Method and system for monitoring and performance prediction of gyratory crusher based on digital twin

Abstract

The present invention relates to a method and system for monitoring and predicting the performance of a gyratory crusher based on digital twins, and belongs to the field of performance prediction of mining machinery and equipment. Use multiple sensors to collect the key parameters of the crusher in real time and monitor the state of the crusher; establish the dynamics and discrete element analysis model of the gyratory crusher in combination with the real parameters and working conditions, simulate the crushing process of the gyratory crusher, and obtain the relevant performance parameter data of the crusher; compare and judge the data collected by the sensor to achieve real-time monitoring of the working state of the gyratory crusher; use the state data obtained by the sensor as twin data and input it into the prediction model to predict the performance parameters of the equipment in real time; use the real-time state data of the crusher as input and input it into the crusher dynamics and discrete element coupling simulation model to perform dynamic simulation of the crushing process. Quickly and accurately predict the performance parameters of the gyratory crusher and reflect the working state of the equipment in real time.

Description

Technical Field

[0001] The present invention relates to the technical field of mining machinery and equipment performance prediction, and in particular to a method and system for monitoring and predicting the performance of a gyratory crusher based on digital twins. Background Art

[0002] With the rapid development of my country's industrial economy, the demand for ore raw materials in various industrial sectors is also growing. Most ore raw materials need to be further processed by crushing, grinding and other processes before they can be directly used in various engineering fields. The crushing and grinding process is accompanied by huge energy consumption. Therefore, the performance of crushing equipment is not only related to product quality, but also directly affects the economic cost of the crushing process. At present, the energy consumption of the crushing and grinding process accounts for more than 10% of my country's total energy consumption, and the annual loss of wear-resistant metals exceeds 3 million tons. Compared with the high-cost and low-efficiency grinding process, the power consumption of the crushing operation is only 8% to 12% of the grinding operation. Therefore, achieving more crushing and less grinding has become an effective way to reduce power consumption and improve efficiency.

[0003] As a crushing equipment with the advantages of high production efficiency, stable operation, and low unit energy consumption, large gyratory crushers are often used in the coarse crushing process of ores. The increasing reduction of high-grade ores and the gradual strengthening of people's awareness of environmental resource protection have put forward higher requirements on the performance level of large gyratory crushers. In engineering practice, performance parameters such as the production capacity of the crusher, the product quality of the ore particles after crushing, and the crushing power are important indicators for judging the performance of the gyratory crusher. Therefore, both the manufacturers of crusher equipment and the users of the equipment pay special attention to the performance parameters of the gyratory crusher. It is expensive and difficult to directly obtain data on performance parameters such as productivity, discharge particle size distribution, crushing power, and liner wear during the crushing process. Therefore, numerical simulation and theoretical formula derivation are often used to obtain performance parameters. However, these methods are greatly different from the actual working conditions and cannot provide real-time feedback on the performance level of the gyratory crusher. Therefore, there is an urgent need for a real-time, fast, accurate and efficient performance parameter prediction system and method for gyratory crushers.

[0004] Digital Twin, also known as digital twin. Digital twin technology cleverly combines real physical entities with information data, and uses technologies such as the Internet of Things and big data analysis to process relevant information such as physical model parameters, real-time sensor data, and actual operating conditions. It completes the mapping of the operation process of the physical entity in the virtual world and establishes a "digital twin" corresponding to the physical entity, ultimately achieving the purpose of reflecting the entire life cycle of the corresponding physical equipment. The development of digital twin technology provides new ideas and methods for solving the above problems. Summary of the invention

[0005] The purpose of the present invention is to provide a method and system for monitoring and predicting the performance of a gyratory crusher based on digital twins, which solves the problems of large amount of calculation, high time cost, and poor immediacy when predicting the performance parameters of existing gyratory crushers. The present invention combines three-dimensional geometric models, dynamic simulation analysis, discrete element simulation analysis, real-time monitoring feedback from sensors, data fusion analysis, and self-learning prediction algorithms to establish a digital twin system for a gyratory crusher, which helps to realize the real-time and visualization of the operating status of the equipment, effectively restore the working process of the gyratory crusher, and then accurately predict the performance parameters. The present invention can reflect the working status of the crushing equipment in real time, quickly predict the performance parameters of the gyratory crusher during the crushing process, synchronously simulate the crushing working process of the crusher, and save the state parameter data and performance parameter data to provide parameter basis for subsequent analysis and research of the equipment.

