Roller press machining process monitoring management system and method based on digitization

Abstract

The invention discloses a roller press machining process monitoring management system and method based on digitization, and relates to the field of machining process monitoring management. A mapping basis is provided for subsequently obtaining corresponding working duration data and electric energy loss data according to the compression roller wear degree grade data and the bearing position wear degree grade data; by predicting compression roller wear degree grade data and bearing position wear degree grade data of to-be-monitored roller press equipment, reasons causing abnormities of working time length data and electric energy loss data are directly reflected, so that problems can be quickly solved; by acquiring the future working duration data set and the future electric energy loss data set, a judgment basis is provided for subsequently judging whether to-be-monitored roller press equipment needs to be repaired or not.

Description

Technical Field

[0001] The present invention belongs to the field of machining process monitoring and management, and in particular, relates to a digital-based roller press machining process monitoring and management system and method. Background Art

[0002] Chinese patent CN115860556B discloses a method for detecting the qualified rate of discharge of a high-pressure roller mill based on multivariate correlation. The method selects features of multiple variables in the grinding operation of the high-pressure roller mill and estimates the optimal time delay vector, constructs a multivariate correlation reconstruction matrix, constructs a grey correlation coefficient matrix, and obtains the grey correlation degree. Relative to the reference variable, the time delay of multiple process variables is estimated, the optimal time delay basis vector is searched, and the fitness function for evaluation is the grey correlation degree, and the qualified rate of discharge of the high-pressure roller mill at each moment is obtained.

[0003] Due to the working characteristics of the roller press, the surface of the roller and the bearing positions are prone to wear, and the degree of wear will continue to increase with the extension of processing time; if these wear cannot be repaired in time in advance, when the degree of wear increases to a certain value, it will have a great impact on its processing efficiency and quality, thereby causing losses to the company. Summary of the invention

[0004] In view of the problems in the related art, the present invention proposes a digital-based roller press processing process monitoring management system and method to overcome the above-mentioned technical problems existing in the existing related art.

[0005] In order to solve the above technical problems, the present invention is achieved through the following technical solutions:

[0006] The present invention is a digital-based roller press process monitoring and management method, comprising the following steps:

[0007] S1. Collect roller wear level data, bearing wear level data, working time data and power loss data of multiple roller press equipment to obtain a level data matrix, a working time data matrix and a power loss data matrix;

[0008] S2, constructing a first final mapping equation and a second final mapping equation using a working time data matrix, a power consumption data matrix and a level data matrix;

[0009] S3, collecting roller wear degree grade data and bearing wear degree grade data of the roller press equipment to be monitored and constructing the first final support vector machine prediction model and the second final support vector machine prediction model; using the first final support vector machine prediction model and the second final support vector machine prediction model to predict the roller wear degree grade data and the bearing wear degree grade data of the roller press equipment to be monitored at a future moment, and obtaining a future roller wear degree grade data set and a future bearing wear degree grade data set;

[0010] S4, using the first final mapping equation and the second final mapping equation to map and classify the future roller wear level data set and the future bearing wear level data set to obtain a future working time data set and a future power loss data set;

[0011] S5. Determine whether the roller press equipment to be monitored needs to be repaired according to the future working time data set and the future power consumption data set;

[0012] Since the roller wear and bearing wear of the roller press equipment will have a certain impact on the efficiency of roller pressing and the power loss, the present scheme collects roller wear degree level data, bearing wear degree level data, working time data and power loss data of multiple roller press equipment, thereby providing data support for the subsequent establishment of mapping equations between roller wear degree level data, bearing wear degree level data and working time data, and mapping equations between roller wear degree level data, bearing wear degree level data and power loss data; by constructing the first final mapping equation and the second final mapping equation, a mapping basis is provided for the subsequent acquisition of the corresponding working time data and power loss data according to the roller wear degree level data and the bearing wear degree level data of the roller press equipment to be monitored; by constructing the first final support vector machine prediction model and the second final support vector machine prediction model, the present scheme collects roller wear degree level data, bearing wear degree level data, working time data and power loss data of multiple roller press equipment, thereby providing data support for the subsequent establishment of mapping equations between roller wear degree level data, bearing wear degree level data and working time data, and mapping equations between roller wear degree level data, bearing wear degree level data and power loss data; by constructing the first final mapping equation and the second final mapping equation, the present scheme collects roller wear degree level data, bearing wear degree level data and working time data, and mapping equations between roller wear degree level data and bearing wear degree level data and working time data, thereby providing data support for the subsequent acquisition of the corresponding working time data and power loss data according to the roller wear degree level data and the bearing wear degree level data of the roller press equipment to be monitored; by constructing the first final support vector machine prediction model and the second final support vector machine prediction model, the present scheme collects roller wear degree level data, bearing wear degree level data and power loss data of multiple roller press equipment, thereby providing data support for the subsequent acquisition of the corresponding working time data and power loss data according to the roller wear degree level data and the bearing wear degree level data of the roller press equipment to be monitored The final support vector machine prediction model provides a prediction model for the subsequent prediction of the roller wear degree level data and the bearing wear degree level data of the monitored roller press equipment at future time points; the reason why the roller wear degree level data and the bearing wear degree level data of the monitored roller press equipment are predicted instead of directly predicting the working time data and the power loss data is that it can directly reflect the reasons that cause the abnormal working time data and the power loss data, so that when the abnormality occurs, targeted repairs can be carried out to quickly solve the problem; by mapping the roller wear degree level data and the bearing wear degree level data of the monitored roller press equipment at future time points, the future working time data set and the future power loss data set are obtained, thereby providing a basis for subsequent judgment on whether the monitored roller press equipment needs to be repaired.

[0013] Preferably, the S1 comprises the following steps:

[0014] S11, setting a plurality of roller press equipments with worn rollers but no worn bearings, roller press equipments with no worn rollers and worn bearings, roller press equipments with worn bearings and worn bearings, and roller press equipments with no worn rollers and no worn bearings, to obtain a first roller press equipment set a 1 ={a 11 ,a 12 ,...,a 1i ,...,a 1a′}、Second roller press equipment set a 2 ={a 21 ,a 22 ,...,a 2i ,...,a 2a′}、Third roller press equipment set a 3 ={a 31 ,a 32 ,...,a 3i ,...,a 3a′} and the fourth roller press equipment set a 4 ={a 41 ,a 42 ,...,a 4i ,...,a 4a′}; a 1i 、a 2i 、a 3i 、a 4i They respectively represent the set i-th roller press equipment with worn rollers and no worn bearings, roller press equipment with no worn rollers and worn bearings, roller press equipment with worn bearings and worn bearings, and roller press equipment with no worn rollers and no worn bearings, and a′ represents the total number of the set roller press equipment with worn rollers and no worn bearings, roller press equipment with no worn rollers and worn bearings, roller press equipment with worn bearings and worn bearings, and roller press equipment with no worn rollers and no worn bearings;

[0015] S12, set the cement raw material to be pressed and the wear level of the press roller Bearing seat wear grade set They represent the set i-th roller wear level and bearing wear level respectively. Respectively represent the total number of set roller wear levels and bearing wear levels;

