A method for monitoring and analyzing the running track of a vibrating screen

Abstract

The application discloses a kind of vibration screen operation trajectory monitoring and fault analysis method, during the operation of vibration screen, the vibration condition of different directions of vibration screen screen frame is monitored in real time by acceleration sensor, monitoring information is transported to data acquisition module, vibration data is output to industrial computer by data acquisition module, the running track of screen frame is calculated according to vibration data, whether vibration screen is in normal operating state is judged by analyzing the running track of the two sides of inlet end and the two sides of discharge end, and fault analysis is carried out by comparison with historical running track, to judge the fault characteristics of vibration screen unbalance load, overload, screen hole blockage, spring failure, etc., timely eliminate fault, prevent vibration screen from being further damaged.The application can remotely monitor the running posture condition of vibration screen in real time and carry out fault analysis.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of vibrating screen, and relates to a vibrating screen operation trajectory monitoring and fault analysis method. BACKGROUND

[0002] The vibrating screen is an indispensable material grading equipment in the metallurgical, coal, mining and other industries. The vibrating screen can be divided into single-shaft vibrating screen, double-shaft vibrating screen and box-type vibrator vibrating screen according to the type of vibrator. The single-shaft vibrating screen is vibrated by a single unbalanced weight, the screen box is vibrated, the screen surface is inclined, and the movement trajectory of the screen box is generally circular or elliptical. The double-shaft or box-type vibrating screen is vibrated by synchronous and opposite rotary double unbalanced weights, the screen surface is horizontal or slightly inclined, and the movement trajectory of the screen box is circular or straight. The monitoring method is suitable for monitoring and analyzing the movement trajectory of all the above vibrating screens.

[0003] The vibrating screen is vibrated by the motor during work, the exciting force generated by the eccentric block in the vibrator drives the screen frame to run at a certain movement trajectory, and the material on the screen frame is screened and graded through the screen mesh according to the designed movement trajectory. However, in the use process of the vibrating screen, some incorrect use methods or faults caused by the use time of the equipment may cause the vibrating screen to fail to work according to the designed vibration movement trajectory during operation. If the vibration abnormality cannot be found in time, the screening efficiency of the vibrating screen will be continuously reduced, and even a series of problems such as screen box cracking, beam breaking and vibrator damage may occur, which will delay the engineering progress and cause huge economic losses.

[0004] The ideal movement trajectory of the existing vibrating screen is determined during the design stage of the drawing, and the different movement trajectories determine the use of the vibrating screen in different use occasions. However, the movement trajectory of the vibrating screen in the actual field working state cannot always be consistent with the designed movement trajectory. The reason is that the actual material running working condition is very complex, and the change of the movement trajectory of the vibrating screen may be caused by the occurrence of unbalanced loading or excessively thick material layer. However, it is difficult for the on-site personnel or camera to find these problems in time. Only when the vibrating screen fails or even cracks caused by long-term unbalanced loading or overload can the on-site personnel find the fault. However, this method of monitoring the running fault of the equipment is lagging and may cause economic losses. SUMMARY

[0005] The purpose of this invention is to solve the problem of existing vibrating screens' inability to effectively monitor the screen frame's trajectory during operation. This prevents the early detection and resolution of abnormal trajectory issues, leading to prolonged periods of unreasonable operation and reduced screening efficiency, ultimately damaging the equipment. According to vibrating screen manufacturers' experience, the vast majority of vibrating screen failures are due to improper on-site use. Uniform and reasonable feeding, along with appropriate feeding height, are fundamental guarantees for a longer equipment lifespan. However, due to the complex and variable conditions at mining production sites, it is impossible to guarantee that vibrating screens will always operate under ideal conditions. Over time, problems such as uneven feeding, screen hole blockage, and spring aging and failure often arise, and the inability to detect these problems in a timely manner often leads to further damage to the vibrating screen. To address the above-described problems, this invention provides a method for monitoring and analyzing the operating trajectory of a vibrating screen.

[0006] The technical solution adopted by this invention to solve its technical problem is: a method for monitoring the running trajectory and analyzing the faults of a vibrating screen. During the operation of the vibrating screen, the vibration of the screen frame in different directions is monitored in real time by an acceleration sensor. The monitoring information is transmitted to a data acquisition module, which outputs the vibration data to an industrial control computer. The running trajectory of the screen frame is calculated based on the vibration data. By analyzing the running trajectories on both sides of the feed end and both sides of the discharge end, it is determined whether the vibrating screen is in a normal operating state. Fault analysis is performed by comparing the running trajectory with historical running trajectories to determine fault characteristics such as uneven load, overload, screen hole blockage, and spring failure, so as to eliminate faults in time and prevent further damage to the vibrating screen.