[0006] The above-mentioned object of the present invention is achieved by the following technical solutions:

[0007] The method for monitoring and performance prediction of a gyratory crusher based on digital twin includes the following steps:

[0008] Step S1, determining the collected gyratory crusher status data according to common faults and key working parameters in the gyratory crusher crushing operation;

[0009] Step S2, establishing a three-dimensional geometric model of the equipment and materials according to parameter information of the gyratory crusher and the ore particles to be crushed;

[0010] Step S3: collect the real-time status data of the equipment based on multiple sensors and high-speed cameras and transmit it to the host computer database;

[0011] Step S4: The host computer determines the real-time status data and monitors the equipment operation status;

[0012] Step S5, storing the state parameter group with warning and the state parameter group without warning respectively;

[0013] Step S6, setting the motion relationship between the components of the gyratory crusher according to the actual situation, and establishing a dynamic model of the crusher;

[0014] Step S7, after setting parameters according to the actual working conditions, a discrete element simulation model of the crusher is established;

[0015] Step S8, establishing a DEM-MBD joint simulation model of the gyratory crusher through the coupling interface of the dynamics simulation software and the discrete element simulation software, inputting the state parameters of the crusher when it is working normally to simulate the crushing process of the gyratory crusher, and obtaining the performance parameter data of the crusher;

[0016] Step S9, combining relevant structural parameters, relevant state parameters and simulation performance results into sample data, and completing the determination of sample points after selective sampling;

[0017] Step S10: training the prediction model using sample point data;

[0018] Step S11, calibrating the prediction model through self-learning function using historical operating condition data;

[0019] Step S12: The state data obtained by each sensor is used as twin data, input into the prediction model to predict the performance parameters, and input into the coupling simulation model to simulate the crushing process;

[0020] Step S13: output and save the performance parameter prediction results.

[0021] Furthermore, in the step S1, common faults include bearing wear, eccentric sleeve failure, filter blockage, crusher blockage, lubricant contamination, lubricant leakage, excessive vibration, and crusher overload; the key working parameters include spindle speed, crushing chamber material level, motor power and crushing chamber pressure; the collected gyratory crusher status data includes bearing vibration frequency f, bearing temperature T1, filter pressure difference P1, hydraulic cylinder pressure P2, lubricating oil temperature T2, cooling pipe oil outlet temperature T3, lubricating oil pressure P3, lubricating oil level H, crusher horizontal vibration value L1 and vertical vibration value L2, spindle speed n, material level height h, crushing chamber pressure p, motor current I, motor voltage U, material mass m at the conveyor belt and material image.

[0022] Furthermore, in step S2, the gyratory crusher parameter information includes the geometric structure parameters and material parameters of the crusher; the ore parameter information to be crushed includes the geometric structure parameters and material parameters of the ore particles. A three-dimensional geometric model is established according to the parameters and assembly relationship of the gyratory crusher using Solidworks software; a three-dimensional geometric model of the ore particles is established according to the parameters and shape characteristics of the ore using Solidworks software.

[0023] Further, in the step S3, the multi-sensor includes a crusher vibration sensor, a bearing temperature sensor, a lubricating oil level sensor, a speed sensor, a material level sensor, a crushing chamber pressure sensor, a current sensor, a voltage sensor, a mass sensor, a bearing vibration sensor, a filter pressure difference sensor, a hydraulic cylinder pressure sensor, a lubricating oil temperature sensor, a cooling pipe temperature sensor, and a lubricating oil pressure sensor;

[0024] The speed sensor is located at the main shaft, the material level height sensor is located at the crusher crossbeam, the crushing chamber pressure sensor is located at the crushing chamber reserved position, the crusher vibration sensor is located outside the crusher body, the mass sensor is located at the discharge conveyor belt, the bearing vibration sensor is located at the bearing to be tested, the bearing temperature sensor is located at the bearing to be tested, the motor current sensor is located at the motor, the motor voltage sensor is located at the motor, the filter pressure difference sensor is located at the lubricating oil filter, the hydraulic cylinder pressure sensor is located at the hydraulic cylinder, the lubricating oil temperature sensor is located at the lubricating oil tank, the lubricating oil pressure sensor is located at the lubricating oil tank, the cooling pipe temperature sensor is located at the cooling oil outlet pipe, and the lubricating oil level sensor is located at the lubricating oil tank;

[0025] The high-speed camera is located at the discharging conveyor belt;

[0026] The state data is exchanged with the host computer through a data interface; the database adopts MySQL database.