[0016] According to the wear degree of the roller Bearing seat wear grade set Get the first roller press equipment set a1 ={a 11 ,a 12 ,...,a 1i ,...,a 1a′} and the second roller press equipment set a 2 ={a 21 ,a 22 ,...,a 2i ,...,a 2a′}、Third roller press equipment set a 3 ={a 31 ,a 32 ,...,a 3i ,...,a 3a′} and the fourth roller press equipment set a 4 ={a 41 ,a 42 ,...,a 4i ,...,a 4a′ The wear degree grade data of the rollers and the bearings of each roller press equipment in} are used to obtain the grade data matrix b; as follows,

[0017]

[0018] b 1i1 、b 1i2 、b 2i1 、b 2i2 、b 3i1 、b 3i2 、b 4i1 、b 4i2 Respectively represent the roller wear degree grade data and the bearing wear degree grade data of the roller press equipment with wear on the i-th roller and no wear on the bearing seat, the roller press equipment with no wear on the roller and wear on the bearing seat, the roller press equipment with wear on the bearing seat and wear on the bearing seat, and the roller press equipment with no wear on the roller and no wear on the bearing seat;

[0019] S13, obtaining the first roller press equipment set a 1 ={a 11 ,a 12 ,...,a 1i ,...,a 1a′}、Second roller press equipment set a 2 ={a 21 ,a 22 ,...,a 2i ,...,a 2a′}、Third roller press equipment set a 3 ={a 31 ,a 32 ,...,a 3i ,...,a3a′} and the fourth roller press equipment set a 4 ={a 41 ,a 42 ,...,a 4i ,...,a 4a′} working time data and power consumption data;

[0020] By setting four types of roller press equipment, namely, roller with wear and no wear on the bearing seat, roller without wear and wear on the bearing seat, bearing seat with wear and wear on the bearing seat, and roller without wear and no wear on the bearing seat, the comprehensiveness of the roller wear degree grade data and bearing seat wear degree grade data of the roller press equipment obtained subsequently is guaranteed, thereby ensuring the universality of the mapping equation subsequently established based on these grade data.

[0021] Preferably, the S13 comprises the following steps:

[0022] S131, setting the fine particle size of the cement raw material to be pressed after pre-crushing, recorded as the first fine particle size;

[0023] S132, respectively using the first roller press equipment set according to the first fine particle size

[0024] a 1 ={a 11 ,a 12 ,...,a 1i ,...,a 1a′}、Second roller press equipment set a 2 ={a 21 ,a 22 ,...,a 2i ,...,a 2a′}、Third roller press equipment set a 3 ={a 31 ,a 32 ,...,a 3i ,...,a 3a′} and the fourth roller press equipment set

[0025] a 4 ={a 41 ,a 42 ,...,a 4i ,...,a 4a′} Performing a pre-crushing operation on the cement raw materials to be pressed of the same initial size, and stopping the pre-crushing operation when the fine particle size of the cement raw materials to be pressed reaches a first fine particle size;

[0026] S133, after the pre-crushing operation is completed, the working time data and power consumption data of each roller press equipment are obtained to obtain the working time data matrix b′1 And the power loss data matrix b′ 2 ; They are as follows,

[0027]

[0028] Among them, b′ 11i , b′ 12i , b′ 13i , b′ 14i b′ respectively represents the working time data of the i-th roller press equipment with worn rollers and no worn bearings, the roller press equipment with no worn rollers and worn bearings, the roller press equipment with worn bearings and worn bearings, and the roller press equipment with no worn rollers and no worn bearings for pre-crushing the cement raw materials to be pressed; 21i , b′ 22i , b′ 23i , b′ 24i Respectively represent the power consumption data of the i-th roller press equipment with worn rollers and no worn bearings, the roller press equipment with no worn rollers and worn bearings, the roller press equipment with worn bearings and worn bearings, and the roller press equipment with no worn rollers and no worn bearings for pre-crushing the cement raw materials to be pressed;

[0029] This solution mainly monitors and optimizes the process of pre-crushing cement raw materials by the roller press. Therefore, by setting the first fine granularity, a unified crushing standard is constructed, so that the working time data and power loss data of the roller press equipment obtained subsequently are generated under the same premise, thereby ensuring the significance of comparing the data.

[0030] Preferably, S2 comprises the following steps:

[0031] S21, respectively setting the level data matrix b and the working time data matrix b′ 1 And the power loss data matrix b′ 2 The data mapping equation between them is used to obtain the first initial mapping equation and the second initial mapping equation; they are as follows:

[0032]

[0033] In the formula, Respectively represent the working time data and the power consumption data; Respectively represent the wear degree grade data of the roller and the bearing seat; α 1 , α 2 , α 4 , α 5α represents the undetermined coefficients of the roller wear level data and the bearing wear level data in the first initial mapping equation and the second initial mapping equation respectively; 3 , α 6 Respectively represent the bias in the first initial mapping equation and the second initial mapping equation;

[0034] S22, according to the level data matrix b and the working time data matrix b′ 1 The first initial mapping equation is optimized and adjusted to obtain the first final mapping equation; according to the level data matrix b and the power loss data matrix b′ 2 Optimizing and adjusting the second initial mapping equation to obtain a second final mapping equation;

[0035] In this solution, the mapping equation adopts the form of a linear function, which ensures the simplicity of the mapping equation and makes the subsequent optimization and adjustment of the parameters of the first initial mapping equation and the second initial mapping equation less complicated.

[0036] Preferably, in S22, according to the level data matrix b and the working time data matrix b′ 1 Optimizing and adjusting the first initial mapping equation to obtain the first final mapping equation includes the following steps:

[0037] S2211, Establishing the first salp population c 1i represents the i-th salp in the first salp population, c′ 1 represents the size of the first salp population; sets the maximum number of iterations of the first salp population to The current number of iterations is are respectively recorded as the first maximum number of iterations and the first current number of iterations; the search space dimension of the first salp population is 3 dimensions;

[0038] S2212: Set the undetermined coefficients of the roller wear level data and the bearing wear level data in the first initial mapping equation and the value range of the offset amount to obtain a first value range set. They respectively represent the lower limit and upper limit of the undetermined coefficient of the roller wear degree grade data in the first initial mapping equation; respectively represent the lower limit and upper limit of the value of the undetermined coefficient of the bearing wear degree grade data in the first initial mapping equation; Respectively represent the lower limit and upper limit of the bias value in the first initial mapping equation;

[0039] The initial position of each salp in the first salp population is set by using the Tent mapping method in conjunction with the first value interval set to obtain a first initial position matrix d 1 ;as follows,

[0040]

[0041] Among them, d 1i1 d 1i2 d 1i3 Respectively represent the components of the initial position of the i-th salp in the first salp population in the undetermined coefficient dimension of the roller wear degree grade data, the components in the undetermined coefficient dimension of the bearing seat wear degree grade data, and the components in the offset dimension;

[0042] The mapping function of the Tent mapping is as follows,

[0043]

[0044] Where, d i ' +1 d i ′ represents the data generated in the i+1th iteration and the ith iteration respectively; β 1 is a random number between 0 and 1; β 1 =0.5;

[0045] S2213, each roller wear level data and bearing wear level data in the level data matrix b are substituted into the first initial mapping equation to obtain the first initial mapping data matrix as follows,

[0046]