[0007] Further optimized, the arrangement of the acceleration sensors is as follows: with the direction of material flow facing the vibrating screen as the reference, acceleration sensors are arranged at four monitoring points on the left side of the feed end, the right side of the feed end, the left side of the discharge end, and the right side of the discharge end of the vibrating screen frame. Three acceleration sensors are arranged at each monitoring point. The three acceleration sensors are aligned with three directions: perpendicular to the screen surface, parallel to the screen surface, and perpendicular to the side plate. The three directions are perpendicular to each other and intersect at a point in the same spatial coordinate system.

[0008] Further preferably, the acceleration sensor is a piezoelectric vibration acceleration sensor.

[0009] Further optimization involves calculating the screen frame's running trajectory as follows: the vibration data perpendicular to the screen surface is plotted as the Y-direction of the screen frame's running trajectory, and the vibration data parallel to the screen surface is plotted as the X-direction of the screen frame's running trajectory. The running trajectory diagrams of the four monitoring points of the vibrating screen frame are then plotted using the X-direction and Y-direction data.

[0010] Further preferably, the method of fault analysis is that: under the condition of empty load or uniform load of the vibrating screen, the running tracks on both sides of the feeding end are basically consistent, the running tracks on both sides of the discharging end are basically consistent, and the arrangement heights of the monitoring points on the left and right sides of the same end are consistent; when monitoring the running track of the screen frame in real time, the long axis, the short axis and the shape of the running tracks on the left and right sides of the same end are compared to obtain the load condition of the vibrating screen; under the condition of partial load, the running tracks on the left and right sides of the same end change, and the change condition of the running tracks is recorded; similarly, when the vibrating screen fails, the change condition of the running tracks is also recorded, and finally a database of the change of the running tracks corresponding to the fault characteristics is formed, and when the running track changes in real time, the corresponding historical running track change data is searched through the vibration data analysis system to judge the fault problem corresponding to the running track change.

[0011] Further preferably, during the operation of the vibrating screen in the production site, the data acquisition module arranged on site acquires the vibration data of the four monitoring points on both sides of the feeding end and both sides of the discharging end of the vibrating screen frame, and transmits the acquired vibration data to the cloud platform through the gateway in a wireless network. The cloud platform deploys a vibration data analysis system, which processes and analyzes the acquired data, uses the vibration data in the vertical screen surface direction and the parallel screen surface direction to draw the running track of the vibrating screen frame, compares the running tracks on both sides of the feeding end and both sides of the discharging end, and judges the difference between the running tracks on both sides by the horizontal displacement value, the vertical displacement value and the angle value of the running track, and compares the running track of the vibrating screen under the normal state to infer the running state of the vibrating screen.

[0012] Further preferably, the logic thinking of the running track analysis system in the running track analysis and judgment is that: the vibration acceleration sensor measuring equipment installed in the sensors on both sides of the feeding end and both sides of the discharging end measures the screen frame running track data under the normal operation state, including the horizontal displacement value (short axis) X0, the vertical displacement value (long axis) Y0 and the angle value a0 of the screen frame running track under the normal operation state; the horizontal displacement value (short axis) X, the vertical displacement value (long axis) Y and the angle value a of the screen frame running track under the real-time operation state are obtained. According to the fluctuation range of the normal running track data of the on-site vibrating screen, the maximum difference value of the horizontal displacement △X max , the maximum difference value of the vertical displacement △Y max , and the maximum difference value of the angle △a max .

[0013] The calculation method of the real-time difference value is:

[0014] The horizontal displacement difference value △X is |X-X0|;

[0015] The vertical displacement difference value △Y is |Y-Y0|;

[0016] The angle value difference value Δa = |a-a0|.

[0017] Compare the real-time difference value of the vibrating screen with the maximum reasonable difference value. Is there a real-time difference value ΔX > ΔX max , ΔY > ΔY max , Δa > Δa max ? If there is a real-time difference value greater than the maximum difference value, it is judged that the vibrating screen is in an abnormal running state, and abnormal comparison parameters including horizontal displacement value, vertical displacement value and angle value and screen frame running track pattern are output; the causes of the screen frame running track abnormality are analyzed and reasonable solutions are proposed, and fault data is archived and fault report is output.

