, position update stage, the failed state identification white shark individual moves in the F search space towards the b p The optimal match or suboptimal match of the f p,u The prey moves its position to update its position in the F search space; The failure state identification white shark individual moves toward the optimal failure state identification white shark individual position in the F search space to get close to the b p The best match for the f p,u The location of the prey; S624, in the fish school behavior stage, the failure state identification white shark population retains the optimal failure state identification white shark individual position in the F search space in the position update stage through the feeding behavior, and updates the other failure state identification white shark individual positions according to the optimal failure state identification white shark individual position to obtain the searched white shark individual position in the F search space that matches the b p Match the f p,u ; S625. When the maximum number of iterations is met, output the b p With the f p,u The spatial three-dimensional solid model feature matching results are performed, and the wear failure state analysis data set of the mill liner is generated G = (g1,…,g p ,…,g σ ), where g p represents the mill liner wear failure state analysis data corresponding to the p-th mill liner; When b p With f p,u If the spatial three-dimensional solid model feature matching is successful, the wear failure state analysis data of the mill liner will be output. p for failure; When b p With f p,u If the spatial 3D solid model feature is not matched successfully, the mill liner wear failure state analysis data g will be output p is normal.

8. The method for intelligent wear monitoring of mill liner according to claim 7, characterized in that: The S7 comprises the following steps: S71, combining the data of A, B and G to construct mill liner wear monitoring result data H=(A, B, G); S72, pushing the H to the mill liner wear monitoring platform through the Internet of Things communication network and displaying and outputting it through a display screen.

9. A mill liner intelligent wear monitoring system based on big data, used to implement a mill liner intelligent wear monitoring method according to any one of claims 1 to 8, characterized in that: The system comprises a mill liner wear monitoring parameter acquisition module, a mill liner wear failure state assessment module, and a mill liner wear monitoring information feedback module.

Images