[0047] in, Respectively represent (b 1i1 ,b 1i2 )、(b 2i1 ,b 2i2 )、(b 3i1 ,b 3i2 )、(b 4i1 ,b 4i2 ) is substituted into the mapping data obtained in the first initial mapping equation;

[0048] According to the first initial mapping data matrix And the working time data matrix b′ 1 Constructing the fitness function of the first salp population as follows,

[0049]

[0050] In the formula, χ 1 is a positive number, indicating the first correction parameter;

[0051] S2214, start iteration, before iteration, set the first current iteration number Set to 1; the fitness function of the first salp population is used in the first iteration Calculate the first initial position matrix d 1 The fitness value of the initial position of each salp in the first fitness value set is obtained, and the first fitness value set is obtained; the maximum fitness value and the corresponding salp position are selected from the first fitness value set to obtain the first global optimal fitness and the first global optimal position; the first initial position matrix d is adjusted according to the first global optimal fitness and the first global optimal position. 1 After the update is completed, the first current iteration number is Add 1 and enter the next iteration;

[0052] In each other iteration, the fitness function of the first salp population is used Calculate the fitness value of each salp position obtained after the update in the previous round of iterations to obtain a second fitness value set; select the largest fitness value and the corresponding salp position from the second fitness value set to obtain a second global optimal fitness and a second global optimal position; update the position of each salp according to the second global optimal fitness and the second global optimal position, and after the update is completed, replace the first current iteration number with the current number of fitness values. Add 1 and enter the next iteration;

[0053] S2215, when When , stop the iteration and obtain the first final global optimal position; substitute the components of the first final global optimal position in the dimensions of the undetermined coefficients of the roller wear degree grade data and the bearing wear degree grade data and the components in the offset dimension into the first initial mapping equation to obtain the optimized first mapping equation; substitute each roller wear degree grade data and bearing wear degree grade data in the grade data matrix b into the optimized first mapping equation respectively to obtain the first optimized mapping data matrix;

[0054] Set the first mapping error threshold e 1 , when the first optimized mapping data matrix does not contain mapping data and working time data matrix b′ 1 The difference between the corresponding working hours data is greater than or equal to e 1 When the difference between the mapping data and the corresponding working time data in the first optimized mapping data matrix is ​​greater than or equal to e1 When the mapping data matrix b′ is not present in the first optimized mapping data matrix, return to S2214 and continue to iterate until the mapping data matrix b′ does not exist in the first optimized mapping data matrix. 1 The difference between the corresponding working hours data is greater than or equal to e 1 until the time;

[0055] The structure of the salp optimization algorithm is simple, and it can quickly converge to the optimal solution. It uses a chain structure to simulate the behavior of the salp population. The leader in the salp population guides the movement of the group, and other salps follow, so that it can effectively perform global search. Based on the above, this scheme uses the salp optimization algorithm to iteratively optimize the parameters of the first initial mapping equation multiple times, and measures the fitness of each salp in the salp population by the difference between its mapped working time data and the actual working time data, so that as the iteration proceeds, the difference between its mapped working time data and the actual working time data becomes smaller and smaller, which means that the mapping effect of the first initial mapping equation is getting better and better.

[0056] Preferably, in S22, according to the level data matrix b and the power consumption data matrix b′ 2 Optimizing and adjusting the second initial mapping equation to obtain the second final mapping equation includes the following steps:

[0057] S2221. Establishment of the second salp population c 2i represents the i-th salp in the second salp population, c′ 2 represents the size of the second salp population; sets the maximum number of iterations of the second salp population to The current number of iterations is They are respectively recorded as the second maximum number of iterations and the second current number of iterations; the search space dimension of the second salp population is 3 dimensions;

[0058] S2222: Set the undetermined coefficients of the roller wear level data and the bearing wear level data in the second initial mapping equation and the value range of the offset amount to obtain a second value range set. respectively represent the lower limit and upper limit of the value of the undetermined coefficient of the roller wear degree grade data in the second initial mapping equation; They respectively represent the lower limit and upper limit of the undetermined coefficient of the bearing seat wear degree grade data in the second initial mapping equation; Respectively represent the lower limit and upper limit of the bias value in the second initial mapping equation;

[0059] The initial position of each salp in the second salp population is set by using the Tent mapping method in conjunction with the second value interval set to obtain a second initial position matrix d 2 ;as follows,

[0060]

[0061] Among them, d 2i1 d 2i2 d 2i3 represent the components of the initial position of the i-th salp in the second salp population in the dimension of the undetermined coefficient of the roller wear degree grade data, the component of the undetermined coefficient of the bearing seat wear degree grade data, and the component of the offset dimension respectively;

[0062] S2223, each roller wear level data and bearing wear level data in the level data matrix b are substituted into the second initial mapping equation to obtain the second initial mapping data matrix as follows,

[0063]

[0064] in, Respectively represent (b 1i1 ,b 1i2 )、(b 2i1 ,b 2i2 )、(b 3i1 ,b 3i2 )、(b 4i1 ,b 4i2 ) is substituted into the mapping data obtained in the second initial mapping equation;

[0065] According to the second initial mapping data matrix And the power loss data matrix b′ 2 Constructing the fitness function of the second salp population as follows,

[0066]

[0067] In the formula, χ 2 is a positive number, indicating the second correction parameter;

[0068] S2224, start iteration, before iteration, set the second current iteration number Set to 1; the fitness function of the second salp population is used in each iteration Calculate the fitness value of each salp position obtained after the update in the previous round of iteration to obtain a third fitness value set; select the largest fitness value and the corresponding salp position from the third fitness value set to obtain a third global optimal fitness and a third global optimal position; update the position of each salp according to the third global optimal fitness and the third global optimal position, and after the update is completed, set the second current iteration number Add 1 and enter the next iteration;

[0069] S2225, when When , the iteration is stopped to obtain the second final global optimal position; the components of the second final global optimal position in the dimensions of the undetermined coefficients of the roller wear degree grade data and the bearing wear degree grade data and the components in the offset dimension are substituted into the second initial mapping equation to obtain the optimized second mapping equation; each roller wear degree grade data and bearing wear degree grade data in the grade data matrix b are substituted into the optimized second mapping equation to obtain the second optimized mapping data matrix;

[0070] Set the second mapping error threshold e 2 , when the second optimized mapping data matrix does not contain mapping data and power loss data matrix b′ 2 The difference between the corresponding power loss data is greater than or equal to e 2 When the difference between the mapping data and the corresponding power loss data in the second optimized mapping data matrix is ​​greater than or equal to e 2 When the mapping data matrix is ​​optimized, the process returns to S2224 and continues to iterate until the difference between the mapping data and the corresponding power loss data is greater than or equal to e in the second optimized mapping data matrix. 2 until the time;

[0071] The salp optimization algorithm is also used to iteratively optimize the parameters of the second initial mapping equation multiple times, and the difference between the mapped energy loss data and the actual energy loss data is used to measure the fitness of each salp in the salp population. As the iteration proceeds, the difference between the mapped energy loss data and the actual energy loss data becomes smaller and smaller, which means that the mapping effect of the second initial mapping equation is getting better and better.