[0018] Further preferably, a running track analysis system is used for fault analysis, and a fault analysis model based on a least squares support vector machine is constructed in the running track analysis system, and the processing steps are as follows:

[0019] Step S1: Taking the running track pattern of the four monitoring points of the vibrating screen screen frame as the characteristic attribute, a vibrating screen screen frame running track data set D is constructed, and the vibrating screen screen frame running track data set is labeled, with normal being 0 and abnormal being 1;

[0020] The vibrating screen screen frame running track data D i of the i-th time period = {d i1 , d i2 , d i3 , d i4}; d i1 , d i2 , d i3 , d i4 are the running tracks of the left side of the inlet end, the right side of the inlet end, the left side of the outlet end and the right side of the outlet end of the vibrating screen screen frame in the i-th time period respectively; the vibrating screen screen frame running track data set of n time periods is represented as follows:

[0021]

[0022] Step S2: The CNN network is used to extract features from the vibrating screen screen frame running track data set, and a vibrating screen screen frame running track feature data set is obtained; the vibrating screen screen frame running track feature data set is divided into a training set and a test set;

[0023] Step S3: The dung beetle optimization algorithm is used to optimize the penalty factor c and the radial basis inner product function parameter g of the least squares support vector machine (LSSVM), and the optimal parameters are obtained;

[0024] Step S4: The training set is used to train the least squares support vector machine (LSSVM) with the optimal parameters, and the test set is used for testing, and the fault analysis model based on the least squares support vector machine is obtained after the test is qualified;

[0025] Step S5: feature extraction of the real-time vibrating screen frame running track is performed through the CNN network, then fault diagnosis is performed through a fault analysis model based on a least square support vector machine, and a fault diagnosis result is output.

[0026] The present application has the beneficial effect that the running posture of the vibrating screen can be remotely monitored in real time, whether the vibrating screen is in a normal running track can be determined in a timely manner, the problem running track of the vibrating screen can be found early and personnel can be reminded to perform maintenance, a series of subsequent damages caused by long-term improper use of on-site equipment are effectively reduced, early fault determination of the vibrating screen is achieved, and predictive maintenance of the vibrating screen is achieved. Unplanned downtime caused by sudden failure of the vibrating screen is effectively reduced, and the loss borne by the user due to unplanned failure downtime is reduced. BRIEF DESCRIPTION OF DRAWINGS

[0027] Figure 1 is a specific structure diagram of the vibrating screen with running track monitoring;

[0028] Figure 2 is a running track diagram of the right side of the discharge end of the vibrating screen;

[0029] Figure 3 is a running track diagram of the left side of the discharge end of the vibrating screen;

[0030] Figure 4 is a vibrating screen running track analysis flowchart of Example 1;

[0031] In the figure: 1-motor, 2-vibration exciter, 3-vibrating screen frame, 4-monitoring point. DETAILED DESCRIPTION

[0032] The present application will be further illustrated in detail below in combination with the drawings and examples.

[0033] Example 1

[0034] Reference Figures 1-4 A vibrating screen running track monitoring and fault analysis method, a vibrating screen frame 3 is provided with a motor 1 and a vibration exciter 2, acceleration sensors are arranged at four monitoring points 4 on the left side of the inlet end, the right side of the inlet end, the left side of the discharge end and the right side of the discharge end of the vibrating screen frame 1 respectively, with the direction facing the material flow direction of the vibrating screen as the reference, three acceleration sensors are arranged at each monitoring point, the three acceleration sensors are respectively aligned in three directions, which are: the direction perpendicular to the screen surface, the direction parallel to the screen surface and the direction perpendicular to the side plate, the three directions are perpendicular to each other and intersect at a same space coordinate system. The acceleration sensors of the present embodiment are piezoelectric vibration acceleration sensors, and therefore 12 piezoelectric vibration acceleration sensors are needed to be arranged at the four monitoring points.

[0035] The piezoelectric vibration acceleration sensor is connected to a data acquisition module through a cable, the data acquisition module transmits the collected vibration data to an industrial computer through a network or transmits the data remotely to the cloud through a wireless network, the vibration data is used to calculate the running track of the screen frame, the running tracks on both sides of the inlet end and both sides of the outlet end are analyzed to determine whether the vibrating screen is in a normal running state, and the historical running track is compared to analyze the fault, so as to determine the fault characteristics of the vibrating screen, such as load imbalance, overload, screen hole blockage, spring failure, etc., to timely eliminate the fault and prevent the vibrating screen from being further damaged.

