Blockage prediction method for crawler-type impact crushing station

Abstract

The invention discloses a material blockage prediction method for a crawler-type counterattack crushing station, and belongs to the technical field of industrial automation, the material blockage prediction method comprises a data acquisition module, a data processing module, an alarm module and a display monitoring module, current characteristics of each motor are extracted, and a decision tree model is established for material blockage prediction; according to the method, weight calculation is carried out on the material blockage prediction result of each motor to obtain a comprehensive material blockage prediction result, and judgment is carried out in combination with the rotating speed early warning threshold value and the low speed threshold value of the crushing motor, so that the technical problem of predicting and early warning material blockage events in the working process of the crawler-type counterattack crushing station is solved, loss caused by sudden shutdown is avoided, and the service life of the crawler-type counterattack crushing station is prolonged. The equipment operation reliability is improved, more accurate material blockage prediction is provided, the equipment can operate in the optimal state, the occurrence frequency of material blockage and faults is reduced, the operation efficiency and production benefits of the equipment are remarkably improved, the service life of the equipment is prolonged, and the maintenance frequency and cost are reduced.

Description

Technical Field

[0001] The invention belongs to the technical field of industrial automation, and in particular relates to a material blocking prediction method for a crawler-type impact crushing station. Background Art

[0002] Crawler mobile impact crushing station is a high-efficiency equipment integrating multiple processes such as receiving, crushing, screening, iron removal, and transportation. It is widely used in mining, construction waste recycling and other fields. It can achieve the crushing, screening and transportation of large pieces of materials through the coordinated work of multiple subsystems such as the receiving system, feeding system, crushing system, power system, conveying system, walking system and hydraulic system. Various motors are mainly used to drive the crushing host, feeder, belt conveyor and other components.

[0003] The existing crawler mobile impact crushing station has the following problems:

[0004] The problem of material blocking is serious: in actual use, once the crusher is blocked, the entire equipment will stop working and workers must enter the crusher to clean up the material. This is not only time-consuming and labor-intensive, but also poses a great safety hazard.

[0005] Lack of real-time monitoring and early warning mechanism: Existing equipment lacks real-time monitoring and effective early warning mechanism for the operating status of each motor. When blockage or other abnormal conditions occur, they cannot be discovered and handled in time, which easily leads to equipment damage and shutdown. Summary of the invention

[0006] The purpose of the present invention is to provide a method for predicting material blocking of a crawler impact crushing station, which solves the technical problem of predicting and warning of material blocking events during the operation of the crawler impact crushing station.

[0007] To achieve the above object, the present invention adopts the following technical solution:

[0008] A method for predicting material blockage in a crawler impact crushing station comprises the following steps:

[0009] Step 1: Construct data acquisition module, data processing module, alarm module and display monitoring module. The data acquisition module collects the current data of the crushing motor, feeder motor and belt conveyor motor in real time, and sends all the current data to the data processing module. The data processing module receives and stores the current data of each motor and generates historical data.

[0010] Step 2: The data processing module extracts the characteristics of the current data of each motor respectively, and establishes a training set for each motor based on historical data; a decision tree model is established for each motor respectively, and the decision tree model is trained using the corresponding training set;

[0011] Step 3: The data processing module uses the regression model to calculate the current value of each motor at the future time t, and inputs the result into the decision tree model corresponding to the motor to obtain the blockage prediction result of each motor;

[0012] Step 4: The data processing module performs weight calculation on the blockage prediction results of each motor to obtain a comprehensive blockage prediction result of the current value;

[0013] Step 5: The data processing module sets a speed warning threshold and a low speed threshold for the crushing motor. The data acquisition module collects the speed of the crushing motor in real time and uploads it to the data processing module. The data acquisition module determines the speed of the crushing motor: when it reaches the warning threshold and does not fall below the low speed threshold, the regression model is used to predict the probability of material blocking, and step 6 is executed; when it falls below the low speed threshold, an alarm command is sent to the alarm module to execute step 8;

[0014] Step 6: Perform weight calculation on the comprehensive blockage prediction result obtained in step 4 and the blockage probability obtained in step 5 to obtain the final blockage prediction result;

[0015] Step 7: The alarm module obtains the final blockage prediction result from the data processing module and issues corresponding alarm information according to the final blockage prediction result;

[0016] Step 8: The data processing module sends the final blockage prediction result to the display monitoring module for display.