[0072] Preferably, S3 comprises the following steps:

[0073] S31, set the roller press equipment to be monitored, and put the roller press equipment to be monitored into working state; set the historical time point collection and future time points e′1i , e′ 2i Respectively represent the set i-th historical time point and future time point; according to the historical time point set Collecting roller wear level data and bearing wear level data of the roller press equipment to be monitored when in working state, and obtaining a roller wear level data set to be predicted and a bearing wear level data set to be predicted;

[0074] S32, constructing a first initial support vector machine prediction model and a second initial support vector machine prediction model, using the training and test data in the to-be-predicted roller wear level data set and the to-be-predicted bearing wear level data set to train and test the first initial support vector machine prediction model and the second initial support vector machine prediction model, obtaining a first final support vector machine prediction model and a second final support vector machine prediction model; with the future time point set The first final support vector machine prediction model and the second final support vector machine prediction model are used to predict the data to be predicted in the roll wear degree grade data set to be predicted and the bearing wear degree grade data set to be predicted, respectively, to obtain a future roll wear degree grade data set and a future bearing wear degree grade data set;

[0075] Preferably, the S32 comprises the following steps:

[0076] S321, setting a first training data ratio, a second training data ratio, a first test data ratio and a second test data ratio; dividing the data set of the roller wear degree grade to be predicted and the data set of the bearing wear degree grade to be predicted according to the first training data ratio, the first test data ratio, the second training data ratio and the second test data ratio, respectively, to obtain a roller wear degree grade training data set, a roller wear degree grade test data set, a roller wear degree grade to be predicted data set, a bearing wear degree grade training data set, a bearing wear degree grade test data set and a bearing wear degree grade to be predicted data set;

[0077] S322, using the roller wear level training data set, roller wear level test data set, bearing wear level training data set, and bearing wear level test data set to respectively train and test the first initial support vector machine prediction model and the second initial support vector machine prediction model to obtain the first final support vector machine prediction model and the second final support vector machine prediction model;

[0078] S323, respectively using the first final support vector machine prediction model and the second final support vector machine prediction model to predict the roller wear degree grade to be predicted data set and the bearing wear degree grade to be predicted data set to obtain a future roller wear degree grade data set and a future bearing wear degree grade data set;

[0079] By using the roller wear degree grade training data set, the roller wear degree grade test data set, the bearing wear degree grade training data set, and the bearing wear degree grade test data set to train and test the first initial support vector machine prediction model and the second initial support vector machine prediction model respectively, the obtained first final support vector machine prediction model and the second final support vector machine prediction model have good prediction capabilities for the roller wear degree grade data and the bearing wear degree grade data respectively, ensuring the accuracy of subsequent predictions of the roller wear degree grade data and the bearing wear degree grade data of the monitored roller press equipment at future times.

[0080] Preferably, S4 comprises the following steps:

[0081] S41, merging the future roller wear level data set and the future bearing wear level data set to obtain a future level data set They respectively represent the predicted roller wear level data and bearing wear level data of the roller press equipment to be monitored at the i-th future time point;

[0082] S42, the future level data set Input into the first final mapping equation and the second final mapping equation respectively to obtain the future working time data set and future power loss datasets f 1i 、f 2i They respectively represent the working time data and the power consumption data of the roller press equipment to be monitored at the i-th future time point;

[0083] By mapping the working time data and power loss data corresponding to the future roller wear level data set and the future bearing wear level data set, a data basis is provided for subsequent determination of whether the monitored roller press equipment needs to be repaired.

[0084] Preferably, S5 comprises the following steps:

[0085] S51, setting a working time safety threshold and a power consumption safety threshold;

[0086] S52: When the future working time data set When there is no working time data less than the working time safety threshold in the data set, there is no need to repair the current roller wear and bearing wear of the roller press equipment to be monitored; otherwise, it is necessary to repair the current roller wear and bearing wear of the roller press equipment to be monitored until the future working time data set is Until there is no future working time data less than the working time safety threshold;

[0087] When the future power consumption data set When there is no power loss data greater than or equal to the power loss safety threshold in the data set, there is no need to repair the current roller wear and bearing wear of the roller press equipment to be monitored; otherwise, it is necessary to repair the current roller wear and bearing wear of the roller press equipment to be monitored until the future power loss data set is reached. Until there is no power loss data greater than or equal to the power loss safety threshold;

[0088] By setting the working time safety threshold and the power consumption safety threshold, it is possible to determine whether the future working time data and power consumption data obtained meet the requirements, and then determine whether the monitored roller press equipment needs to be repaired.

[0089] A digital-based roller press processing monitoring and management system includes a first roller press equipment wear level data acquisition module, a second roller press equipment wear level data acquisition module, a mapping equation construction module, a second roller press equipment wear level data acquisition module, a prediction model construction module, a prediction module, a mapping classification module and a final judgment module.

[0090] The present invention has the following beneficial effects:

[0091] 1. In the present invention, by constructing the first final mapping equation and the second final mapping equation, a mapping basis is provided for subsequently obtaining the corresponding working time data and electric energy loss data according to the roller wear degree level data and the bearing wear degree level data; by predicting the roller wear degree level data and the bearing wear degree level data of the monitored roller press equipment, the reasons causing the abnormalities in the working time data and the electric energy loss data are directly reflected, and when the abnormalities occur, targeted repairs can be carried out to quickly solve the problems; by mapping the roller wear degree level data and the bearing wear degree level data of the monitored roller press equipment at a future time point, the future working time data set and the future electric energy loss data set are obtained, thereby providing a basis for subsequent judgment on whether the monitored roller press equipment needs to be repaired.

[0092] 2. In the present invention, the parameters of the first initial mapping equation and the second initial mapping equation are optimized multiple times by adopting the Salp Unica optimization algorithm. As the iteration proceeds, the difference between the mapped working time data and power consumption data and the actual working time data and power consumption data becomes smaller and smaller, which means that the mapping effects of the first initial mapping equation and the second initial mapping equation are getting better and better.

[0093] 3. The mapping equation in the present invention adopts the form of a linear function, which ensures the simplicity of the mapping equation and makes the subsequent optimization and adjustment of the parameters of the first initial mapping equation and the second initial mapping equation less complicated.

[0094] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0095] In order to more clearly illustrate the technical solutions of the embodiments of the invention, the drawings required for describing the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the invention. For ordinary technicians in this field, they can also obtain drawings based on these drawings without paying creative work.

[0096] Figure 1 The present invention is a schematic diagram of a process flow of monitoring and managing a roller press processing process by a digital roller press processing process monitoring and management system. DETAILED DESCRIPTION

[0097] The following will be combined with the drawings in the embodiments of the invention to clearly and completely describe the technical solutions in the embodiments of the invention. Obviously, the described embodiments are only part of the embodiments of the invention, not all of the embodiments. Based on the embodiments in the invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the invention.

[0098] In the description of the present invention, it is necessary to understand that the terms "opening", "upper", "lower", "top", "middle", "inside" and the like indicating orientation or positional relationship are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the components or elements referred to must have a specific orientation, be constructed and operate in a specific orientation, and therefore cannot be understood as limiting the invention.