[0027] Furthermore, in step S4, the judgment contents include bearing vibration frequency f, bearing temperature T1, filter pressure difference P1, hydraulic cylinder pressure P2, lubricating oil temperature T2, cooling pipe oil outlet temperature T3, lubricating oil pressure P3, lubricating oil level H, crusher horizontal vibration value L1 and vertical vibration value L2, spindle speed n, material level height h, crushing chamber pressure p and motor power P after the upper computer converts the motor current I and motor voltage U, and the material particle size d at the discharge port obtained after the upper computer processes the image obtained by the high-speed camera. The image processing process is to perform scale calibration, binarization, Gaussian filtering, image enhancement, and particle size determination on the captured image.

[0028] If the bearing vibration frequency f exceeds the set vibration frequency value f u , a warning of excessive bearing vibration frequency is issued; if the bearing temperature T1 exceeds the set bearing temperature value T1 u , a warning of excessive bearing temperature is issued; if the filter pressure difference P1 exceeds the set filter pressure difference P1 u , a filter pressure differential over-large warning is issued; if the filter pressure differential P1 is lower than the set filter pressure differential P1 d , a warning of too low filter pressure difference is issued; if the hydraulic cylinder pressure P2 exceeds the set hydraulic cylinder pressure P2 u , a hydraulic cylinder overpressure warning is issued; if the lubricating oil temperature T2 exceeds the set lubricating oil temperature T2 u , a lubricating oil temperature over-high warning is issued; if the lubricating oil temperature T2 is lower than the set lubricating oil temperature T2 d , a low lubricating oil temperature warning is issued; if the cooling pipe oil outlet temperature T3 exceeds the set lubricating oil temperature T3 u, a warning of excessive cooling pipe oil outlet temperature is issued; if the cooling pipe oil outlet temperature T3 is lower than the set cooling pipe oil outlet temperature T3 d , a warning of low oil outlet temperature of the cooling pipe is issued; if the lubricating oil pressure P3 exceeds the set lubricating oil pressure P3 u , a warning of excessive lubricating oil pressure is issued; if the lubricating oil pressure P3 is lower than the set lubricating oil pressure P3 d , a warning of too low lubricating oil pressure is issued; the lubricating oil level H is lower than the set lubricating oil level H d , a low lubricating oil level warning is issued; if the crusher horizontal vibration value L1 is higher than the set crusher horizontal vibration value L1 u , a warning of excessive vibration of the crusher is issued; if the vertical vibration value L2 of the crusher is higher than the set horizontal vibration value L2 of the crusher u , a warning of excessive crusher vibration is issued; if the particle size d exceeds the set maximum allowable particle size d u If the crushing chamber pressure p exceeds the set crushing chamber maximum allowable pressure p u , a warning of excessive pressure in the crushing chamber is issued; if the motor power P exceeds the set value P u , a motor overpower warning is issued.

[0029] Furthermore, in step S6, the dynamic model is established in Recurdyn software. In step S7, the discrete element simulation model is established in EDEM software. In step S8, the performance parameter data of the crusher include the productivity s(t), the particle size distribution d(t), the liner wear w(t) and the power P(t) of the gyratory crusher.

[0030] Furthermore, in step S9, the relevant structural parameters include the size of the feed opening a, the size of the discharge opening b, the height of the straight section cavity h l , Curve cavity height h s , the curvature of the curve segment cavity r, the eccentricity angle γ, the bottom angle of the moving cone α; the relevant state parameters include the spindle speed n, the material level height h and the crushing chamber pressure p; the simulation performance results include the productivity s, the discharge particle size d, the liner wear w and the power P of the gyratory crusher. The sampling steps include,

[0031] S91, divide the sample into intervals every one minute, and divide the sample space into i-dimensional intervals;

[0032] S92, sampling the sample data in each dimensional sample space;

[0033] S93. Save and transmit the extracted samples in a matrix form.

[0034] Furthermore, in step S10, the prediction model is a Kriging model, and the expression is as follows:

[0035] y(x)=g(x) T β+z(x)

[0036] Where g(x) is the polynomial function of the variable x, that is, the basis function of the model; β is the regression coefficient; z(x) is a Gaussian random process with a mean of 0.

[0037] Furthermore, in step S10, the self-learning function is a U function, and the expression of the function U(X) is as follows:

[0038]

[0039] In the formula, is an estimate of Y(X); is the corresponding standard deviation.