[0017] Preferably, the data acquisition module and the alarm module are both PLC controllers, the display monitoring module is a display screen, the data processing module is a micro industrial computer, the data acquisition module and the alarm module both communicate with the data processing module via a 485 bus, and the display monitoring module communicates with the data processing module via an HDMI data cable.

[0018] Preferably, when executing step 2, the collected current data of the crushing motor is set to I C , the current data of the feeder motor is I F , the current data of the belt conveyor motor is I B The training sets constructed are crusher motor training set AI C , Crushing motor training set AI F , Crushing motor training set AI B ;

[0019] The features of the current data extracted from each motor include the current change rate, the current mean value and the current standard deviation;

[0020] The feature extraction data set of the training set of each motor is expressed as follows:

[0021]

[0022] Among them, Features 破碎 is the training set of the crushing motor, ΔI 破碎 is the current change rate of the crushing motor, is the mean current of the crushing motor, is the current standard deviation of the crushing motor; 给料 is the training set of the feeder motor, ΔI 给料 is the current change rate of the feeder motor, is the mean current of the feeder motor, is the current standard deviation of the feeder motor; 皮带 is the training set of belt conveyor motor, ΔI 皮带 is the current change rate of the belt conveyor motor, is the mean current of the belt conveyor motor, is the current standard deviation of the belt conveyor motor;

[0023] The formula for training the decision tree model of each motor is as follows:

[0024]

[0025] Among them, P′ 堵料，破碎 is the training result of the decision tree model for the crushing motor, P′ 堵料，给料 is the training result of the decision tree model of the feeder motor, P′ 堵料，皮带 is the training result of the decision tree model of the belt conveyor motor, n is the number of decision trees, i represents the i-th decision tree model, I 历史，破碎 represents the data in the training set of the broken motor, I 历史，给料 represents the data in the training set of the feeder motor, I 历史，皮带 Represents the data in the training set for conveyor belt motors.

[0026] Preferably, when executing step 3, the data processing module uses a regression model to predict the current values ​​of each motor at the future time t, and the specific formula is as follows:

[0027] I 破碎 (t) = I 破碎，0 +α 破碎 t+β 破碎 t 2 +∈ 破碎 ;

[0028] I 给料 (t) = I 给料，0 +α 给料 t+β 给料 t 2 +∈ 给料 ;

[0029] I 皮带 (t) = I 皮带，0 +α 皮带 t+β 皮带 t 2 +∈ 皮带 ,

[0030] Among them, I 破碎(t) is the current of the crushing motor at time t, I 破碎，0 is the initial current, indicating the current value of the crushing motor at the initial moment, α 破碎 and β 破碎 are regression coefficients, used to describe the linear and nonlinear changes of current over time, ∈ 破碎 is the error term, I 给料(t) is the current of the feeder motor at time t, I 给料，0 is the initial current, indicating the current value of the feeder motor at the initial moment, α 给料 and β 给料 are regression coefficients, used to describe the linear and nonlinear changes of current over time, ∈ 给料 is the error term, I 皮带(t) is the current of the belt conveyor motor at time t, I 皮带，0 is the initial current, indicating the current value of the belt conveyor motor at the initial moment, α 皮带 and β 皮带 are regression coefficients, used to describe the linear and nonlinear changes of current over time, ∈ 皮带 is the error term;

[0031] After inputting into the decision tree model, the blockage prediction results of each motor are obtained as shown in the following formula:

[0032]

[0033] P 堵料，破碎 is the prediction result of the decision tree model of the crushing motor, P 堵料，给料 is the prediction result of the decision tree model of the feeder motor, P 堵料，皮带 This is the prediction result of the decision tree model for the belt conveyor motor.