[0099] Embodiment 1

[0100] This embodiment is a digital roller press process monitoring and management method, comprising the following steps:

[0101] S1. Collect roller wear level data, bearing wear level data, working time data and power loss data of multiple roller press equipment to obtain a level data matrix, a working time data matrix and a power loss data matrix;

[0102] The S1 comprises the following steps:

[0103] S11, setting a plurality of roller press equipments with worn rollers but no worn bearings, roller press equipments with no worn rollers and worn bearings, roller press equipments with worn bearings and worn bearings, and roller press equipments with no worn rollers and no worn bearings, to obtain a first roller press equipment set a 1 ={a 11 ,a 12 ,...,a 1i ,...,a 1a′}、Second roller press equipment set a 2 ={a 21 ,a 22 ,...,a 2i ,...,a 2a′}、Third roller press equipment set a 3 ={a 31 ,a 32 ,...,a 3i ,...,a 3a′} and the fourth roller press equipment set a 4 ={a 41 ,a 42 ,...,a 4i ,...,a 4a′}; a 1i 、a 2i 、a 3i 、a 4i They respectively represent the set i-th roller press equipment with worn rollers and no worn bearings, roller press equipment with no worn rollers and worn bearings, roller press equipment with worn bearings and worn bearings, and roller press equipment with no worn rollers and no worn bearings, and a′ represents the total number of the set roller press equipment with worn rollers and no worn bearings, roller press equipment with no worn rollers and worn bearings, roller press equipment with worn bearings and worn bearings, and roller press equipment with no worn rollers and no worn bearings;

[0104] S12, set the cement raw material to be pressed and the wear level of the press roller Bearing seat wear grade set They represent the set i-th roller wear level and bearing wear level respectively. Respectively represent the total number of set roller wear levels and bearing wear levels;

[0105] According to the wear degree of the pressure roller Bearing seat wear grade set Get the first roller press equipment set a 1 ={a 11 ,a 12 ,...,a 1i ,...,a 1a′} and the second roller press equipment set a 2 ={a 21 ,a 22 ,...,a 2i ,...,a 2a′}、Third roller press equipment set a 3 ={a 31 ,a 32 ,...,a 3i ,...,a 3a′} and the fourth roller press equipment set a 4 ={a 41 ,a 42 ,...,a 4i ,...,a 4a′ The wear degree grade data of the rollers and the bearings of each roller press equipment in} are used to obtain the grade data matrix b; as follows,

[0106]

[0107] b 1i1 , b 1i2 , b 2i1 , b 2i2 , b 3i1 , b 3i2 , b 4i1 , b 4i2 Respectively represent the roller wear degree grade data and the bearing wear degree grade data of the roller press equipment with wear on the i-th roller and no wear on the bearing seat, the roller press equipment with no wear on the roller and wear on the bearing seat, the roller press equipment with wear on the bearing seat and wear on the bearing seat, and the roller press equipment with no wear on the roller and no wear on the bearing seat;

[0108] S13, obtaining the first roller press equipment set a 1 ={a 11 ,a 12 ,...,a 1i ,...,a 1a′}、Second roller press equipment set a 2 ={a 21 ,a 22,...,a 2i ,...,a 2a′}、Third roller press equipment set a 3 ={a 31 ,a 32 ,...,a 3i ,...,a 3a′} and the fourth roller press equipment set a 4 ={a 41 ,a 42 ,...,a 4i ,...,a 4a′} working time data and power consumption data;

[0109] The S13 comprises the following steps:

[0110] S131, setting the fine particle size of the cement raw material to be pressed after pre-crushing, recorded as the first fine particle size;

[0111] S132, respectively using the first roller press equipment set according to the first fine particle size

[0112] a 1 ={a 11 ,a 12 ,...,a 1i ,...,a 1a′}、Second roller press equipment set a 2 ={a 21 ,a 22 ,...,a 2i ,...,a 2a′}、Third roller press equipment set a 3 ={a 31 ,a 32 ,...,a 3i ,...,a 3a′} and the fourth roller press equipment set

[0113] a 4 ={a 41 ,a 42 ,...,a 4i ,...,a 4a′} Performing a pre-crushing operation on the cement raw materials to be pressed of the same initial size, and stopping the pre-crushing operation when the fine particle size of the cement raw materials to be pressed reaches a first fine particle size;

[0114] S133, after the pre-crushing operation is completed, the working time data and power consumption data of each roller press equipment are obtained to obtain the working time data matrix b′ 1 And the power loss data matrix b′ 2 ; They are as follows,

[0115]

[0116] Among them, b′ 11i , b′ 12i , b′ 13i , b′ 14i b′ respectively represents the working time data of the i-th roller press equipment with worn rollers and no worn bearings, the roller press equipment with no worn rollers and worn bearings, the roller press equipment with worn bearings and worn bearings, and the roller press equipment with no worn rollers and no worn bearings for pre-crushing the cement raw materials to be pressed; 21i , b′ 22i , b′ 23i , b′ 24i Respectively represent the power consumption data of the i-th roller press equipment with worn rollers and no worn bearings, the roller press equipment with no worn rollers and worn bearings, the roller press equipment with worn bearings and worn bearings, and the roller press equipment with no worn rollers and no worn bearings for pre-crushing the cement raw materials to be pressed;

[0117] S2, constructing a first final mapping equation and a second final mapping equation using a working time data matrix, a power consumption data matrix and a level data matrix;

[0118] The S2 comprises the following steps:

[0119] S21, respectively setting the level data matrix b and the working time data matrix b′ 1 And the power loss data matrix b′ 2 The data mapping equation between them is used to obtain the first initial mapping equation and the second initial mapping equation; they are as follows:

[0120]

[0121] In the formula, Respectively represent the working time data and the power consumption data; Respectively represent the wear degree grade data of the roller and the bearing seat; α 1 , α 2 , α 4 , α 5 α represents the undetermined coefficients of the roller wear level data and the bearing wear level data in the first initial mapping equation and the second initial mapping equation respectively; 3 , α 6 Respectively represent the bias in the first initial mapping equation and the second initial mapping equation;

[0122] S22, according to the level data matrix b and the working time data matrix b′ 1 The first initial mapping equation is optimized and adjusted to obtain the first final mapping equation; according to the level data matrix b and the power loss data matrix b′ 2 Optimizing and adjusting the second initial mapping equation to obtain a second final mapping equation;

[0123] In S22, according to the level data matrix b and the working time data matrix b′ 1 Optimizing and adjusting the first initial mapping equation to obtain the first final mapping equation includes the following steps:

[0124] S2211, Establishing the first salp population c 1i represents the i-th salp in the first salp population, c′ 1 represents the size of the first salp population; sets the maximum number of iterations of the first salp population to The current number of iterations is are respectively recorded as the first maximum number of iterations and the first current number of iterations; the search space dimension of the first salp population is 3 dimensions;

[0125] S2212: Set the undetermined coefficients of the roller wear level data and the bearing wear level data in the first initial mapping equation and the value range of the offset amount to obtain a first value range set. They respectively represent the lower limit and upper limit of the undetermined coefficient of the roller wear degree grade data in the first initial mapping equation; respectively represent the lower limit and upper limit of the value of the undetermined coefficient of the bearing wear degree grade data in the first initial mapping equation; Respectively represent the lower limit and upper limit of the bias value in the first initial mapping equation;

[0126] The initial position of each salp in the first salp population is set by using the Tent mapping method in conjunction with the first value interval set to obtain a first initial position matrix d 1 ;as follows,

[0127]