[0040] Furthermore, in step S12, when the prediction model predicts the performance parameter, the prediction step includes:

[0041] S121, generating initial sample points using the Latin hypercube method;

[0042] S122, performing simulation calculation on the crushing process according to the extracted sample point data to obtain corresponding performance parameter values, and saving and transmitting them in a matrix form;

[0043] S123, using 70% of the sample points and the corresponding response values ​​to train the Kriging prediction model;

[0044] S124, using the Kriging model to perform performance parameter prediction calculations on the remaining 30% of the sample points to obtain predicted response values;

[0045] S125, using the self-learning function U function to calculate the predicted response value to obtain the U value;

[0046] S126. If the U value is less than the set value, the establishment of the Kriging prediction model is completed;

[0047] S127, if the U value is greater than the set value, the Kriging model is calibrated and updated using historical operating data and then the process returns to step S125;

[0048] S128. Complete the establishment of the Kriging performance parameter prediction model.

[0049] Furthermore, in step S13, the predicted performance parameters are displayed via a display device and stored in a data storage module.

[0050] Another object of the present invention is to provide a system for gyratory crusher state monitoring and performance prediction based on digital twin technology, including the following subsystems: state monitoring system, performance prediction system, simulation system; each subsystem is composed of a physical entity module, a data transmission and storage module, and a virtual model module.

[0051] The physical entity module includes the entity and three-dimensional geometric model of each device;

[0052] The virtual model module is used to predict the performance parameters of the gyratory crusher equipment in real time and to perform synchronous simulation of the crushing process by establishing a digital twin;

[0053] The data transmission and storage module completes the real-time data interaction between the physical entity module and the virtual model module, and is used for real-time transmission and storage of the status data and predicted performance parameter data of the gyratory crusher equipment.

[0054] The beneficial effects of the present invention are:

[0055] 1. Quickly and accurately predict the performance parameters of the gyratory crusher: Combine the equipment entity, crushing process of the crusher, simulation model and digital twin technology, use the Latin hypercube sampling method to sample and select the sample data after eliminating invalid sample points, use the selected sample data to train the Kriging prediction model and use the historical working condition sample data to self-learn and calibrate the prediction model. When the prediction accuracy meets the requirements, the prediction model is established. After inputting the state parameters into the prediction model, the predicted value of the performance parameter of the gyratory crusher can be quickly obtained. The required calculation amount is small, the time cost is low, and the immediacy is good.

[0056] 2. Real-time response to the working status of the equipment: The equipment status data can be output in real time through the installed multiple sensors and high-speed camera equipment. When the status data is transmitted to the host computer through the port for analysis and judgment, the working status of the equipment is monitored and timely feedback is given on the working status of the equipment through the display; when the status data is transmitted to the virtual model module through the data transmission module, it is used as the input data of the prediction model to obtain the prediction parameters.

[0057] 3. Comprehensively consider the productivity, motor power and discharge particle size of the gyratory crusher, collect and monitor multi-sensor data in real time, provide a data basis for in-depth research on the working mechanism of the gyratory crusher, facilitate subsequent theoretical analysis of the gyratory crusher, and contribute to the research and development of efficient, energy-saving and safe gyratory crushers.

[0058] 4. The present invention proposes a method and system for state monitoring and performance prediction of a gyratory crusher based on digital twin technology. Taking into account the working characteristics of the gyratory crusher, it integrates multiple information and multiple modules, which plays an important role in improving the intelligence and digitalization level of crushing equipment, and has a driving effect on achieving high-quality development of my country's manufacturing industry. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] The drawings described herein are used to provide further understanding of the present invention and constitute a part of this application. The illustrative examples of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0060] Figure 1 A step diagram of a method for monitoring and predicting performance of a gyratory crusher based on digital twins of the present invention;

[0061] Figure 2 It is a system block diagram of gyratory crusher monitoring and performance prediction based on digital twin of the present invention;

[0062] Figure 3 It is an overall flow chart of the method for monitoring and performance prediction of a gyratory crusher based on digital twin of the present invention;

[0063] Figure 4 It is a schematic diagram of signal acquisition of the present invention. DETAILED DESCRIPTION