[0034] Preferably, when executing step 4, the specific formula for weight calculation is as follows:

[0035] P 堵料，综合 =w 破碎 P 堵料，破碎 +w 给料 P 堵料，给料 +w 皮带 P 堵料，皮带 ;

[0036] Among them, P is the comprehensive plugging prediction result, w 破碎 ω 给料 and w 皮带 They are the weights of the prediction results of the decision tree models corresponding to the crushing motor, feeder motor and belt conveyor motor respectively.

[0037] Preferably, when executing step 5, the specific formula for predicting the probability of material blocking by the regression model is as follows:

[0038]

[0039] Among them, γ0, γ1 and γ2 are the parameters of the Logistic regression model, P 堵料,转速 is the probability of material blocking predicted based on the rotation speed and current, R is the current rotation speed of the crusher, and I is the current current of the crushing motor.

[0040] Preferably, when executing step 6, the specific formula for weighting the comprehensive blockage prediction result obtained in step 4 and the blockage probability obtained in step 5 is as follows:

[0041] P 最终堵料 =w 电流 P 堵料,综合 +w 转速 P 堵料,转速 ;

[0042] Among them, P 最终堵料 is the final blockage prediction result, w 电流 is the weight of the comprehensive blockage prediction result, w 转速 is the weight of the probability of blockage.

[0043] The method for predicting material blockages in a crawler impact crushing station described in the present invention solves the technical problem of predicting and warning of material blockage events during the operation of the crawler impact crushing station. The present invention extracts the current characteristics of each motor and establishes a decision tree model to predict material blockages, so that the equipment can warn of potential failures in advance and avoid losses caused by sudden shutdowns. The decision tree model is combined with the prediction results of the regression model to accurately judge the risk of material blockages in the future and improve the reliability of equipment operation. The material blockage prediction results of each motor are weighted to obtain a comprehensive material blockage prediction result, and combined with the speed warning threshold and low speed threshold judgment of the crushing motor, a more accurate material blockage prediction is provided, so that the equipment can operate in the best state, reduce the frequency of material blockages and failures, and significantly improve the operating efficiency and production benefits of the equipment. Real-time monitoring and intelligent prediction reduce the number of unexpected shutdowns of the equipment, extend the service life of the equipment, and reduce the maintenance frequency and cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 It is the main flow chart of the present invention;

[0045] Figure 2 It is a flow chart of step 5 of the present invention. DETAILED DESCRIPTION

[0046] Depend on Figure 1-Figure 2 A method for predicting material blockage of a crawler impact crushing station shown in the figure comprises the following steps:

[0047] Step 1: Construct a data acquisition module, a data processing module, an alarm module and a display monitoring module. The data acquisition module and the alarm module are both PLC controllers, the display monitoring module is a display screen, and the data processing module is a micro industrial computer. The data acquisition module and the alarm module both communicate with the data processing module through a 485 bus, and the display monitoring module communicates with the data processing module through an HDMI data cable.

[0048] In this embodiment, the data acquisition module and the alarm module are both implemented using Siemens S7-1200 series PLC controllers, specifically Siemens S7-1200 CPU 1212C (6ES7212-1BE40-0XB0).

[0049] The micro industrial computer may be UNO-2271G micro industrial computer.