[0036] The calculation method of the running track of the screen frame is as follows: the vibration data in the vertical screen surface direction is drawn as the Y direction of the running track of the screen frame, and the vibration data in the parallel screen surface direction is drawn as the X direction of the running track of the screen frame, and the running track diagram of the four monitoring points of the screen frame of the vibrating screen is drawn through the X direction and Y direction data. Figure 2 The running track diagram of the right side of the outlet end of the vibrating screen is shown, the Y direction vibration data is drawn as the vertical screen surface direction of the monitoring point on the right side of the outlet end of the vibrating screen, and the X direction vibration data is drawn as the parallel screen surface direction of the monitoring point on the right side of the outlet end of the vibrating screen. Figure 3 The running track diagram of the left side of the outlet end of the vibrating screen is shown, the Y direction vibration data is drawn as the vertical screen surface direction of the monitoring point on the left side of the outlet end of the vibrating screen, and the X direction vibration data is drawn as the parallel screen surface direction of the monitoring point on the left side of the outlet end of the vibrating screen.

[0037] The fault analysis method is as follows: under the condition of no load or uniform load of the vibrating screen, the running tracks on both sides of the inlet end remain basically the same, the running tracks on both sides of the outlet end remain basically the same, and the arrangement heights of the monitoring points on the left and right sides of the same end remain the same; when the running track of the screen frame is monitored in real time, the long axis, the short axis and the shape of the running track on the same end are compared to obtain the load condition of the vibrating screen; in the case of load imbalance, the running tracks on the left and right sides of the same end change, and the running track change condition is recorded, including the characteristic running track, the time, the corresponding actual device load imbalance picture and video record. Similarly, when the screen hole is blocked, the running track change condition includes the characteristic running track, the time, the corresponding actual device load imbalance picture and video record, and finally a database of the running track change corresponding to the fault characteristics is formed. When the running track changes in real time, the corresponding historical running track change data is searched through the vibration data analysis system, and the fault problem corresponding to the running track change can be directly determined. Through continuous accumulation and learning of field data, the reliability of the determination method will also be continuously improved, which can be used as a reliable basis for remote equipment state health diagnosis. The running track change condition of the vibrating screen is monitored in real time, and in the case that the running tracks on both sides of the same end differ greatly, an alarm is given to remind that the running state is abnormal and output a report to determine the fault problem and give reasonable suggestions.

[0038] In the operation process of the production site vibrating screen, the data acquisition module arranged on site acquires vibration data of four monitoring points on both sides of the inlet end and both sides of the discharge end of the vibrating screen screen frame, and transmits the acquired vibration data to the cloud platform through the gateway in a wireless network. The cloud platform deploys a vibration data analysis system, which processes and analyzes the collected data, uses the vibration data in the vertical screen surface direction and the parallel screen surface direction to draw the running track trend of the vibrating screen screen frame, compares the running tracks on both sides of the inlet end and the running tracks on both sides of the discharge end, and judges the difference between the running tracks on both sides by the horizontal displacement value, the vertical displacement value and the angle value of the running track, compares the running track of the vibrating screen with the running track of the vibrating screen under the normal state, so as to infer the running state of the vibrating screen. When the running track of the screen frame appears obvious abnormality, alarm information is output in time, and the comparison result of the running track of the screen frame is pushed to the user in the form of a report, which includes the motion running track fault data of the equipment, the comparison result of each value, the possible fault cause, the reliability of the cause and the corresponding solution and other contents.

[0039] The logic thinking of the running track analysis system for running track analysis and judgment is that the vibration acceleration sensor in the sensor unit 4 installed on both sides of the inlet end and both sides of the discharge end measures the screen frame running track data under the normal running state of the equipment, including the horizontal displacement value (short axis) X0, the vertical displacement value (long axis) Y0 and the angle value a0 of the screen frame running track under the normal running state. The horizontal displacement value (short axis) X, the vertical displacement value (long axis) Y and the angle value a of the screen frame running track under the real-time running state are obtained. According to the fluctuation range of the normal running track data of the site vibrating screen, the maximum difference value of the horizontal displacement ΔX max , the maximum difference value of the vertical displacement ΔY max , and the maximum difference value of the angle value Δa max .

[0040] The calculation method of the real-time difference value is:

[0041] The horizontal displacement difference value ΔX is |X-X0|;

[0042] The vertical displacement difference value ΔY is |Y-Y0|;

[0043] The angle difference value Δa is |a-a0|.