[0128] Among them, d 1i1 d 1i2 d 1i3 Respectively represent the components of the initial position of the i-th salp in the first salp population in the undetermined coefficient dimension of the roller wear degree grade data, the components in the undetermined coefficient dimension of the bearing seat wear degree grade data, and the components in the offset dimension;

[0129] The mapping function of the Tent mapping is as follows,

[0130]

[0131] Where, d i ' +1 d i ′ represents the data generated in the i+1th iteration and the ith iteration respectively; β 1 is a random number between 0 and 1; β 1 =0.5;

[0132] S2213, each roller wear level data and bearing wear level data in the level data matrix b are substituted into the first initial mapping equation to obtain the first initial mapping data matrix as follows,

[0133]

[0134] in, Respectively represent (b 1i1 ,b 1i2 )、(b 2i1 ,b 2i2 )、(b 3i1 ,b 3i2 )、(b 4i1 ,b 4i2 ) is substituted into the mapping data obtained in the first initial mapping equation;

[0135] According to the first initial mapping data matrix And the working time data matrix b′ 1 Constructing the fitness function of the first salp population as follows,

[0136]

[0137] In the formula, χ 1 is a positive number, indicating the first correction parameter;

[0138] S2214, start iteration, before iteration, set the first current iteration number Set to 1; the fitness function of the first salp population is used in the first iteration Calculate the first initial position matrix d 1The fitness value of the initial position of each salp in the first fitness value set is obtained, and the first fitness value set is obtained; the maximum fitness value and the corresponding salp position are selected from the first fitness value set to obtain the first global optimal fitness and the first global optimal position; the first initial position matrix d is adjusted according to the first global optimal fitness and the first global optimal position. 1 After the update is completed, the first current iteration number is Add 1 and enter the next iteration;

[0139] In each other iteration, the fitness function of the first salp population is used Calculate the fitness value of each salp position obtained after the update in the previous round of iterations to obtain a second fitness value set; select the largest fitness value and the corresponding salp position from the second fitness value set to obtain a second global optimal fitness and a second global optimal position; update the position of each salp according to the second global optimal fitness and the second global optimal position, and after the update is completed, replace the first current iteration number with the current number of fitness values. Add 1 and enter the next iteration;

[0140] S2215, when When , stop the iteration and obtain the first final global optimal position; substitute the components of the first final global optimal position in the dimensions of the undetermined coefficients of the roller wear degree grade data and the bearing wear degree grade data and the components in the offset dimension into the first initial mapping equation to obtain the optimized first mapping equation; substitute each roller wear degree grade data and bearing wear degree grade data in the grade data matrix b into the optimized first mapping equation respectively to obtain the first optimized mapping data matrix;

[0141] Set the first mapping error threshold e 1 , when the first optimized mapping data matrix does not contain mapping data and working time data matrix b′ 1 The difference between the corresponding working hours data is greater than or equal to e 1 When the difference between the mapping data and the corresponding working time data in the first optimized mapping data matrix is ​​greater than or equal to e 1 When the mapping data matrix b′ is not present in the first optimized mapping data matrix, return to S2214 and continue to iterate until the mapping data matrix b′ does not exist in the first optimized mapping data matrix. 1 The difference between the corresponding working hours data is greater than or equal to e 1 until the time;

[0142] In S22, according to the level data matrix b and the power consumption data matrix b′2 Optimizing and adjusting the second initial mapping equation to obtain the second final mapping equation includes the following steps:

[0143] S2221. Establishment of the second salp population c 2i represents the i-th salp in the second salp population, c′ 2 represents the size of the second salp population; sets the maximum number of iterations of the second salp population to The current number of iterations is They are respectively recorded as the second maximum number of iterations and the second current number of iterations; the search space dimension of the second salp population is 3 dimensions;

[0144] S2222: Set the undetermined coefficients of the roller wear level data and the bearing wear level data in the second initial mapping equation and the value range of the offset amount to obtain a second value range set. respectively represent the lower limit and upper limit of the value of the undetermined coefficient of the roller wear degree grade data in the second initial mapping equation; They respectively represent the lower limit and upper limit of the undetermined coefficient of the bearing seat wear degree grade data in the second initial mapping equation; Respectively represent the lower limit and upper limit of the bias value in the second initial mapping equation;

[0145] The initial position of each salp in the second salp population is set by using the Tent mapping method in conjunction with the second value interval set to obtain a second initial position matrix d 2 ;as follows,

[0146]

[0147] Among them, d 2i1 d 2i2 d 2i3 represent the components of the initial position of the i-th salp in the second salp population in the dimension of the undetermined coefficient of the roller wear degree grade data, the component of the undetermined coefficient of the bearing seat wear degree grade data, and the component of the offset dimension respectively;

[0148] S2223, each roller wear level data and bearing wear level data in the level data matrix b are substituted into the second initial mapping equation to obtain the second initial mapping data matrix as follows,

[0149]

[0150] in, Respectively represent (b 1i1 ,b 1i2 )、(b 2i1 ,b 2i2 )、(b 3i1 ,b 3i2 )、(b 4i1 ,b 4i2 ) is substituted into the mapping data obtained in the second initial mapping equation;

[0151] According to the second initial mapping data matrix And the power loss data matrix b′ 2 Constructing the fitness function of the second salp population as follows,

[0152]

[0153] In the formula, χ 2 is a positive number, indicating the second correction parameter;

[0154] S2224, start iteration, before iteration, set the second current iteration number Set to 1; the fitness function of the second salp population is used in each iteration Calculate the fitness value of each salp position obtained after the update in the previous round of iteration to obtain a third fitness value set; select the largest fitness value and the corresponding salp position from the third fitness value set to obtain a third global optimal fitness and a third global optimal position; update the position of each salp according to the third global optimal fitness and the third global optimal position, and after the update is completed, set the second current iteration number Add 1 and enter the next iteration;

[0155] S2225, when When , the iteration is stopped to obtain the second final global optimal position; the components of the second final global optimal position in the dimensions of the undetermined coefficients of the roller wear degree grade data and the bearing wear degree grade data and the components in the offset dimension are substituted into the second initial mapping equation to obtain the optimized second mapping equation; each roller wear degree grade data and bearing wear degree grade data in the grade data matrix b are substituted into the optimized second mapping equation to obtain the second optimized mapping data matrix;

[0156] Set the second mapping error threshold e 2 , when the second optimized mapping data matrix does not contain mapping data and power loss data matrix b′ 2 The difference between the corresponding power loss data is greater than or equal to e 2When the difference between the mapping data and the corresponding power loss data in the second optimized mapping data matrix is ​​greater than or equal to e 2 When the mapping data matrix is ​​optimized, the process returns to S2224 and continues to iterate until the difference between the mapping data and the corresponding power loss data is greater than or equal to e in the second optimized mapping data matrix. 2 until the time;

[0157] S3, collecting roller wear degree grade data and bearing wear degree grade data of the roller press equipment to be monitored and constructing the first final support vector machine prediction model and the second final support vector machine prediction model; using the first final support vector machine prediction model and the second final support vector machine prediction model to predict the roller wear degree grade data and the bearing wear degree grade data of the roller press equipment to be monitored at a future moment, and obtaining a future roller wear degree grade data set and a future bearing wear degree grade data set;

[0158] The S3 comprises the following steps:

[0159] S31, set the roller press equipment to be monitored, and put the roller press equipment to be monitored into working state; set the historical time point collection and future time points e′ 1i , e′ 2i Respectively represent the set i-th historical time point and future time point; according to the historical time point set Collect roller wear level data and bearing wear level data of the roller press equipment to be monitored when it is in working state, and obtain a roller wear level data set to be predicted and a bearing wear level data set to be predicted;

[0160] S32, constructing a first initial support vector machine prediction model and a second initial support vector machine prediction model, using the training and test data in the to-be-predicted roller wear level data set and the to-be-predicted bearing wear level data set to train and test the first initial support vector machine prediction model and the second initial support vector machine prediction model, obtaining a first final support vector machine prediction model and a second final support vector machine prediction model; with the future time point set The first final support vector machine prediction model and the second final support vector machine prediction model are used to predict the data to be predicted in the roll wear degree grade data set to be predicted and the bearing wear degree grade data set to be predicted, respectively, to obtain a future roll wear degree grade data set and a future bearing wear degree grade data set;

[0161] The S32 comprises the following steps:

[0162] S321, setting a first training data ratio, a second training data ratio, a first test data ratio and a second test data ratio; dividing the data set of the roller wear degree grade to be predicted and the data set of the bearing wear degree grade to be predicted according to the first training data ratio, the first test data ratio, the second training data ratio and the second test data ratio, respectively, to obtain a roller wear degree grade training data set, a roller wear degree grade test data set, a roller wear degree grade to be predicted data set, a bearing wear degree grade training data set, a bearing wear degree grade test data set and a bearing wear degree grade to be predicted data set;

[0163] S322, using the roller wear level training data set, roller wear level test data set, bearing wear level training data set, and bearing wear level test data set to respectively train and test the first initial support vector machine prediction model and the second initial support vector machine prediction model to obtain the first final support vector machine prediction model and the second final support vector machine prediction model;

[0164] The S322 includes the following steps:

[0165] S3221, setting a first training error threshold and a second training error threshold;

[0166] The roller wear level training data set is input into the first initial support vector machine prediction model for training; during the training process, when the training error is less than the first training error threshold, the training is stopped to obtain the trained first support vector machine prediction model; otherwise, the training is continued until the training error is less than the first training error threshold;

[0167] The bearing wear level training data set is input into the second initial support vector machine prediction model for training; during the training process, when the training error is less than the second training error threshold, the training is stopped to obtain the trained second support vector machine prediction model; otherwise, the training is continued until the training error is less than the second training error threshold;

[0168] S3222, setting a first accuracy threshold and a second accuracy threshold; inputting the roller wear level test data set and the bearing wear level test data set into the trained first support vector machine prediction model and the trained second support vector machine prediction model for testing, respectively, to obtain a first accuracy rate and a second accuracy rate;

[0169] When the first accuracy rate is greater than or equal to the first accuracy rate threshold, the trained first support vector machine prediction model is used as the first final support vector machine prediction model; otherwise, return to S3221 to continue training the trained first support vector machine prediction model until the first accuracy rate is greater than or equal to the first accuracy rate threshold;

[0170] When the second accuracy rate is greater than or equal to the second accuracy rate threshold, the trained second support vector machine prediction model is used as the second final support vector machine prediction model; otherwise, return to S3221 to continue training the trained second support vector machine prediction model until the second accuracy rate is greater than or equal to the second accuracy rate threshold;

[0171] S323, respectively using the first final support vector machine prediction model and the second final support vector machine prediction model to predict the roller wear degree grade to be predicted data set and the bearing wear degree grade to be predicted data set to obtain a future roller wear degree grade data set and a future bearing wear degree grade data set;

[0172] S4, using the first final mapping equation and the second final mapping equation to map and classify the future roller wear level data set and the future bearing wear level data set to obtain a future working time data set and a future power loss data set;

[0173] The S4 comprises the following steps:

[0174] S41, merging the future roller wear level data set and the future bearing wear level data set to obtain a future level data set They respectively represent the predicted roller wear level data and bearing wear level data of the roller press equipment to be monitored at the i-th future time point;

[0175] S42, the future level data set Input into the first final mapping equation and the second final mapping equation respectively to obtain the future working time data set and future power loss datasets f 1i 、f 2i They respectively represent the working time data and the power consumption data of the roller press equipment to be monitored at the i-th future time point;

[0176] S5. Determine whether the roller press equipment to be monitored needs to be repaired according to the future working time data set and the future power consumption data set;

[0177] The S5 comprises the following steps:

[0178] S51, setting a working time safety threshold and a power consumption safety threshold;

[0179] S52: When the future working time data set When there is no working time data less than the working time safety threshold in the data set, there is no need to repair the current roller wear and bearing wear of the roller press equipment to be monitored; otherwise, it is necessary to repair the current roller wear and bearing wear of the roller press equipment to be monitored until the future working time data set is Until there is no future working time data less than the working time safety threshold;

[0180] When the future power consumption data set When there is no power loss data greater than or equal to the power loss safety threshold in the data set, there is no need to repair the current roller wear and bearing wear of the roller press equipment to be monitored; otherwise, it is necessary to repair the current roller wear and bearing wear of the roller press equipment to be monitored until the future power loss data set is reached. Until there is no power loss data greater than or equal to the power loss safety threshold.

[0181] Embodiment 2

[0182] This embodiment discloses a digital-based roller press machining process monitoring and management system, the system can implement the method of the above embodiment, including a first roller press equipment wear level data acquisition module, a first roller press equipment data acquisition module, a mapping equation construction module, a second roller press equipment wear level data acquisition module, a prediction model construction module, a prediction module, a mapping classification module and a final determination module;

[0183] The first roller press equipment wear level data acquisition module is used to collect roller wear level data and bearing wear level data of multiple roller press equipment to obtain a level data matrix;

[0184] The first roller press equipment data acquisition module is used to collect the working time data and power consumption data of multiple roller press equipment to obtain the working time data matrix and the power consumption data matrix;

[0185] The mapping equation construction module is used to construct a first final mapping equation and a second final mapping equation by using the working time data matrix, the power consumption data matrix and the level data matrix;

[0186] The second roller press equipment wear level data acquisition module is used to collect roller wear level data and bearing wear level data at multiple historical time points of the roller press equipment to be monitored, and obtain a roller wear level data set to be predicted and a bearing wear level data set to be predicted;

[0187] The prediction model building module is used to construct a first final support vector machine prediction model and a second final support vector machine prediction model using a data set of the wear degree level of the roller to be predicted and a data set of the wear degree level of the bearing seat to be predicted;

[0188] The prediction module is used to use the first final support vector machine prediction model and the second final support vector machine prediction model to predict the roller wear degree level data and the bearing wear degree level data of the roller press equipment to be monitored at a future moment, and obtain a future roller wear degree level data set and a future bearing wear degree level data set;

[0189] The mapping classification module is used to map and classify the future roller wear level data set and the future bearing wear level data set using the first final mapping equation and the second final mapping equation to obtain the future working time data set and the future power loss data set;

[0190] The final determination module is used to determine whether the roller press equipment to be monitored needs to be repaired according to the future working time data set and the future power consumption data set.

[0191] In the description of this specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0192] The preferred embodiments of the invention disclosed above are only used to help explain the invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to the specific implementation methods described. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the invention, so that those skilled in the art can understand and use the invention well.