[0050] The data acquisition module collects the current data of the crushing motor, feeder motor and belt conveyor motor in real time, and sends all the current data to the data processing module. The data processing module receives and stores the current data of each motor and generates historical data;

[0051] Step 2: The data processing module extracts the characteristics of the current data of each motor respectively, and establishes a training set for each motor based on historical data; a decision tree model is established for each motor respectively, and the decision tree model is trained using the corresponding training set;

[0052] When executing step 2, set the collected current data of the crushing motor to I C , the current data of the feeder motor is I F , the current data of the belt conveyor motor is I B The training sets constructed are crusher motor training set AI C , Crushing motor training set AI F , Crushing motor training set AI B ;

[0053] The features of the current data extracted from each motor include the current change rate, the current mean value and the current standard deviation;

[0054] The feature extraction data set of the training set of each motor is expressed as follows:

[0055]

[0056] Among them, Features 破碎 is the training set of the crushing motor, ΔI 破碎 is the current change rate of the crushing motor, is the mean current of the crushing motor, is the current standard deviation of the crushing motor; 给料 is the training set of the feeder motor, ΔI 给料 is the current change rate of the feeder motor, is the mean current of the feeder motor, is the current standard deviation of the feeder motor; 皮带 is the training set of belt conveyor motor, ΔI 皮带 is the current change rate of the belt conveyor motor, is the mean current of the belt conveyor motor, is the current standard deviation of the belt conveyor motor;

[0057] The formula for training the decision tree model of each motor is as follows:

[0058]

[0059] Among them, P′ 堵料,破碎 is the training result of the decision tree model for the crushing motor, P′ 堵料,给料 is the training result of the decision tree model of the feeder motor, P′ 堵料,皮带 is the training result of the decision tree model of the belt conveyor motor, n is the number of decision trees, i represents the i-th decision tree model, I 历史，破碎 represents the data in the training set of the broken motor, I 历史，给料 represents the data in the training set of the feeder motor, I 历史，皮带 Represents the data in the training set for conveyor belt motors.

[0060] In this embodiment, the data processing module extracts the features of each motor current data, including current change rate, mean value, standard deviation, etc. The extracted feature set is expressed as:

[0061] Feature set motor = {ΔI 电机 , I 电机 ,σI 电机};

[0062] Build training sets for each motor based on historical data.

[0063] A decision tree model is established for each motor and trained using the corresponding training set. These models are used to determine whether blockage will occur.

[0064] Step 3: The data processing module uses the regression model to calculate the current value of each motor at the future time t, and inputs the result into the decision tree model corresponding to the motor to obtain the blockage prediction result of each motor;

[0065] The data processing module uses a regression model to predict the current values ​​of each motor at the future time t. The specific formula is as follows:

[0066] I 破碎 (t) = I 破碎，0 +α 破碎 t+β 破碎 t 2 +∈ 破碎 ;

[0067] I 给料 (t) = I 给料，0 +α 给料 t+β 给料 t 2 +∈ 给料 ;

[0068] I 皮带 (t) = I 皮带，0 +α 皮带 t+β 皮带 t 2 +∈ 皮带 ;

[0069] Among them, I 破碎(t) is the current of the crushing motor at time t, I 破碎，0 is the initial current, indicating the current value of the crushing motor at the initial moment, α 破碎 and β 破碎 are regression coefficients, used to describe the linear and nonlinear changes of current over time, ∈ 破碎 is the error term, I 给料(t) is the current of the feeder motor at time t, I 给料，0 is the initial current, indicating the current value of the feeder motor at the initial moment, α 给料 and β 给料 are regression coefficients, used to describe the linear and nonlinear changes of current over time, ∈ 给料 is the error term, I 皮带(t) is the current of the belt conveyor motor at time t, I 皮带，0 is the initial current, indicating the current value of the belt conveyor motor at the initial moment, α 皮带 and β 皮带 are regression coefficients, used to describe the linear and nonlinear changes of current over time, ∈ 皮带 is the error term;

[0070] After inputting into the decision tree model, the blockage prediction results of each motor are obtained as shown in the following formula:

[0071]

[0072] P 堵料，破碎 is the prediction result of the decision tree model of the crushing motor, P 堵料，给料 is the prediction result of the decision tree model of the feeder motor, P 堵料，皮带 This is the prediction result of the decision tree model for the belt conveyor motor.

[0073] In this embodiment, a decision tree model is first established for each motor to judge the blockage situation of each motor, and then the decision tree model is trained to obtain reliable judgment results. This predicted model will play a role in accurately judging and identifying blockage events in subsequent prediction calculations.