[0044] Compare the real-time difference value of the vibrating screen with the maximum reasonable difference value. Whether there is real-time difference value ΔX> ΔX max , ΔY> ΔY max , and Δa> Δa maxIf there is a real-time difference greater than the maximum difference, it is determined that the vibrating screen is in an abnormal operating state, and an abnormal comparison parameter is output, including horizontal displacement value, vertical displacement value, and angle value, and a screen frame operating trajectory graph. The causes of the abnormal screen frame operating trajectory are analyzed, and reasonable solutions are proposed. The fault data is archived and a fault report is output.

[0045] Fault type one: abnormal values of each comparison parameter of the screen frame operating trajectory: if the horizontal displacement value X or the vertical displacement value Y is too small, it indicates that the vibrating screen is overloaded or there is interference, causing abnormal vibration of the screen. If the horizontal displacement value X or the vertical displacement value Y is too large, it indicates that the spring of the vibrating screen is failed or enters resonance, causing abnormal vibration of the screen.

[0046] Fault type two: large difference between the operating trajectory graphs of the two sides of the inlet or the two sides of the outlet: if the horizontal displacement value X, the vertical displacement value Y, and the angle value of the operating trajectory graphs of the two sides of the inlet or the two sides of the outlet are greatly different, the abnormal operating trajectory can be clearly seen, indicating that there is a problem of material unbalanced loading.

[0047] The vibration data analysis system outputs an analysis report, which includes a comparison of each numerical parameter of the operating trajectory graphs of the two sides of the inlet or the two sides of the outlet, lists abnormal data, gives corresponding fault judgments, and gives the reliability of the judgments, proposes maintenance solutions and improvement suggestions. The analysis report can help customers judge the abnormality of the equipment in advance and make timely maintenance plans, avoid further damage to the vibrating screen, and prevent non-planned shutdowns that cause significant losses to users.

[0048] In this embodiment, a piezoelectric vibration acceleration sensor is selected. When the sensor vibrates, the piezoelectric ceramic inside the sensor deforms, and the built-in circuit in the sensor converts the vibration signal into an IEPE voltage signal. The advantages of this sensor are that the piezoelectric crystal can work for a long time at a temperature above 400°C, the instantaneous (10 millisecond level) impact temperature can reach above 4000K, it can withstand very high surface pressure, has high stiffness, good linearity, small hysteresis, constant sensitivity in a wide temperature range, wide frequency response range, can withstand almost infinite load cycles, and has very high insulation impedance. The disadvantages are that some piezoelectric materials need moisture-proof measures, and the direct current response of the output is poor, which requires the use of high input impedance circuits or charge amplifiers to overcome this defect. It is widely used in long-term real-time monitoring of vibration acceleration, vibration speed, vibration displacement, and other parameters of large machinery bearings.

[0049] Embodiment 2

[0050] In embodiment 1, the corresponding historical operating trajectory change data is searched for fault analysis, and the corresponding fault problem of the operating trajectory change can be directly determined. Through historical data statistics, the maximum difference in horizontal displacement is ΔX max , the maximum difference in vertical displacement is ΔY max, the maximum difference of angle values Δa max Then, the fault type is determined according to the comparison of the real-time horizontal displacement value, the vertical displacement value and the angle value. The direct determination method adopted in Embodiment 1 is compared with data. When there is no same historical running track change data, the similarity judgment should be added to determine the fault type, and the closest historical running track graph is given. In order to overcome the problem of insufficient historical data, a fault analysis model based on least squares support vector machine is constructed in the running track analysis system.

[0051] The processing steps of the fault analysis model based on least squares support vector machine are as follows:

[0052] Step S1: The running track graph of the four monitoring points of the vibrating screen screen frame is taken as the characteristic attribute to construct the vibrating screen screen frame running track data set D, and the vibrating screen screen frame running track data set is labeled, normal is 0 and abnormal is 1;

[0053] The vibrating screen screen frame running track data D of the i th time period i ={d i1 ,d i2 ,d i3 ,d i4};d i1 ,d i2 ,d i3 ,d i4 are the running tracks of the left side of the inlet end, the right side of the inlet end, the left side of the outlet end and the right side of the outlet end of the vibrating screen screen frame respectively; The vibrating screen screen frame running track data set of n time periods is represented as follows:

[0054]