[0064] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than 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. In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0065] See also Figures 1 to 4As shown, the method and system for monitoring and predicting the performance of a gyratory crusher based on digital twins of the present invention, according to the working characteristics and common fault types of the gyratory crusher, uses multi-sensor equipment and high-speed cameras to collect the key parameters of the crusher in real time and monitor the state of the crusher; combines the real parameters and working conditions to establish the dynamics and discrete element analysis model of the gyratory crusher, simulates the crushing process of the gyratory crusher, and obtains the relevant performance parameter data of the crusher; uses the relevant geometric parameters, state parameters and performance parameter data of the crusher as sample data, selects and processes the sample data, and then trains and establishes the performance prediction model of the gyratory crusher, and performs self-learning correction on the prediction model in combination with the historical working condition sample data to complete the determination of the prediction model. Real-time monitoring of the working state of the gyratory crusher is achieved by comparing and judging the data collected by the sensor; the performance parameters of the equipment are predicted in real time by taking the state data obtained by the sensor as twin data and inputting it into the prediction model; the dynamic simulation of the crushing process is performed by taking the real-time state data of the crusher as input and inputting it into the crusher dynamics and discrete element coupling simulation model. In response to the problem that the prior art lacks real-time status monitoring of gyratory crushers and real-time prediction of crushing performance levels, the present invention combines digital twin technology with gyratory crusher equipment, fully integrates multi-sensor equipment, and establishes a status monitoring system and digital twin model for the gyratory crusher, thereby completing real-time monitoring of the working status of the gyratory crusher and real-time prediction of the performance level.

[0066] See also Figure 1 As shown, the method for state monitoring and performance prediction of a gyratory crusher based on digital twin technology of the present invention has the following specific implementation steps:

[0067] Step S1, determining the collected gyratory crusher status data according to common faults and key working parameters in the gyratory crusher crushing operation;

[0068] Step S2, establishing a three-dimensional geometric model of the equipment and materials according to parameter information of the gyratory crusher and the ore particles to be crushed;

[0069] Step S3: collect the real-time status data of the equipment based on multiple sensors and high-speed cameras and transmit it to the host computer database;

[0070] Step S4: The host computer determines the real-time status data and monitors the equipment operation status;

[0071] Step S5, storing the state parameter group with warning and the state parameter group without warning respectively;

[0072] Step S6, setting the motion relationship between the components of the gyratory crusher according to the actual situation, and establishing a dynamic model of the crusher;

[0073] Step S7: After setting parameters according to the actual working conditions, a discrete element simulation model of the crusher is established.

[0074] Step S8, establishing a DEM-MBD joint simulation model of the gyratory crusher through the coupling interface of the dynamics simulation software and the discrete element simulation software, inputting the state parameters of the crusher when it is working normally to simulate the crushing process of the gyratory crusher, and obtaining the performance parameter data of the crusher;

[0075] Step S9, combining relevant structural parameters, relevant state parameters and simulation performance results into sample data, and completing the determination of sample points after selective sampling;

[0076] Step S10: training the prediction model using sample point data;

[0077] Step S11, self-learning and calibrating the prediction model through historical operating condition data;

[0078] Step S12: using the state data obtained by the sensor as twin data, inputting it into the prediction model to predict the performance parameters, and inputting it into the coupling simulation model to simulate the crushing process;

[0079] Step S13: output and save the performance parameter prediction results.

[0080] See also Figure 2 As shown, the system for state monitoring and performance prediction of a gyratory crusher based on digital twin technology described in the present invention includes the following subsystems: a state monitoring system, a performance prediction system, and a simulation system; each subsystem is respectively composed of a physical entity module, a data transmission and storage module, and a virtual model module.

[0081] The physical entity module includes physical devices such as gyratory crusher equipment, sensor equipment, high-speed camera equipment, simulation computing equipment, data transmission processing equipment, and display equipment, as well as a three-dimensional model of the equipment constructed based on real structural parameters, material parameters, and state parameters.

[0082] The virtual model module includes a virtual entity, a behavior mapping of the equipment working process in a virtual space, and performance parameter information of a digital twin.

[0083] The data transmission and storage module includes not only the physical equipment required for data transmission and storage, but also the standard protocol for transmitting interactive data and the stored data information. It is connected to the physical entity module and the virtual model module to achieve connection and interaction between the modules.

[0084] The existing prediction of gyratory crusher performance parameters requires a large amount of calculation, high time cost, and the immediacy needs to be improved. The purpose of the present invention is to build a digital twin model of the gyratory crusher crushing process, so as to make real-time predictions of the productivity performance parameters, particle size distribution performance parameters, and power performance parameters of the gyratory crusher equipment. To achieve the above purpose, a method and system for predicting gyratory crusher performance based on digital twin are proposed. The function implementation process is as follows: Figure 3 shown.

[0085] The three-dimensional geometric model building process uses Solidworks software to build a three-dimensional geometric model of the gyratory crusher according to the structural parameters and assembly relationship of the equipment; uses Solidworks software to build a three-dimensional geometric model of the gyratory crusher ore particles according to the geometric parameters and shape characteristics of the ore.