[0074] Step 4: The data processing module performs weight calculation on the blockage prediction results of each motor to obtain a comprehensive blockage prediction result of the current value;

[0075] The specific formula for weight calculation is as follows:

[0076] P 堵料，综合 =w 破碎 P 堵料，破碎 +w 给料堵料，给料 +w 皮带 P 堵料，皮带 ;

[0077] Among them, P blocking material, comprehensive is the comprehensive blocking material prediction result, w 破碎 w 给料 and w 皮带 They are the weights of the prediction results of the decision tree models corresponding to the crushing motor, feeder motor and belt conveyor motor respectively.

[0078] In this embodiment, models are established for the three motors to determine the blockage conditions of the three motors respectively, and then the importance of each motor in predicting the blockage event is allocated according to the weight, and finally the blockage prediction result is obtained, which greatly increases the accuracy of the final result.

[0079] Step 5: The data processing module sets a speed warning threshold and a low speed threshold for the crushing motor. The data acquisition module collects the speed of the crushing motor in real time and uploads it to the data processing module. The data acquisition module determines the speed of the crushing motor: when it reaches the warning threshold and does not fall below the low speed threshold, the regression model is used to predict the probability of material blocking, and step 6 is executed; when it falls below the low speed threshold, an alarm command is sent to the alarm module to execute step 8;

[0080] The specific formula for predicting the probability of material blockage using the regression model is as follows:

[0081]

[0082] Among them, γ0, γ1 and γ2 are the parameters of the Logistic regression model, P 堵料,转速 is the probability of material blocking predicted based on the rotation speed and current, R is the current rotation speed of the crusher, and I is the current current of the crushing motor.

[0083] γ0 is the intercept term, which represents the basic probability when there is no influence of other characteristics;

[0084] γ1 is the regression coefficient of the rotation speed, which indicates the influence weight of the rotation speed on the probability of material blockage;

[0085] γ2 is the regression coefficient of current, which represents the influence weight of current on the probability of blockage.

[0086] γ0, γ1 and γ2 can all be obtained through training with historical data.

[0087] In this embodiment, a speed warning mechanism is established. By setting a speed warning threshold, once the speed of the crushing motor reaches the threshold, the system immediately starts the warning mechanism and prepares to predict the probability of blockage. In this way, the speed is associated with the prediction of blockage. According to the current speed data of the crushing motor, the system can predict the probability of blockage in combination with the regression model, thereby improving the accuracy and timeliness of the prediction. At the same time, this embodiment also sets a low-speed threshold. When the speed of the crushing motor drops below the low-speed threshold, the system immediately sends an alarm instruction to the alarm module, indicating that there may be abnormal equipment operation or blockage. Combining the two situations, the system can respond to changes in the motor state in real time, ensuring that a warning can be issued in time before blockage occurs, effectively avoiding the risk of equipment damage or production interruption, reducing losses and maintenance costs caused by equipment blockage, improving the reliability and stability of the equipment, and providing strong support for subsequent data analysis and equipment maintenance, which helps to optimize equipment operation strategies and production plans.

[0088] Step 6: Perform weight calculation on the comprehensive blockage prediction result obtained in step 4 and the blockage probability obtained in step 5 to obtain the final blockage prediction result. The specific formula is as follows:

[0089] P 最终堵料 =w 电流 P 堵料,综合 +w 转速 P 堵料,转速 ;

[0090] Among them, P 最终堵料 is the final blockage prediction result, w 电流 is the weight of the comprehensive blockage prediction result, w 转速 is the weight of the probability of blockage.

[0091] Step 7: The alarm module obtains the final material blocking prediction result from the data processing module, and issues corresponding alarm information according to the final material blocking prediction result; in this embodiment, the alarm information is the final material blocking prediction result, and the user can customize the control rules according to the result.