[0055] Step S2: The CNN network is used to extract the features of the vibrating screen screen frame running track data set, and the vibrating screen screen frame running track feature data set is obtained; The vibrating screen screen frame running track feature data set is divided into a training set and a test set;

[0056] Step S3: The dung beetle optimization algorithm is used to optimize the penalty factor c and the radial basis inner product function parameter g of the least squares support vector machine (LSSVM), and the optimal parameters are obtained;

[0057] Step S4: The training set is used to train the least squares support vector machine (LSSVM) with the optimal parameters, and the test set is used for testing, and the fault analysis model based on the least squares support vector machine is obtained;

[0058] Step S5: The CNN network is used to extract the features of the real-time vibrating screen screen frame running track, and then the fault analysis model based on the least squares support vector machine is used for fault diagnosis, and the fault diagnosis result is output.

[0059] The idea of Scarab Optimization Algorithm (SOA) comes from the inspiration of the rolling ball, dancing, foraging, breeding and stealing behaviors of Scarab. Five different update rules are designed to help find high-quality solutions. Each Scarab colony is composed of four different agent Scarabs, namely rolling Scarab, breeding Scarab (breeding ball), small Scarab and stealing Scarab.

[0060] Scarab Optimization Algorithm has strong optimization ability, can effectively search for optimal solutions, and does not require a large amount of computing resources and complex parameter settings, and can find better solutions in a short time. The optimization process of Scarab Optimization Algorithm is shown in Figure 2 .

[0061] I. Rolling Scarab

[0062] Scarab has an interesting habit of making feces into a ball and then rolling it to an ideal position. During the rolling process, Scarab needs to keep the fecal ball rolling in a straight line by using celestial clues (such as the position of the sun or the direction of the wind). In order to simulate the rolling behavior, Scarab needs to move in a given direction in the entire search space. During the rolling process, the position of the rolling Scarab is updated, and the rolling mathematical model can be expressed as:

[0063] x i (t+1)=x i (t)+α×β×x i (t-1)+b×Δx;

[0064] Δx=|x i (t)-X w |;

[0065]

[0066] In the formula: t represents the current iteration number, x i (t) represents the position information of the i-th Scarab at the t-th iteration, x i (t+1) represents the position information of the i-th Scarab at the t+1-th iteration, R1 is a random number with a value range of [0, 1], α is a natural coefficient, and the value is -1 or 1, β∈(0, 0.2] represents the constant of the deflection coefficient, b represents a constant belonging to (0, 1), X w represents the global worst position, and Δx is used to simulate the change of light intensity.

[0067] When the dung beetle encounters an obstacle and cannot move forward, it needs to adjust its direction by dancing to obtain a new route. The tangent function is used to simulate the dancing behavior of the dung beetle to obtain a new rolling direction. Once the dung beetle successfully determines a new direction, it will continue to roll the ball forward. Therefore, the position of the dung beetle dancing behavior is defined as follows:

[0068] x i (t+1)=x i (t)+tan(R2π)|x i (t)-x i (t-1)|;

[0069] wherein: R2 is a random number with a value range of [0, 1], when R2 is 0, 0.5 or 1, the position of the dung beetle will not be updated.

[0070] When searching, the rolling dung beetle adjusts the angle randomly, which causes the algorithm to miss some better solutions and reduces the convergence accuracy. In order to enhance the thoroughness of the search and eliminate the negative impact of local optimal solution, a dynamic mapping strategy is introduced:

[0071]

[0072]

[0073] wherein: R3 is a random number in the interval (0, 1), L is a dynamic adjustment parameter, Lb and Ub represent the lower bound and upper limit of the optimization problem respectively.

[0074] II. Reproductive dung beetle (reproductive ball)

[0075] In nature, the dung ball is rolled to a safe place and hidden by the dung beetle. In order to provide a safe environment for their offspring, it is crucial for the dung beetle to choose a suitable oviposition site. Inspired by the above discussion, a boundary selection strategy is proposed to simulate the area where the female dung beetle lays eggs, which is:

[0076]

[0077] wherein: X * represents the current local optimal position, Lb * and Ub * represent the lower bound and upper bound of the oviposition area respectively, wherein the parameter A = (1-t / T max ) 3 , T max represents the maximum number of iterations, Lb and Ub represent the lower bound and upper bound of the optimization problem respectively.