[0086] The simulation system is composed of RecurDyn software and EDEM software. The steps of the coupled simulation analysis process include: importing the three-dimensional model of the gyratory crusher into the simulation software in the required format of the software; setting the parameters required for the model according to the actual working state of the equipment. The parameter setting includes: the constraint relationship of each component; the motion parameter setting, the moving parts of the gyratory crusher include the main shaft and the moving cone, and the motion parameters include the suspension point position, the rotation speed, and the start time of the movement; the contact model, the contact model includes the normal force contact model and the tangential force contact model; the material parameters of the components and the ore, including the material density, Young's modulus, and Poisson's ratio; the material contact parameters, the contact objects include between components, between components and ore, and between ore and ore; the parameters include the static friction coefficient, the dynamic friction coefficient, and the collision recovery coefficient; the falling material parameter setting, the falling material parameter includes the ore falling position and the ore falling amount per unit time; the simulation parameter setting, the simulation parameter includes the simulation time, the simulation step size, the crushing behavior start time, and the number of CPUs used for calculation.

[0087] The performance prediction system is determined by sample point selection training and self-learning calibration. In the sample point selection process, the method used is Latin hypercube sampling, and the selection sampling step includes:

[0088] S91, divide the sample into intervals every one minute, and divide the sample space into m-dimensional intervals;

[0089] S92, sampling the sample data in each dimensional sample space;

[0090] S93. Save and transmit the extracted samples in a matrix form.

[0091] In the process of establishing the prediction model, the prediction model is a Kriging model, and its expression is as follows:

[0092] y(x)=g(x) T β+z(x)

[0093] From the above expression of the Kriging model:

[0094] g(x) is the polynomial function of the variable x, i.e. the basis function of the model;

[0095] β is the regression coefficient;

[0096] z(x) is a Gaussian random process with mean 0.

[0097] When judging whether the accuracy of the prediction model meets the requirements, the self-learning function U function is used for calculation and judgment, and its expression is as follows:

[0098]

[0099] In the expression of the above self-learning function U(X):

[0100] is an estimate of Y(X);

[0101] is the corresponding standard deviation.

[0102] The state monitoring detection is completed after the host computer processes and judges the signals received by the sensor equipment and the pictures taken by the high-speed camera. The sensor equipment includes a vibration sensor, a temperature sensor, an oil level sensor, a speed sensor, an oil level sensor, a material level sensor, a pressure sensor, a mass sensor, a current sensor, and a voltage sensor; the state data is exchanged with the host computer through a data interface; the database adopts a MySQL database; the host computer compares and judges the received data with the set value to complete real-time monitoring of the working state of the crushing equipment.

[0103] The present invention adopts multiple sensors and high-speed cameras to collect and monitor relevant signals in real time, wherein the speed sensor is located at the main shaft, the material level height sensor is located at the crusher crossbeam, the crushing chamber pressure sensor is located at the crushing chamber reserved position, the crusher vibration sensor is located outside the crusher body, the mass sensor is located at the discharging conveyor belt, the bearing vibration sensor is located at the bearing to be tested, the bearing temperature sensor is located at the bearing to be tested, the motor current sensor is located at the motor, the motor voltage sensor is located at the motor, the filter pressure difference sensor is located at the lubricating oil filter, the hydraulic cylinder pressure sensor is located at the hydraulic cylinder, the lubricating oil temperature sensor is located at the lubricating oil tank, the lubricating oil pressure sensor is located at the lubricating oil tank, the cooling pipe temperature sensor is located at the cooling oil outlet pipe, and the lubricating oil level sensor is located at the lubricating oil tank;

[0104] The high-speed camera is located at the discharge conveyor belt.