Claims

1. A digital roller press process monitoring and management method, characterized in that: The following steps are involved: S1. Collect roller wear level data, bearing wear level data, working time data and power loss data of multiple roller press equipment to obtain a level data matrix, a working time data matrix and a power loss data matrix; S2, constructing a first final mapping equation and a second final mapping equation using a working time data matrix, a power consumption data matrix and a level data matrix; S3, collecting roller wear degree grade data and bearing wear degree grade data of the roller press equipment to be monitored and constructing the first final support vector machine prediction model and the second final support vector machine prediction model; using the first final support vector machine prediction model and the second final support vector machine prediction model to predict the roller wear degree grade data and the bearing wear degree grade data of the roller press equipment to be monitored at a future moment, and obtaining a future roller wear degree grade data set and a future bearing wear degree grade data set; S4, using the first final mapping equation and the second final mapping equation to map and classify the future roller wear level data set and the future bearing wear level data set to obtain a future working time data set and a future power loss data set; S5. Determine whether the roller press equipment to be monitored needs to be repaired according to the future working time data set and the future power consumption data set.

2. A method for monitoring and managing the roller press process based on digitization according to claim 1, characterized in that: The S1 comprises the following steps: S11, setting a plurality of roller press equipments with worn rollers and no worn bearings, roller press equipments with no worn rollers and worn bearings, roller press equipments with worn bearings and worn bearings, and roller press equipments with no worn rollers and no worn bearings, to obtain a first roller press equipment set, a second roller press equipment set, a third roller press equipment set, and a fourth roller press equipment set; S12, setting the cement raw material to be pressed, the roller wear degree level set and the bearing wear degree level set; obtaining the roller wear degree level data and the bearing wear degree level data of the first roller press equipment set, the second roller press equipment set, the third roller press equipment set and the fourth roller press equipment set according to the roller wear degree level set and the bearing wear degree level set, and obtaining a level data matrix; S13, obtaining the working time data and power consumption data of the first roller press equipment set, the second roller press equipment set, the third roller press equipment set and the fourth roller press equipment set.

3. The method for monitoring and managing the roller press process based on digitization according to claim 2 is characterized in that: The S13 comprises the following steps: S131, setting the fine particle size of the cement raw material to be pressed after pre-crushing, recorded as the first fine particle size; S132, according to the first fine particle size, respectively using the first roller press equipment set, the second roller press equipment set, the third roller press equipment set and the fourth roller press equipment set to perform pre-crushing operations on the cement raw materials to be pressed with the same initial size, and when the fine particle size of the cement raw materials to be pressed reaches the first fine particle size, stopping the pre-crushing operation; S133. After the pre-crushing operation is completed, the working time data and the power consumption data of each roller press equipment are obtained to obtain a working time data matrix and a power consumption data matrix.

4. The method for monitoring and managing the roller press process based on digitization according to claim 3 is characterized in that: The S2 comprises the following steps: S21, respectively setting data mapping equations between the level data matrix and the working time data matrix and the power consumption data matrix to obtain a first initial mapping equation and a second initial mapping equation; S22. Optimize and adjust the first initial mapping equation according to the level data matrix and the working time data matrix to obtain a first final mapping equation; optimize and adjust the second initial mapping equation according to the level data matrix and the power loss data matrix to obtain a second final mapping equation.

5. The method for monitoring and managing the roller press process based on digitization according to claim 4 is characterized in that: In S22, optimizing and adjusting the first initial mapping equation according to the level data matrix and the working time data matrix to obtain the first final mapping equation includes the following steps: S2211, constructing a first salp population; setting the maximum number of iterations of the first salp population to The current number of iterations is S2212, setting the unknown coefficients of the roller wear level data and the bearing wear level data in the first initial mapping equation and the value range of the offset amount to obtain a first value range set; using the Tent mapping method to set the initial position of each salp in the first value range set to obtain a first initial position matrix; S2213, respectively substituting each roller wear level data and bearing wear level data in the level data matrix into the first initial mapping equation to obtain a first initial mapping data matrix; constructing a fitness function of the first salp population according to the first initial mapping data matrix and the working time data matrix; S2214, start iteration; in each round of iteration, use the fitness function of the first salp population to calculate the fitness value of each salp position obtained after the update in the previous round of iteration, and update each salp position; S2215, when , stop iterating and obtain the first final global optimal position; substitute the first final global optimal position into the first initial mapping equation to obtain the optimized first mapping equation; substitute the wear level data of each roller and the wear level data of the bearing position in the level data matrix into the optimized first mapping equation respectively to obtain the first optimized mapping data matrix; set the first mapping error threshold e1, when the difference between the mapping data that does not exist in the first optimized mapping data matrix and the corresponding working time data in the working time data matrix is ​​greater than or equal to e1, use the optimized first mapping equation as the first final mapping equation; otherwise, return to S2214 to continue iterating until the difference between the mapping data that does not exist in the first optimized mapping data matrix and the corresponding working time data in the working time data matrix is ​​greater than or equal to e1.

6. A method for monitoring and managing the roller press process based on digitization according to claim 5, characterized in that: The S3 comprises the following steps: S31, setting the roller press equipment to be monitored, the historical time point set and the future time point set; collecting the roller wear level data and the bearing wear level data of the roller press equipment to be monitored in the working state according to the historical time point set, and obtaining the roller wear level data set to be predicted and the bearing wear level data set to be predicted; S32. Construct a first initial support vector machine prediction model and a second initial support vector machine prediction model, and use the training and test data in the data set of the wear degree level of the pressure roller to be predicted and the data set of the wear degree level of the bearing to be predicted to train and test the first initial support vector machine prediction model and the second initial support vector machine prediction model to obtain a first final support vector machine prediction model and a second final support vector machine prediction model; use the first final support vector machine prediction model and the second final support vector machine prediction model in conjunction with the future time point set to predict the data to be predicted in the data set of the wear degree level of the pressure roller to be predicted and the data set of the wear degree level of the bearing to be predicted, respectively, to obtain a future data set of the wear degree level of the pressure roller and a future data set of the wear degree level of the bearing.

7. A method for monitoring and managing the roller press process based on digitization according to claim 6, characterized in that: The S4 comprises the following steps: S41, merging the future roller wear degree grade data set and the future bearing seat wear degree grade data set to obtain a future grade data set; S42, inputting the future level data set into the first final mapping equation and the second final mapping equation respectively to obtain a future working time data set and a future electric energy consumption data set.

8. The method for monitoring and managing the roller press process based on digitization according to claim 7 is characterized in that: The S5 comprises the following steps: S51, setting a working time safety threshold and a power consumption safety threshold; S52. When there is no working duration data less than the working duration safety threshold in the future working duration data set and there is no power loss data greater than or equal to the power loss safety threshold in the future power loss data set, there is no need to repair the current roller wear and bearing wear of the roller press equipment to be monitored; otherwise, it is necessary to repair the current roller wear degree and bearing wear degree of the roller press equipment to be monitored until there is no future working duration data less than the working duration safety threshold in the future working duration data set and there is no power loss data greater than or equal to the power loss safety threshold in the future power loss data set.

9. A system for implementing the digital-based roller press machining process monitoring and management method as described in any one of claims 1 to 8.

Images