[0078] Once the oviposition area is determined, the female dung beetle will choose the area to lay eggs, and each female dung beetle will lay only one egg in each iteration. The boundary of the oviposition area is dynamically changed, which is mainly determined by the value of A. Therefore, the position of the breeding ball is also dynamic during the iteration process, which is represented as:

[0079] B p (t+1) = X * + b1(B p (t) - Lb * ) + b2(B p (t) - Ub * );

[0080] where B p (t) is the position information of the pth breeding ball at the tth iteration, b1 and b2 are two independent random vectors of size 1 x D, and D represents the dimension of the optimization problem. The position of the breeding ball is strictly limited within a certain range, i.e., the oviposition area.

[0081] Three, small dung beetles

[0082] Some dung beetles have grown into adults and crawled out of the ground to find food, which we call small dung beetles. In addition, we also need to establish the best foraging area to guide the dung beetles to forage, which simulates the foraging process of these dung beetles in nature. Specifically, the boundary of the best foraging area is defined as follows:

[0083]

[0084] where X b represents the global optimal position, Lb b and Ub b represent the lower and upper limits of the best foraging area, and R4 is a random number in the interval (0, 1). The position update of the small dung beetle is as follows:

[0085] y j (t+1) = y j (t) + C1 x (y j (t) - Lb b ) + C2 x (y j (t) - Ub b );

[0086] where y j (t) represents the position information of the jth small dung beetle at the tth iteration, y j (t+1) represents the position information of the jth small dung beetle at the t+1th iteration, C1 represents a random number following a normal distribution, and C2 represents a random vector belonging to (0, 1).

[0087] Normal individuals are difficult to search for the optimal solution when the fitness is low, and the individuals with the lowest fitness are guided to search for the optimal solution globally by increasing the average position of the current individuals.

[0088] y' min (t)=y min (t)-C·R5·[y av -y min (t)];

[0089]

[0090] In the formula, y min (t) is the position of the individual with the lowest fitness in the tth iteration, y' min (t) is the updated position of the individual y min (t), y av is the average position of all individuals in the current iteration, R5 is a random number in the interval (0, 1), and C is a dynamic weighting factor.

[0091] Four, thief scarab

[0092] Some scarabs, called thieves, steal fecal balls from other scarabs. X b is the best food source, so we assume that X b is the best place to fight for food. During the iteration process, the position information of the thief scarab is updated, which can be described as follows:

[0093] z k (t+1)=X b +S×g×(|z k (t)-X * |+|z k (t)-X b |);

[0094] In the formula, z k (t) represents the position information of the kth thief scarab in the tth iteration, g is a random vector with a size of 1x D obeying normal distribution, and S represents a constant.

[0095] The above only describes the embodiments of the present application, and does not limit the patent range of the present application, and any equivalent flow transformation using the content of the specification and drawings, or direct or indirect application in other related technical fields, are also included in the patent protection range of the present application.

Claims

1. A method for monitoring the running trajectory and analyzing the faults of a vibrating screen, characterized in that, During the operation of the vibrating screen, the vibration of the screen frame in different directions is monitored in real time by an accelerometer. The monitoring information is transmitted to the data acquisition module, which outputs the vibration data to the industrial control computer. The screen frame running trajectory is calculated based on the vibration data. The running trajectory of the screen frame is analyzed on both sides of the feed end and both sides of the discharge end to determine whether the vibrating screen is in normal operating condition. Fault analysis is performed by comparing the running trajectory with the historical running trajectory to determine the fault characteristics. The accelerometers are arranged as follows: with the direction of material flow facing the vibrating screen as the reference, accelerometers are arranged at four monitoring points on the left side of the feed end, the right side of the feed end, the left side of the discharge end, and the right side of the discharge end of the vibrating screen frame. Three accelerometers are arranged at each monitoring point. The three accelerometers are aligned with three directions: perpendicular to the screen surface, parallel to the screen surface, and perpendicular to the side plate. The three directions are perpendicular to each other and intersect at a point on the same origin in the same spatial coordinate system. The calculation method for the screen frame running trajectory is as follows: the vibration data in the direction perpendicular to the screen surface is plotted as the Y direction of the screen frame running trajectory, and the vibration data in the direction parallel to the screen surface is plotted as the X direction of the screen frame running trajectory. The running trajectory diagram of the four monitoring points of the vibrating screen frame is plotted using the X and Y direction data. The fault analysis method is as follows: When the vibrating screen is unloaded or uniformly loaded, the running trajectories on both sides of the feed end are basically consistent, the running trajectories on both sides of the discharge end are basically consistent, and the height of the monitoring points on both sides of the same end is consistent; when monitoring the running trajectory of the screen frame in real time, the load condition of the vibrating screen is obtained by comparing the major axis, minor axis, and shape of the running trajectories on both sides of the same end; when the screen is unbalanced, the running trajectories on both sides of the same end change, and the changes in running trajectory are recorded; similarly, when the vibrating screen malfunctions, the changes in running trajectory are also recorded, and a database of fault characteristics corresponding to changes in running trajectory is finally formed. When the real-time running trajectory changes, the corresponding historical running trajectory change data is retrieved through the vibration data analysis system to determine the fault problem corresponding to the change in running trajectory. During the operation of the vibrating screen at the production site, vibration data from four monitoring points—both sides of the feed end and both sides of the discharge end—are collected using data acquisition modules installed on-site. This data is then transmitted wirelessly to a cloud platform via a gateway. A vibration data analysis system is deployed on the cloud platform to process and analyze the collected data. Using vibration data perpendicular and parallel to the screen surface, the system plots the running trajectory of the vibrating screen frame. By comparing the running trajectories on both sides of the feed end and the discharge end, and using the horizontal, vertical, and angle values ​​of the running trajectories, the system determines the differences between the two sides. By comparing this with the material-carrying running trajectory of the vibrating screen under normal conditions, the operating status of the vibrating screen can be inferred.