[0105] If the bearing vibration frequency f exceeds the set vibration frequency value f u , a warning of excessive bearing vibration frequency is issued; if the bearing temperature T1 exceeds the set bearing temperature value T1 u , a warning of excessive bearing temperature is issued; if the filter pressure difference P1 exceeds the set filter pressure difference P1 u , a filter pressure differential over-large warning is issued; if the filter pressure differential P1 is lower than the set filter pressure differential P1 d , a warning of too low filter pressure difference is issued; if the hydraulic cylinder pressure P2 exceeds the set hydraulic cylinder pressure P2 u , a hydraulic cylinder overpressure warning is issued; if the lubricating oil temperature T2 exceeds the set lubricating oil temperature T2 u , a lubricating oil temperature over-high warning is issued; if the lubricating oil temperature T2 is lower than the set lubricating oil temperature T2 d , a low lubricating oil temperature warning is issued; if the cooling pipe oil outlet temperature T3 exceeds the set lubricating oil temperature T3 u , a warning of excessive cooling pipe oil outlet temperature is issued; if the cooling pipe oil outlet temperature T3 is lower than the set cooling pipe oil outlet temperature T3 d , a warning of low oil outlet temperature of the cooling pipe is issued; if the lubricating oil pressure P3 exceeds the set lubricating oil pressure P3 u , a warning of excessive lubricating oil pressure is issued; if the lubricating oil pressure P3 is lower than the set lubricating oil pressure P3 d , a warning of too low lubricating oil pressure is issued; the lubricating oil level H is lower than the set lubricating oil level H d , a low lubricating oil level warning is issued; if the crusher horizontal vibration value L1 is higher than the set crusher horizontal vibration value L1 u , a warning of excessive vibration of the crusher is issued; if the vertical vibration value L2 of the crusher is higher than the set horizontal vibration value L2 of the crusher u , a warning of excessive crusher vibration is issued; if the particle size d exceeds the set maximum allowable particle size d u If the crushing chamber pressure p exceeds the set crushing chamber maximum allowable pressure p u , a warning of excessive pressure in the crushing chamber is issued; if the motor power P exceeds the set value P u , a warning of motor overpower is issued. The sensor collects signals and transmits them to the host computer through the data interface after corresponding processing.

[0106] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made to the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for monitoring and performance prediction of a gyratory crusher based on digital twin, characterized in that: The following steps are involved: Step S1, determining the collected gyratory crusher status data according to common faults and key working parameters in the gyratory crusher crushing operation; Step S2, establishing a three-dimensional geometric model of the equipment and materials according to parameter information of the gyratory crusher and the ore particles to be crushed; Step S3: collect the real-time status data of the equipment based on multiple sensors and high-speed cameras and transmit it to the host computer database; Step S4: The host computer processes and determines the real-time status data and monitors the operation status of the equipment; Step S5, storing the state parameter group with warning and the state parameter group without warning respectively; Step S6, setting the motion relationship between the components of the gyratory crusher according to the actual situation, and establishing a dynamic model of the crusher; Step S7, after setting parameters according to the actual working conditions, a discrete element simulation model of the crusher is established; Step S8, establishing a DEM-MBD joint simulation model of the gyratory crusher through the coupling interface of the dynamics simulation software and the discrete element simulation software, inputting the state parameters of the crusher when it is working normally to simulate the crushing process of the gyratory crusher, and obtaining the performance parameter data of the crusher; Step S9, combining relevant structural parameters, relevant state parameters and simulation performance results into sample data, and completing the determination of sample points after selective sampling; Step S10: training the prediction model using sample point data; Step S11, calibrating the prediction model through self-learning function using historical operating condition data; Step S12: The state data obtained by each sensor is used as twin data, input into the prediction model to predict the performance parameters, and input into the coupling simulation model to simulate the crushing process; S121, generating initial sample points using the Latin hypercube method; S122, performing simulation calculation on the crushing process according to the extracted sample point data to obtain corresponding performance parameter values, and saving and transmitting them in a matrix form; S123, using 70% of the sample points and the corresponding response values ​​to train the Kriging prediction model; S124, using the Kriging model to perform performance parameter prediction calculations on the remaining 30% of the sample points to obtain predicted response values; S125, using self-learning function U The function calculates the predicted response value U value; S126, if U If the value is less than the set value, the establishment of the Kriging prediction model is completed; S127, if U If the value is greater than the set value, the Kriging model is calibrated and updated using the historical operating condition data and then returns to step S125; S128, completing the establishment of the Kriging performance parameter prediction model; Step S13: output and save the performance parameter prediction results.

2. The method for monitoring and predicting performance of a gyratory crusher based on digital twin according to claim 1, characterized in that: The common faults described in step S1 include bearing wear, eccentric sleeve failure, filter blockage, crusher blockage, lubricant contamination, lubricant leakage, excessive vibration, and crusher overload; the key working parameters include spindle speed, crushing chamber material level, motor power and crushing chamber pressure; the collected gyratory crusher status data include bearing vibration frequency f, bearing temperature T1, filter pressure difference P1, hydraulic cylinder pressure P2, lubricating oil temperature T2, cooling pipe oil outlet temperature T3, lubricating oil pressure P3, lubricating oil level H, crusher horizontal vibration value L1 and vertical vibration value L2, spindle speed n, material level height h, crushing chamber pressure p, motor current I, motor voltage U, material mass m at the discharge conveyor belt, and material image.