[0092] Step 8: The data processing module sends the final blockage prediction result to the display monitoring module for display.

[0093] The method for predicting material blockages in a crawler impact crushing station described in the present invention solves the technical problem of predicting and warning of material blockage events during the operation of the crawler impact crushing station. The present invention extracts the current characteristics of each motor and establishes a decision tree model to predict material blockages, so that the equipment can warn of potential failures in advance and avoid losses caused by sudden shutdowns. The decision tree model is combined with the prediction results of the regression model to accurately judge the risk of material blockages in the future and improve the reliability of equipment operation. The material blockage prediction results of each motor are weighted to obtain a comprehensive material blockage prediction result, and combined with the speed warning threshold and low speed threshold judgment of the crushing motor, a more accurate material blockage prediction is provided, so that the equipment can operate in the best state, reduce the frequency of material blockages and failures, and significantly improve the operating efficiency and production benefits of the equipment. Real-time monitoring and intelligent prediction reduce the number of unexpected shutdowns of the equipment, extend the service life of the equipment, and reduce the maintenance frequency and cost.

Claims

1. A method for predicting material blockage in a crawler impact crushing station, characterized in that: The steps include: Step 1: Construct data acquisition module, data processing module, alarm module and display monitoring module. The data acquisition module collects the current data of the crushing motor, feeder motor and belt conveyor motor in real time, and sends all the current data to the data processing module. The data processing module receives and stores the current data of each motor and generates historical data. Step 2: The data processing module extracts the characteristics of the current data of each motor respectively, and establishes a training set for each motor based on historical data; a decision tree model is established for each motor respectively, and the decision tree model is trained using the corresponding training set; Step 3: The data processing module uses the regression model to calculate the current value of each motor at the future time t, and inputs the result into the decision tree model corresponding to the motor to obtain the blockage prediction result of each motor; Step 4: The data processing module performs weight calculation on the blockage prediction results of each motor to obtain a comprehensive blockage prediction result of the current value; Step 5: The data processing module sets a speed warning threshold and a low speed threshold for the crushing motor. The data acquisition module collects the speed of the crushing motor in real time and uploads it to the data processing module. The data acquisition module determines the speed of the crushing motor: when it reaches the warning threshold and does not fall below the low speed threshold, the regression model is used to predict the probability of material blocking and step 6 is executed; when it falls below the low speed threshold, an alarm command is sent to the alarm module to execute step 8; Step 6: Perform weight calculation on the comprehensive blockage prediction result obtained in step 4 and the blockage probability obtained in step 5 to obtain the final blockage prediction result; Step 7: The alarm module obtains the final blockage prediction result from the data processing module and issues corresponding alarm information according to the final blockage prediction result; Step 8: The data processing module sends the final blockage prediction result to the display monitoring module for display.

2. The method for predicting material blockage in a crawler impact crushing station according to claim 1, characterized in that: The data acquisition module and the alarm module are both PLC controllers, the display monitoring module is a display screen, and the data processing module is a micro industrial computer. The data acquisition module and the alarm module both communicate with the data processing module via a 485 bus, and the display monitoring module communicates with the data processing module via an HDMI data line.

3. The method for predicting material blockage of a crawler impact crushing station according to claim 1, characterized in that: When executing step 2, set the collected current data of the crushing motor to I C , the current data of the feeder motor is I F , the current data of the belt conveyor motor is I B The training sets constructed are crusher motor training set AI C , Crushing motor training set AI F , Crushing motor training set AI B ; The features of the current data extracted from each motor include the current change rate, the current mean value and the current standard deviation; The feature extraction data set of the training set of each motor is expressed as follows: Among them, Features 破碎 is the training set of the crushing motor, ΔI 破碎 is the current change rate of the crushing motor, is the mean current of the crushing motor, is the current standard deviation of the crushing motor; 给料 is the training set of the feeder motor, ΔI 给料 is the current change rate of the feeder motor, is the mean current of the feeder motor, is the current standard deviation of the feeder motor; 皮带 is the training set of belt conveyor motor, ΔI 皮带 is the current change rate of the belt conveyor motor, is the mean current of the belt conveyor motor, is the current standard deviation of the belt conveyor motor; The formula for training the decision tree model of each motor is as follows: Among them, P′ 堵料,破碎 is the training result of the decision tree model for the crushing motor, P′ 堵料,给料 is the training result of the decision tree model of the feeder motor, P′ 堵料,皮带 is the training result of the decision tree model of the belt conveyor motor, n is the number of decision trees, i represents the i-th decision tree model, I 历史，破碎 represents the data in the training set of the broken motor, I 历史,给料 represents the data in the training set of the feeder motor, I 历史,皮带 Represents the data in the training set for conveyor belt motors.