2. The method for monitoring the operating trajectory and analyzing the faults of a vibrating screen according to claim 1, characterized in that, The acceleration sensor is a piezoelectric vibration acceleration sensor.

3. The method for monitoring the running trajectory and analyzing the faults of a vibrating screen according to claim 2, characterized in that, The logical thinking behind the operation trajectory analysis system is as follows: It measures the screen frame's operational trajectory data under normal operating conditions using piezoelectric vibration acceleration sensors installed at monitoring points on both sides of the inlet and outlet ends. This includes the horizontal displacement value X0, vertical displacement value Y0, and angle value a0 of the screen frame's operational trajectory under normal operating conditions. It then acquires the horizontal displacement value X, vertical displacement value Y, and angle value a0 of the screen frame's operational trajectory under real-time operating conditions. Input the maximum horizontal displacement difference △X based on the fluctuation range of the normal operating trajectory data of the vibrating screen. max Maximum vertical displacement difference ΔY max Maximum difference in angle values ​​△a max The calculation method for real-time difference is as follows: Horizontal angular displacement difference ΔX = |X - X0|; Vertical angular displacement difference ΔY = |Y - Y0|; The angle difference Δa = |a - a0|; Compare the real-time difference of the vibrating screen with the maximum reasonable difference; does a real-time difference ΔX > ΔX exist? max , △Y>△Y max , △a>△a max If a real-time difference is greater than the maximum difference, the vibrating screen is judged to be in an abnormal operating state, and abnormal comparison parameters are output, including horizontal displacement value, vertical displacement value, angle value, and screen frame running trajectory graphic. Analyze the causes of abnormal screen frame running trajectory and reasonable solutions, archive the fault data and output fault reports.

4. The method for monitoring the operating trajectory and analyzing the faults of a vibrating screen according to claim 1, characterized in that, Fault analysis is performed using an operational trajectory analysis system. The system constructs a fault analysis model based on least squares support vector machines, and the processing steps are as follows: Step S1: Construct a vibrating screen frame running trajectory dataset D using the running trajectory diagrams of the four monitoring points of the vibrating screen frame as feature attributes, and label the vibrating screen frame running trajectory dataset, with 0 for normal and 1 for abnormal. Vibrating screen frame running trajectory data for the i-th time period ; The images show the running trajectories of the screen frame at the left, right, left, and right sides of the feed end, respectively, during the i-th time period. The dataset of the running trajectories of the vibrating screen frame over n time periods is represented as follows: ； Step S2: Extract features from the vibrating screen frame running trajectory dataset using a CNN network; divide the vibrating screen frame running trajectory feature dataset into a training set and a test set; Step S3: Apply the dung beetle optimization algorithm to the penalty factor of the least squares support vector machine. c and radial basis inner product function parameters g Optimize to obtain the optimal parameters; Step S4: Train the least squares support vector machine with optimal parameters using the training set, and test it using the test set. If the test is successful, the fault analysis model based on the least squares support vector machine is obtained. Step S5: Extract features from the real-time vibrating screen frame running trajectory using a CNN network, then perform fault diagnosis using a fault analysis model based on least squares support vector machine, and output the fault diagnosis results.

Images