3. The method for monitoring and performance prediction of a gyratory crusher based on digital twin according to claim 1, characterized in that: The gyratory crusher parameter information described in step S2 includes the geometric structure parameters and material parameters of the crusher; the parameter information of the ore particles to be crushed includes the geometric structure parameters and material parameters of the ore particles; using Solidworks software to establish a three-dimensional geometric model according to the parameters and assembly relationship of the gyratory crusher; using Solidworks software to establish a three-dimensional geometric model of the ore particles according to the parameters and shape characteristics of the ore.

4. The method for monitoring and predicting performance of a gyratory crusher based on digital twin according to claim 1, characterized in that: The multi-sensor described in step S3 includes a crusher vibration sensor, a bearing temperature sensor, a lubricating oil level sensor, a speed sensor, a material level sensor, a crushing chamber pressure sensor, a current sensor, a voltage sensor, a mass sensor, a bearing vibration sensor, a filter pressure difference sensor, a hydraulic cylinder pressure sensor, a lubricating oil temperature sensor, a cooling pipe oil outlet temperature sensor, and a lubricating oil pressure sensor; The speed sensor is located at the main shaft, the material level height sensor is located at the crusher crossbeam, the crushing chamber pressure sensor is located at the crushing chamber reserved position, the crusher vibration sensor is located outside the crusher body, the mass sensor is located at the discharge conveyor belt, the bearing vibration sensor is located at the bearing to be tested, the bearing temperature sensor is located at the bearing to be tested, the motor current sensor is located at the motor, the motor voltage sensor is located at the motor, the filter pressure difference sensor is located at the lubricating oil filter, the hydraulic cylinder pressure sensor is located at the hydraulic cylinder, the lubricating oil temperature sensor is located at the lubricating oil tank, the lubricating oil pressure sensor is located at the lubricating oil tank, the cooling pipe temperature sensor is located at the cooling oil outlet pipe, and the lubricating oil level sensor is located at the lubricating oil tank; The high-speed camera is located at the discharging conveyor belt; The state data is exchanged with the host computer through a data interface; the database adopts MySQL database.

5. The method for monitoring and predicting performance of a gyratory crusher based on digital twin according to claim 1, characterized in that: The host computer described in step S4 judges the real-time status data, and the judgment content includes bearing vibration frequency f, bearing temperature T1, filter pressure difference P1, hydraulic cylinder pressure P2, lubricating oil temperature T2, cooling pipe oil outlet temperature T3, lubricating oil pressure P3, lubricating oil level H, crusher horizontal vibration value L1 and vertical vibration value L2, spindle speed n, material level height h, crushing chamber pressure p, motor power P after the host computer converts the motor current I and voltage U, and a warning is issued when the value of the material particle size d at the discharge port obtained by the host computer after the image obtained by the high-speed camera is processed exceeds or falls below the set value; the image processing process is to perform scale calibration processing, binarization processing, Gaussian filtering processing, image enhancement processing, and particle size determination on the captured image.

6. The method for monitoring and predicting performance of a gyratory crusher based on digital twin according to claim 1, characterized in that: The dynamic model described in step S6 is established in Recurdyn software; the discrete element simulation model described in step S7 is established in EDEM software; the performance parameter data of the crusher described in step S8 includes the productivity s(t), the discharge particle size d(t), the liner wear w(t) and the power P(t) of the gyratory crusher.

7. The method for monitoring and predicting performance of a gyratory crusher based on digital twin according to claim 1, characterized in that: The relevant structural parameters described in step S9 include the size of the feed port a, the size of the discharge port b, the height of the straight section cavity h l , Curve cavity height h s , curvature r of the curved segment cavity, eccentricity angle , bottom angle of moving cone ;Related state parameters include spindle speed n, material level h and crushing chamber pressure p; The simulation performance results include the gyratory crusher’s productivity s, discharge particle size d, liner wear w, and power P; The steps for selecting a sample include: S91, divide the sample into intervals every one minute, and divide the sample space into i Dimension interval; S92, sampling the sample data in each dimensional sample space; S93. Save and transmit the extracted samples in a matrix form.

8. The method for monitoring and predicting performance of a gyratory crusher based on digital twin according to claim 1, characterized in that: The prediction model described in step S10 is a Kriging model, and the expression is as follows: ； In the formula, For variables The polynomial function of , which is the basis function of the model; is the regression coefficient; is a Gaussian random process with a mean of 0; The self-learning function described in step S11 is U Function, Function The expression is as follows: ； In the formula, yes estimates; is the corresponding standard deviation.

Images