4. A method for predicting material blockage in a crawler impact crushing station according to claim 3, characterized in that: When executing step 3, the data processing module uses the regression model to predict the current values ​​of each motor at the future time t. The specific formula is as follows: I 破碎 (t)=I 破碎，0 +a 破碎 t+b 破碎t 2 +∈ 破碎 ; I 给料( t)=I 给料，0 +α 给料 t+β 给料 t 2 +∈ 给料 ; I 皮带 (t)=I 皮带，0 +α 皮带 t+β 皮带 t 2 +∈ 皮带 ; Among them, I 破碎(t) is the current of the crushing motor at time t, I 破碎,0 is the initial current, indicating the current value of the crushing motor at the initial moment, α 破碎 and β 破碎 are regression coefficients, used to describe the linear and nonlinear changes of current over time, ∈ 破碎 is the error term, I 给料(t) is the current of the feeder motor at time t, I 给料,0 is the initial current, indicating the current value of the feeder motor at the initial moment, α 给料 and β 给料 are regression coefficients, used to describe the linear and nonlinear changes of current over time, ∈ 给料 is the error term, I 皮带(t) is the current of the belt conveyor motor at time t, I 皮带,0 is the initial current, indicating the current value of the belt conveyor motor at the initial moment, α 皮带 and β 皮带 are regression coefficients, used to describe the linear and nonlinear changes of current over time, ∈ 皮带 is the error term; After inputting into the decision tree model, the blockage prediction results of each motor are obtained as shown in the following formula: P 堵料,破碎 is the prediction result of the decision tree model of the crushing motor, P 堵料,给料 is the prediction result of the decision tree model of the feeder motor, P 堵料,皮带 This is the prediction result of the decision tree model for the belt conveyor motor.

5. The method for predicting material blockage of a crawler impact crushing station according to claim 4, characterized in that: When executing step 4, the specific formula for weight calculation is as follows: P 堵料，综合 =w 破碎 P 堵料，破碎 +w 给料 P 堵料，给料 +w 皮带 P 堵料，皮带 ; Among them, P blocking material, comprehensive is the comprehensive blocking material prediction result, w 破碎 、w 给料 and w 皮带 They are the weights of the prediction results of the decision tree models corresponding to the crushing motor, feeder motor and belt conveyor motor respectively.

6. The method for predicting material blockage in a crawler impact crushing station according to claim 5, characterized in that: When executing step 5, the specific formula for predicting the probability of material blocking by the regression model is as follows: Among them, γ0, γ1 and γ2 are the parameters of the Logistic regression model, P 堵料,转速 is the probability of blockage predicted based on the rotation speed and current, R is the current rotation speed of the crusher, and I is the current current of the crushing motor.

7. A method for predicting material blockage in a crawler impact crushing station according to claim 6, characterized in that: When executing step 6, the specific formula for weighting the comprehensive blockage prediction result obtained in step 4 and the blockage probability obtained in step 5 is as follows: P 最终堵料 =w 电流 P 堵料，综合 +w 转速 P 堵料，转速 ; Among them, P 最终堵料 is the final blockage prediction result, w 电流 is the weight of the comprehensive blockage prediction result, w 转速 is the weight of the probability of blockage.

Images